{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 29704,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00033665499596014,
      "grad_norm": 2.0698454535440955,
      "learning_rate": 9.999999552565099e-06,
      "loss": 0.8854928970336914,
      "num_input_tokens_seen": 156912,
      "step": 5,
      "train_runtime": 15.0055,
      "train_tokens_per_second": 10456.944
    },
    {
      "epoch": 0.00067330999192028,
      "grad_norm": 1.595348631388974,
      "learning_rate": 9.999997734860948e-06,
      "loss": 0.6845027446746826,
      "num_input_tokens_seen": 310632,
      "step": 10,
      "train_runtime": 27.7655,
      "train_tokens_per_second": 11187.707
    },
    {
      "epoch": 0.0010099649878804202,
      "grad_norm": 1.4987275661962902,
      "learning_rate": 9.999994518923374e-06,
      "loss": 0.6747148513793946,
      "num_input_tokens_seen": 466600,
      "step": 15,
      "train_runtime": 40.4341,
      "train_tokens_per_second": 11539.762
    },
    {
      "epoch": 0.00134661998384056,
      "grad_norm": 1.2961275674842117,
      "learning_rate": 9.999989904753278e-06,
      "loss": 0.6179544925689697,
      "num_input_tokens_seen": 625032,
      "step": 20,
      "train_runtime": 54.0293,
      "train_tokens_per_second": 11568.382
    },
    {
      "epoch": 0.0016832749798007002,
      "grad_norm": 1.3270010040761284,
      "learning_rate": 9.999983892351947e-06,
      "loss": 0.6241541385650635,
      "num_input_tokens_seen": 785400,
      "step": 25,
      "train_runtime": 66.8662,
      "train_tokens_per_second": 11745.851
    },
    {
      "epoch": 0.0020199299757608404,
      "grad_norm": 1.2944622140890807,
      "learning_rate": 9.999976481721066e-06,
      "loss": 0.5376529693603516,
      "num_input_tokens_seen": 940640,
      "step": 30,
      "train_runtime": 79.3784,
      "train_tokens_per_second": 11850.078
    },
    {
      "epoch": 0.0023565849717209803,
      "grad_norm": 1.2313305414677842,
      "learning_rate": 9.999967672862704e-06,
      "loss": 0.5632558822631836,
      "num_input_tokens_seen": 1095816,
      "step": 35,
      "train_runtime": 91.858,
      "train_tokens_per_second": 11929.456
    },
    {
      "epoch": 0.00269323996768112,
      "grad_norm": 1.0541117790236472,
      "learning_rate": 9.999957465779326e-06,
      "loss": 0.5211737155914307,
      "num_input_tokens_seen": 1259144,
      "step": 40,
      "train_runtime": 104.769,
      "train_tokens_per_second": 12018.284
    },
    {
      "epoch": 0.0030298949636412605,
      "grad_norm": 1.3166924160252729,
      "learning_rate": 9.999945860473788e-06,
      "loss": 0.5659154415130615,
      "num_input_tokens_seen": 1414720,
      "step": 45,
      "train_runtime": 117.875,
      "train_tokens_per_second": 12001.87
    },
    {
      "epoch": 0.0033665499596014004,
      "grad_norm": 1.5398853879452326,
      "learning_rate": 9.999932856949333e-06,
      "loss": 0.5425772190093994,
      "num_input_tokens_seen": 1573368,
      "step": 50,
      "train_runtime": 130.9149,
      "train_tokens_per_second": 12018.25
    },
    {
      "epoch": 0.0037032049555615404,
      "grad_norm": 1.1138942943585206,
      "learning_rate": 9.999918455209599e-06,
      "loss": 0.5183193683624268,
      "num_input_tokens_seen": 1731392,
      "step": 55,
      "train_runtime": 143.6109,
      "train_tokens_per_second": 12056.13
    },
    {
      "epoch": 0.004039859951521681,
      "grad_norm": 1.2511809161167968,
      "learning_rate": 9.99990265525861e-06,
      "loss": 0.5496606349945068,
      "num_input_tokens_seen": 1889256,
      "step": 60,
      "train_runtime": 157.0216,
      "train_tokens_per_second": 12031.822
    },
    {
      "epoch": 0.004376514947481821,
      "grad_norm": 1.3649377833425977,
      "learning_rate": 9.999885457100789e-06,
      "loss": 0.5638967514038086,
      "num_input_tokens_seen": 2043608,
      "step": 65,
      "train_runtime": 170.0156,
      "train_tokens_per_second": 12020.119
    },
    {
      "epoch": 0.0047131699434419605,
      "grad_norm": 1.2134449261239417,
      "learning_rate": 9.999866860740943e-06,
      "loss": 0.5087326526641845,
      "num_input_tokens_seen": 2197176,
      "step": 70,
      "train_runtime": 183.0566,
      "train_tokens_per_second": 12002.712
    },
    {
      "epoch": 0.005049824939402101,
      "grad_norm": 1.109519206634432,
      "learning_rate": 9.999846866184274e-06,
      "loss": 0.521860408782959,
      "num_input_tokens_seen": 2361016,
      "step": 75,
      "train_runtime": 195.5143,
      "train_tokens_per_second": 12075.922
    },
    {
      "epoch": 0.00538647993536224,
      "grad_norm": 1.1342972606494852,
      "learning_rate": 9.99982547343637e-06,
      "loss": 0.49937987327575684,
      "num_input_tokens_seen": 2506120,
      "step": 80,
      "train_runtime": 209.4003,
      "train_tokens_per_second": 11968.084
    },
    {
      "epoch": 0.005723134931322381,
      "grad_norm": 1.2085456562956363,
      "learning_rate": 9.999802682503217e-06,
      "loss": 0.4897414207458496,
      "num_input_tokens_seen": 2667224,
      "step": 85,
      "train_runtime": 222.3909,
      "train_tokens_per_second": 11993.405
    },
    {
      "epoch": 0.006059789927282521,
      "grad_norm": 1.0987132862115228,
      "learning_rate": 9.999778493391186e-06,
      "loss": 0.5461398124694824,
      "num_input_tokens_seen": 2827728,
      "step": 90,
      "train_runtime": 235.4162,
      "train_tokens_per_second": 12011.612
    },
    {
      "epoch": 0.0063964449232426605,
      "grad_norm": 1.249478349933596,
      "learning_rate": 9.999752906107043e-06,
      "loss": 0.5407850265502929,
      "num_input_tokens_seen": 2988616,
      "step": 95,
      "train_runtime": 248.305,
      "train_tokens_per_second": 12036.066
    },
    {
      "epoch": 0.006733099919202801,
      "grad_norm": 2.567034402628015,
      "learning_rate": 9.999725920657944e-06,
      "loss": 0.5400688648223877,
      "num_input_tokens_seen": 3144096,
      "step": 100,
      "train_runtime": 261.0944,
      "train_tokens_per_second": 12041.989
    },
    {
      "epoch": 0.007069754915162941,
      "grad_norm": 1.1398675016545547,
      "learning_rate": 9.999697537051432e-06,
      "loss": 0.5188553810119629,
      "num_input_tokens_seen": 3298368,
      "step": 105,
      "train_runtime": 274.325,
      "train_tokens_per_second": 12023.578
    },
    {
      "epoch": 0.007406409911123081,
      "grad_norm": 1.227717719283168,
      "learning_rate": 9.999667755295449e-06,
      "loss": 0.49572024345397947,
      "num_input_tokens_seen": 3455432,
      "step": 110,
      "train_runtime": 287.1766,
      "train_tokens_per_second": 12032.43
    },
    {
      "epoch": 0.007743064907083221,
      "grad_norm": 1.303289307700912,
      "learning_rate": 9.999636575398319e-06,
      "loss": 0.5078877449035645,
      "num_input_tokens_seen": 3612472,
      "step": 115,
      "train_runtime": 300.1408,
      "train_tokens_per_second": 12035.923
    },
    {
      "epoch": 0.008079719903043361,
      "grad_norm": 1.39142518218521,
      "learning_rate": 9.999603997368766e-06,
      "loss": 0.4322662353515625,
      "num_input_tokens_seen": 3767256,
      "step": 120,
      "train_runtime": 312.8849,
      "train_tokens_per_second": 12040.389
    },
    {
      "epoch": 0.008416374899003501,
      "grad_norm": 1.2287739937650264,
      "learning_rate": 9.999570021215895e-06,
      "loss": 0.4877100944519043,
      "num_input_tokens_seen": 3927448,
      "step": 125,
      "train_runtime": 326.8862,
      "train_tokens_per_second": 12014.727
    },
    {
      "epoch": 0.008753029894963642,
      "grad_norm": 1.134810485220154,
      "learning_rate": 9.99953464694921e-06,
      "loss": 0.502169418334961,
      "num_input_tokens_seen": 4081760,
      "step": 130,
      "train_runtime": 339.6024,
      "train_tokens_per_second": 12019.231
    },
    {
      "epoch": 0.009089684890923782,
      "grad_norm": 1.3207295749847312,
      "learning_rate": 9.999497874578606e-06,
      "loss": 0.4972585678100586,
      "num_input_tokens_seen": 4244320,
      "step": 135,
      "train_runtime": 352.2974,
      "train_tokens_per_second": 12047.549
    },
    {
      "epoch": 0.009426339886883921,
      "grad_norm": 1.2630112814754622,
      "learning_rate": 9.999459704114361e-06,
      "loss": 0.4974041938781738,
      "num_input_tokens_seen": 4405272,
      "step": 140,
      "train_runtime": 366.1704,
      "train_tokens_per_second": 12030.661
    },
    {
      "epoch": 0.00976299488284406,
      "grad_norm": 1.095641717852237,
      "learning_rate": 9.999420135567153e-06,
      "loss": 0.5314162254333497,
      "num_input_tokens_seen": 4564624,
      "step": 145,
      "train_runtime": 379.5194,
      "train_tokens_per_second": 12027.381
    },
    {
      "epoch": 0.010099649878804202,
      "grad_norm": 1.1191641925266658,
      "learning_rate": 9.999379168948047e-06,
      "loss": 0.5017687797546386,
      "num_input_tokens_seen": 4725648,
      "step": 150,
      "train_runtime": 392.5065,
      "train_tokens_per_second": 12039.667
    },
    {
      "epoch": 0.010436304874764341,
      "grad_norm": 1.1510811912223,
      "learning_rate": 9.999336804268496e-06,
      "loss": 0.4947249412536621,
      "num_input_tokens_seen": 4888368,
      "step": 155,
      "train_runtime": 405.3102,
      "train_tokens_per_second": 12060.808
    },
    {
      "epoch": 0.01077295987072448,
      "grad_norm": 1.1750854077516668,
      "learning_rate": 9.999293041540351e-06,
      "loss": 0.5044156551361084,
      "num_input_tokens_seen": 5048056,
      "step": 160,
      "train_runtime": 418.3097,
      "train_tokens_per_second": 12067.747
    },
    {
      "epoch": 0.011109614866684622,
      "grad_norm": 1.0712399661489687,
      "learning_rate": 9.999247880775849e-06,
      "loss": 0.5079492568969727,
      "num_input_tokens_seen": 5205448,
      "step": 165,
      "train_runtime": 431.6836,
      "train_tokens_per_second": 12058.48
    },
    {
      "epoch": 0.011446269862644761,
      "grad_norm": 1.2383086918864776,
      "learning_rate": 9.999201321987617e-06,
      "loss": 0.463563346862793,
      "num_input_tokens_seen": 5365176,
      "step": 170,
      "train_runtime": 444.459,
      "train_tokens_per_second": 12071.251
    },
    {
      "epoch": 0.011782924858604901,
      "grad_norm": 1.1021188326109024,
      "learning_rate": 9.999153365188676e-06,
      "loss": 0.5227638244628906,
      "num_input_tokens_seen": 5526232,
      "step": 175,
      "train_runtime": 457.9486,
      "train_tokens_per_second": 12067.362
    },
    {
      "epoch": 0.012119579854565042,
      "grad_norm": 1.170848709161684,
      "learning_rate": 9.99910401039244e-06,
      "loss": 0.45966787338256837,
      "num_input_tokens_seen": 5679600,
      "step": 180,
      "train_runtime": 470.9486,
      "train_tokens_per_second": 12059.915
    },
    {
      "epoch": 0.012456234850525182,
      "grad_norm": 1.2694832577032766,
      "learning_rate": 9.999053257612708e-06,
      "loss": 0.46491479873657227,
      "num_input_tokens_seen": 5840616,
      "step": 185,
      "train_runtime": 483.4766,
      "train_tokens_per_second": 12080.453
    },
    {
      "epoch": 0.012792889846485321,
      "grad_norm": 1.2155432705353497,
      "learning_rate": 9.999001106863672e-06,
      "loss": 0.5444772243499756,
      "num_input_tokens_seen": 6000464,
      "step": 190,
      "train_runtime": 496.987,
      "train_tokens_per_second": 12073.683
    },
    {
      "epoch": 0.013129544842445462,
      "grad_norm": 1.097304634737417,
      "learning_rate": 9.998947558159916e-06,
      "loss": 0.5072970390319824,
      "num_input_tokens_seen": 6163560,
      "step": 195,
      "train_runtime": 510.6688,
      "train_tokens_per_second": 12069.583
    },
    {
      "epoch": 0.013466199838405602,
      "grad_norm": 1.2223737546636617,
      "learning_rate": 9.99889261151642e-06,
      "loss": 0.4785150051116943,
      "num_input_tokens_seen": 6315248,
      "step": 200,
      "train_runtime": 523.2587,
      "train_tokens_per_second": 12069.074
    },
    {
      "epoch": 0.013802854834365741,
      "grad_norm": 1.1080866067904105,
      "learning_rate": 9.998836266948542e-06,
      "loss": 0.47400674819946287,
      "num_input_tokens_seen": 6473920,
      "step": 205,
      "train_runtime": 536.7226,
      "train_tokens_per_second": 12061.947
    },
    {
      "epoch": 0.014139509830325882,
      "grad_norm": 1.2536449494794142,
      "learning_rate": 9.998778524472043e-06,
      "loss": 0.47736215591430664,
      "num_input_tokens_seen": 6627424,
      "step": 210,
      "train_runtime": 549.4742,
      "train_tokens_per_second": 12061.393
    },
    {
      "epoch": 0.014476164826286022,
      "grad_norm": 1.1320332661833703,
      "learning_rate": 9.99871938410307e-06,
      "loss": 0.4658658981323242,
      "num_input_tokens_seen": 6786896,
      "step": 215,
      "train_runtime": 563.8482,
      "train_tokens_per_second": 12036.744
    },
    {
      "epoch": 0.014812819822246161,
      "grad_norm": 1.096574211089579,
      "learning_rate": 9.99865884585816e-06,
      "loss": 0.5220664978027344,
      "num_input_tokens_seen": 6941568,
      "step": 220,
      "train_runtime": 577.1139,
      "train_tokens_per_second": 12028.072
    },
    {
      "epoch": 0.015149474818206303,
      "grad_norm": 1.2726189219492656,
      "learning_rate": 9.998596909754245e-06,
      "loss": 0.4895173072814941,
      "num_input_tokens_seen": 7100664,
      "step": 225,
      "train_runtime": 590.0872,
      "train_tokens_per_second": 12033.246
    },
    {
      "epoch": 0.015486129814166442,
      "grad_norm": 1.1458655876594552,
      "learning_rate": 9.998533575808642e-06,
      "loss": 0.4762932777404785,
      "num_input_tokens_seen": 7256336,
      "step": 230,
      "train_runtime": 603.4384,
      "train_tokens_per_second": 12024.981
    },
    {
      "epoch": 0.015822784810126583,
      "grad_norm": 1.0631508251033874,
      "learning_rate": 9.998468844039064e-06,
      "loss": 0.4685684680938721,
      "num_input_tokens_seen": 7414968,
      "step": 235,
      "train_runtime": 617.0903,
      "train_tokens_per_second": 12016.019
    },
    {
      "epoch": 0.016159439806086723,
      "grad_norm": 1.084732059086051,
      "learning_rate": 9.998402714463613e-06,
      "loss": 0.5053768634796143,
      "num_input_tokens_seen": 7577360,
      "step": 240,
      "train_runtime": 630.1816,
      "train_tokens_per_second": 12024.09
    },
    {
      "epoch": 0.016496094802046862,
      "grad_norm": 1.0156196358315555,
      "learning_rate": 9.998335187100782e-06,
      "loss": 0.5120949745178223,
      "num_input_tokens_seen": 7739440,
      "step": 245,
      "train_runtime": 643.3035,
      "train_tokens_per_second": 12030.776
    },
    {
      "epoch": 0.016832749798007002,
      "grad_norm": 1.1229637578221257,
      "learning_rate": 9.998266261969453e-06,
      "loss": 0.504401445388794,
      "num_input_tokens_seen": 7899936,
      "step": 250,
      "train_runtime": 657.508,
      "train_tokens_per_second": 12014.966
    },
    {
      "epoch": 0.01716940479396714,
      "grad_norm": 1.2876741817370467,
      "learning_rate": 9.998195939088905e-06,
      "loss": 0.4625980854034424,
      "num_input_tokens_seen": 8056808,
      "step": 255,
      "train_runtime": 670.0585,
      "train_tokens_per_second": 12024.037
    },
    {
      "epoch": 0.017506059789927284,
      "grad_norm": 1.0368124284942595,
      "learning_rate": 9.998124218478797e-06,
      "loss": 0.4769831657409668,
      "num_input_tokens_seen": 8213304,
      "step": 260,
      "train_runtime": 682.9,
      "train_tokens_per_second": 12027.096
    },
    {
      "epoch": 0.017842714785887424,
      "grad_norm": 1.1997697453880334,
      "learning_rate": 9.998051100159192e-06,
      "loss": 0.487578010559082,
      "num_input_tokens_seen": 8374848,
      "step": 265,
      "train_runtime": 695.4867,
      "train_tokens_per_second": 12041.708
    },
    {
      "epoch": 0.018179369781847563,
      "grad_norm": 1.201893381999332,
      "learning_rate": 9.997976584150533e-06,
      "loss": 0.4916214466094971,
      "num_input_tokens_seen": 8530464,
      "step": 270,
      "train_runtime": 708.2783,
      "train_tokens_per_second": 12043.944
    },
    {
      "epoch": 0.018516024777807703,
      "grad_norm": 1.0347058674493075,
      "learning_rate": 9.997900670473662e-06,
      "loss": 0.45754213333129884,
      "num_input_tokens_seen": 8694304,
      "step": 275,
      "train_runtime": 720.7436,
      "train_tokens_per_second": 12062.963
    },
    {
      "epoch": 0.018852679773767842,
      "grad_norm": 1.4544548395931927,
      "learning_rate": 9.997823359149804e-06,
      "loss": 0.46946043968200685,
      "num_input_tokens_seen": 8852488,
      "step": 280,
      "train_runtime": 733.9296,
      "train_tokens_per_second": 12061.768
    },
    {
      "epoch": 0.01918933476972798,
      "grad_norm": 1.1584230365138628,
      "learning_rate": 9.997744650200579e-06,
      "loss": 0.5050385951995849,
      "num_input_tokens_seen": 9012816,
      "step": 285,
      "train_runtime": 746.4205,
      "train_tokens_per_second": 12074.716
    },
    {
      "epoch": 0.01952598976568812,
      "grad_norm": 1.113884799754799,
      "learning_rate": 9.997664543648e-06,
      "loss": 0.48331484794616697,
      "num_input_tokens_seen": 9174280,
      "step": 290,
      "train_runtime": 760.3185,
      "train_tokens_per_second": 12066.364
    },
    {
      "epoch": 0.019862644761648264,
      "grad_norm": 1.019614337642649,
      "learning_rate": 9.997583039514469e-06,
      "loss": 0.47200641632080076,
      "num_input_tokens_seen": 9335248,
      "step": 295,
      "train_runtime": 772.846,
      "train_tokens_per_second": 12079.053
    },
    {
      "epoch": 0.020199299757608404,
      "grad_norm": 1.4804768664527224,
      "learning_rate": 9.997500137822776e-06,
      "loss": 0.4960646152496338,
      "num_input_tokens_seen": 9495888,
      "step": 300,
      "train_runtime": 785.7977,
      "train_tokens_per_second": 12084.392
    },
    {
      "epoch": 0.020535954753568543,
      "grad_norm": 1.1427763401474098,
      "learning_rate": 9.997415838596106e-06,
      "loss": 0.4839154243469238,
      "num_input_tokens_seen": 9654848,
      "step": 305,
      "train_runtime": 798.7343,
      "train_tokens_per_second": 12087.684
    },
    {
      "epoch": 0.020872609749528682,
      "grad_norm": 1.0110138422928279,
      "learning_rate": 9.997330141858032e-06,
      "loss": 0.5144286155700684,
      "num_input_tokens_seen": 9814600,
      "step": 310,
      "train_runtime": 813.0583,
      "train_tokens_per_second": 12071.213
    },
    {
      "epoch": 0.021209264745488822,
      "grad_norm": 1.105136204965642,
      "learning_rate": 9.997243047632519e-06,
      "loss": 0.4798739433288574,
      "num_input_tokens_seen": 9975904,
      "step": 315,
      "train_runtime": 826.9079,
      "train_tokens_per_second": 12064.105
    },
    {
      "epoch": 0.02154591974144896,
      "grad_norm": 1.1324864490597988,
      "learning_rate": 9.997154555943923e-06,
      "loss": 0.46115951538085936,
      "num_input_tokens_seen": 10135000,
      "step": 320,
      "train_runtime": 840.0743,
      "train_tokens_per_second": 12064.409
    },
    {
      "epoch": 0.021882574737409104,
      "grad_norm": 1.3005230058569233,
      "learning_rate": 9.997064666816992e-06,
      "loss": 0.45470561981201174,
      "num_input_tokens_seen": 10294608,
      "step": 325,
      "train_runtime": 852.5464,
      "train_tokens_per_second": 12075.129
    },
    {
      "epoch": 0.022219229733369244,
      "grad_norm": 1.2818583347173096,
      "learning_rate": 9.996973380276858e-06,
      "loss": 0.5623361587524414,
      "num_input_tokens_seen": 10453864,
      "step": 330,
      "train_runtime": 865.3451,
      "train_tokens_per_second": 12080.572
    },
    {
      "epoch": 0.022555884729329383,
      "grad_norm": 1.2407139272421785,
      "learning_rate": 9.996880696349053e-06,
      "loss": 0.4563101291656494,
      "num_input_tokens_seen": 10611128,
      "step": 335,
      "train_runtime": 877.9259,
      "train_tokens_per_second": 12086.588
    },
    {
      "epoch": 0.022892539725289523,
      "grad_norm": 1.207825477896831,
      "learning_rate": 9.996786615059497e-06,
      "loss": 0.5624344825744629,
      "num_input_tokens_seen": 10774352,
      "step": 340,
      "train_runtime": 891.5374,
      "train_tokens_per_second": 12085.137
    },
    {
      "epoch": 0.023229194721249662,
      "grad_norm": 1.0946064123587629,
      "learning_rate": 9.996691136434498e-06,
      "loss": 0.44589719772338865,
      "num_input_tokens_seen": 10935440,
      "step": 345,
      "train_runtime": 904.0983,
      "train_tokens_per_second": 12095.41
    },
    {
      "epoch": 0.023565849717209802,
      "grad_norm": 0.9586302067686466,
      "learning_rate": 9.996594260500753e-06,
      "loss": 0.510599422454834,
      "num_input_tokens_seen": 11096032,
      "step": 350,
      "train_runtime": 917.0355,
      "train_tokens_per_second": 12099.893
    },
    {
      "epoch": 0.023902504713169945,
      "grad_norm": 3.970662818569927,
      "learning_rate": 9.996495987285358e-06,
      "loss": 0.48085312843322753,
      "num_input_tokens_seen": 11257408,
      "step": 355,
      "train_runtime": 930.1712,
      "train_tokens_per_second": 12102.512
    },
    {
      "epoch": 0.024239159709130084,
      "grad_norm": 1.1902431964261908,
      "learning_rate": 9.996396316815793e-06,
      "loss": 0.46994600296020506,
      "num_input_tokens_seen": 11421040,
      "step": 360,
      "train_runtime": 943.2767,
      "train_tokens_per_second": 12107.836
    },
    {
      "epoch": 0.024575814705090224,
      "grad_norm": 1.267567993582934,
      "learning_rate": 9.99629524911993e-06,
      "loss": 0.4939387798309326,
      "num_input_tokens_seen": 11580896,
      "step": 365,
      "train_runtime": 956.953,
      "train_tokens_per_second": 12101.845
    },
    {
      "epoch": 0.024912469701050363,
      "grad_norm": 1.1519649584363651,
      "learning_rate": 9.996192784226033e-06,
      "loss": 0.454149866104126,
      "num_input_tokens_seen": 11733552,
      "step": 370,
      "train_runtime": 970.5711,
      "train_tokens_per_second": 12089.327
    },
    {
      "epoch": 0.025249124697010503,
      "grad_norm": 1.0319156097649216,
      "learning_rate": 9.996088922162755e-06,
      "loss": 0.47430944442749023,
      "num_input_tokens_seen": 11893936,
      "step": 375,
      "train_runtime": 983.1279,
      "train_tokens_per_second": 12098.056
    },
    {
      "epoch": 0.025585779692970642,
      "grad_norm": 1.1432443630299396,
      "learning_rate": 9.995983662959141e-06,
      "loss": 0.4670549392700195,
      "num_input_tokens_seen": 12053464,
      "step": 380,
      "train_runtime": 995.6173,
      "train_tokens_per_second": 12106.524
    },
    {
      "epoch": 0.025922434688930785,
      "grad_norm": 1.1662824248938957,
      "learning_rate": 9.995877006644627e-06,
      "loss": 0.48958525657653806,
      "num_input_tokens_seen": 12212832,
      "step": 385,
      "train_runtime": 1008.0928,
      "train_tokens_per_second": 12114.79
    },
    {
      "epoch": 0.026259089684890925,
      "grad_norm": 1.0514476882370498,
      "learning_rate": 9.995768953249038e-06,
      "loss": 0.4802351951599121,
      "num_input_tokens_seen": 12375632,
      "step": 390,
      "train_runtime": 1021.6958,
      "train_tokens_per_second": 12112.834
    },
    {
      "epoch": 0.026595744680851064,
      "grad_norm": 1.0063667401265024,
      "learning_rate": 9.995659502802594e-06,
      "loss": 0.47361931800842283,
      "num_input_tokens_seen": 12536216,
      "step": 395,
      "train_runtime": 1034.8136,
      "train_tokens_per_second": 12114.468
    },
    {
      "epoch": 0.026932399676811204,
      "grad_norm": 1.0493703848726703,
      "learning_rate": 9.995548655335899e-06,
      "loss": 0.48278303146362306,
      "num_input_tokens_seen": 12695352,
      "step": 400,
      "train_runtime": 1048.6981,
      "train_tokens_per_second": 12105.822
    },
    {
      "epoch": 0.027269054672771343,
      "grad_norm": 1.1152602747254188,
      "learning_rate": 9.995436410879951e-06,
      "loss": 0.524066162109375,
      "num_input_tokens_seen": 12855464,
      "step": 405,
      "train_runtime": 1061.8738,
      "train_tokens_per_second": 12106.396
    },
    {
      "epoch": 0.027605709668731482,
      "grad_norm": 1.074711634836309,
      "learning_rate": 9.99532276946614e-06,
      "loss": 0.4452226161956787,
      "num_input_tokens_seen": 13012488,
      "step": 410,
      "train_runtime": 1074.7137,
      "train_tokens_per_second": 12107.865
    },
    {
      "epoch": 0.027942364664691625,
      "grad_norm": 1.086494646437696,
      "learning_rate": 9.995207731126246e-06,
      "loss": 0.4838810920715332,
      "num_input_tokens_seen": 13176328,
      "step": 415,
      "train_runtime": 1087.1581,
      "train_tokens_per_second": 12119.974
    },
    {
      "epoch": 0.028279019660651765,
      "grad_norm": 1.0498946914612328,
      "learning_rate": 9.99509129589244e-06,
      "loss": 0.46650209426879885,
      "num_input_tokens_seen": 13337808,
      "step": 420,
      "train_runtime": 1101.109,
      "train_tokens_per_second": 12113.068
    },
    {
      "epoch": 0.028615674656611904,
      "grad_norm": 1.0337410324279293,
      "learning_rate": 9.994973463797279e-06,
      "loss": 0.46839241981506347,
      "num_input_tokens_seen": 13498832,
      "step": 425,
      "train_runtime": 1114.3492,
      "train_tokens_per_second": 12113.646
    },
    {
      "epoch": 0.028952329652572044,
      "grad_norm": 0.959037281710558,
      "learning_rate": 9.99485423487372e-06,
      "loss": 0.43642430305480956,
      "num_input_tokens_seen": 13656128,
      "step": 430,
      "train_runtime": 1127.1305,
      "train_tokens_per_second": 12115.836
    },
    {
      "epoch": 0.029288984648532183,
      "grad_norm": 1.0570573377555286,
      "learning_rate": 9.994733609155097e-06,
      "loss": 0.4827410697937012,
      "num_input_tokens_seen": 13813816,
      "step": 435,
      "train_runtime": 1140.299,
      "train_tokens_per_second": 12114.206
    },
    {
      "epoch": 0.029625639644492323,
      "grad_norm": 1.0550529434197375,
      "learning_rate": 9.994611586675149e-06,
      "loss": 0.46020212173461916,
      "num_input_tokens_seen": 13976104,
      "step": 440,
      "train_runtime": 1153.2254,
      "train_tokens_per_second": 12119.144
    },
    {
      "epoch": 0.029962294640452466,
      "grad_norm": 1.1474716966097382,
      "learning_rate": 9.994488167468e-06,
      "loss": 0.47103166580200195,
      "num_input_tokens_seen": 14134544,
      "step": 445,
      "train_runtime": 1166.8123,
      "train_tokens_per_second": 12113.811
    },
    {
      "epoch": 0.030298949636412605,
      "grad_norm": 0.9834376602833634,
      "learning_rate": 9.994363351568158e-06,
      "loss": 0.5048730850219727,
      "num_input_tokens_seen": 14291656,
      "step": 450,
      "train_runtime": 1180.3802,
      "train_tokens_per_second": 12107.672
    },
    {
      "epoch": 0.030635604632372745,
      "grad_norm": 1.0083779942246125,
      "learning_rate": 9.994237139010532e-06,
      "loss": 0.43943028450012206,
      "num_input_tokens_seen": 14452536,
      "step": 455,
      "train_runtime": 1194.4152,
      "train_tokens_per_second": 12100.094
    },
    {
      "epoch": 0.030972259628332884,
      "grad_norm": 1.0794597287319003,
      "learning_rate": 9.994109529830414e-06,
      "loss": 0.4924136161804199,
      "num_input_tokens_seen": 14616376,
      "step": 460,
      "train_runtime": 1206.8605,
      "train_tokens_per_second": 12111.073
    },
    {
      "epoch": 0.03130891462429303,
      "grad_norm": 0.9226732783084329,
      "learning_rate": 9.993980524063494e-06,
      "loss": 0.44845852851867674,
      "num_input_tokens_seen": 14775768,
      "step": 465,
      "train_runtime": 1220.0841,
      "train_tokens_per_second": 12110.45
    },
    {
      "epoch": 0.03164556962025317,
      "grad_norm": 0.9438662462795622,
      "learning_rate": 9.993850121745842e-06,
      "loss": 0.44841880798339845,
      "num_input_tokens_seen": 14933696,
      "step": 470,
      "train_runtime": 1233.4337,
      "train_tokens_per_second": 12107.416
    },
    {
      "epoch": 0.031982224616213306,
      "grad_norm": 1.2188633174885952,
      "learning_rate": 9.99371832291393e-06,
      "loss": 0.453628396987915,
      "num_input_tokens_seen": 15096136,
      "step": 475,
      "train_runtime": 1246.7957,
      "train_tokens_per_second": 12107.947
    },
    {
      "epoch": 0.032318879612173446,
      "grad_norm": 1.0759831528718735,
      "learning_rate": 9.993585127604612e-06,
      "loss": 0.4647801399230957,
      "num_input_tokens_seen": 15251304,
      "step": 480,
      "train_runtime": 1260.0513,
      "train_tokens_per_second": 12103.717
    },
    {
      "epoch": 0.032655534608133585,
      "grad_norm": 0.9560841791603152,
      "learning_rate": 9.993450535855136e-06,
      "loss": 0.45053730010986326,
      "num_input_tokens_seen": 15403056,
      "step": 485,
      "train_runtime": 1272.996,
      "train_tokens_per_second": 12099.847
    },
    {
      "epoch": 0.032992189604093725,
      "grad_norm": 1.1090118863756386,
      "learning_rate": 9.993314547703142e-06,
      "loss": 0.4434460163116455,
      "num_input_tokens_seen": 15563008,
      "step": 490,
      "train_runtime": 1286.256,
      "train_tokens_per_second": 12099.464
    },
    {
      "epoch": 0.033328844600053864,
      "grad_norm": 1.0830839786557283,
      "learning_rate": 9.993177163186655e-06,
      "loss": 0.4664647102355957,
      "num_input_tokens_seen": 15720104,
      "step": 495,
      "train_runtime": 1299.3267,
      "train_tokens_per_second": 12098.654
    },
    {
      "epoch": 0.033665499596014004,
      "grad_norm": 0.9820160136655524,
      "learning_rate": 9.993038382344097e-06,
      "loss": 0.4616697311401367,
      "num_input_tokens_seen": 15881912,
      "step": 500,
      "train_runtime": 1312.931,
      "train_tokens_per_second": 12096.532
    },
    {
      "epoch": 0.03400215459197414,
      "grad_norm": 1.0673159099564704,
      "learning_rate": 9.992898205214279e-06,
      "loss": 0.4627403259277344,
      "num_input_tokens_seen": 16043688,
      "step": 505,
      "train_runtime": 1326.1273,
      "train_tokens_per_second": 12098.151
    },
    {
      "epoch": 0.03433880958793428,
      "grad_norm": 0.9816448509608875,
      "learning_rate": 9.992756631836397e-06,
      "loss": 0.4467434883117676,
      "num_input_tokens_seen": 16202248,
      "step": 510,
      "train_runtime": 1338.7163,
      "train_tokens_per_second": 12102.825
    },
    {
      "epoch": 0.03467546458389442,
      "grad_norm": 0.9239724345264948,
      "learning_rate": 9.992613662250044e-06,
      "loss": 0.4464684009552002,
      "num_input_tokens_seen": 16362368,
      "step": 515,
      "train_runtime": 1351.278,
      "train_tokens_per_second": 12108.81
    },
    {
      "epoch": 0.03501211957985457,
      "grad_norm": 1.0208266697634873,
      "learning_rate": 9.992469296495201e-06,
      "loss": 0.47051296234130857,
      "num_input_tokens_seen": 16524408,
      "step": 520,
      "train_runtime": 1363.9154,
      "train_tokens_per_second": 12115.42
    },
    {
      "epoch": 0.03534877457581471,
      "grad_norm": 0.9247632887609387,
      "learning_rate": 9.992323534612238e-06,
      "loss": 0.4322772979736328,
      "num_input_tokens_seen": 16684832,
      "step": 525,
      "train_runtime": 1376.7774,
      "train_tokens_per_second": 12118.758
    },
    {
      "epoch": 0.03568542957177485,
      "grad_norm": 1.0983886927927062,
      "learning_rate": 9.99217637664192e-06,
      "loss": 0.4783305644989014,
      "num_input_tokens_seen": 16843328,
      "step": 530,
      "train_runtime": 1389.9174,
      "train_tokens_per_second": 12118.222
    },
    {
      "epoch": 0.03602208456773499,
      "grad_norm": 0.9935756434492818,
      "learning_rate": 9.992027822625397e-06,
      "loss": 0.47626543045043945,
      "num_input_tokens_seen": 17002608,
      "step": 535,
      "train_runtime": 1403.7351,
      "train_tokens_per_second": 12112.405
    },
    {
      "epoch": 0.036358739563695126,
      "grad_norm": 1.0577254554025446,
      "learning_rate": 9.99187787260421e-06,
      "loss": 0.44533343315124513,
      "num_input_tokens_seen": 17161128,
      "step": 540,
      "train_runtime": 1417.6336,
      "train_tokens_per_second": 12105.475
    },
    {
      "epoch": 0.036695394559655266,
      "grad_norm": 0.951402445225616,
      "learning_rate": 9.991726526620294e-06,
      "loss": 0.44655756950378417,
      "num_input_tokens_seen": 17323816,
      "step": 545,
      "train_runtime": 1430.3487,
      "train_tokens_per_second": 12111.604
    },
    {
      "epoch": 0.037032049555615405,
      "grad_norm": 1.0264088800049578,
      "learning_rate": 9.991573784715973e-06,
      "loss": 0.48914365768432616,
      "num_input_tokens_seen": 17485416,
      "step": 550,
      "train_runtime": 1443.3081,
      "train_tokens_per_second": 12114.819
    },
    {
      "epoch": 0.037368704551575545,
      "grad_norm": 1.551942927190482,
      "learning_rate": 9.99141964693396e-06,
      "loss": 0.47848825454711913,
      "num_input_tokens_seen": 17643336,
      "step": 555,
      "train_runtime": 1456.0373,
      "train_tokens_per_second": 12117.365
    },
    {
      "epoch": 0.037705359547535684,
      "grad_norm": 0.9768543659315769,
      "learning_rate": 9.99126411331736e-06,
      "loss": 0.45151920318603517,
      "num_input_tokens_seen": 17806512,
      "step": 560,
      "train_runtime": 1468.938,
      "train_tokens_per_second": 12122.031
    },
    {
      "epoch": 0.038042014543495824,
      "grad_norm": 1.1094768464979605,
      "learning_rate": 9.991107183909665e-06,
      "loss": 0.48669800758361814,
      "num_input_tokens_seen": 17967352,
      "step": 565,
      "train_runtime": 1482.238,
      "train_tokens_per_second": 12121.773
    },
    {
      "epoch": 0.03837866953945596,
      "grad_norm": 0.9882371790231235,
      "learning_rate": 9.990948858754763e-06,
      "loss": 0.4621871471405029,
      "num_input_tokens_seen": 18123856,
      "step": 570,
      "train_runtime": 1495.8694,
      "train_tokens_per_second": 12115.935
    },
    {
      "epoch": 0.0387153245354161,
      "grad_norm": 0.9516495724438331,
      "learning_rate": 9.990789137896927e-06,
      "loss": 0.4640963554382324,
      "num_input_tokens_seen": 18284408,
      "step": 575,
      "train_runtime": 1508.3987,
      "train_tokens_per_second": 12121.734
    },
    {
      "epoch": 0.03905197953137624,
      "grad_norm": 1.00191074793403,
      "learning_rate": 9.990628021380823e-06,
      "loss": 0.4907210826873779,
      "num_input_tokens_seen": 18444864,
      "step": 580,
      "train_runtime": 1521.2353,
      "train_tokens_per_second": 12124.925
    },
    {
      "epoch": 0.03938863452733639,
      "grad_norm": 1.1160586745314365,
      "learning_rate": 9.990465509251508e-06,
      "loss": 0.4765368938446045,
      "num_input_tokens_seen": 18599416,
      "step": 585,
      "train_runtime": 1534.4103,
      "train_tokens_per_second": 12121.54
    },
    {
      "epoch": 0.03972528952329653,
      "grad_norm": 0.9988597526211553,
      "learning_rate": 9.990301601554425e-06,
      "loss": 0.49938344955444336,
      "num_input_tokens_seen": 18753096,
      "step": 590,
      "train_runtime": 1547.2795,
      "train_tokens_per_second": 12120.045
    },
    {
      "epoch": 0.04006194451925667,
      "grad_norm": 1.2643211743220946,
      "learning_rate": 9.990136298335414e-06,
      "loss": 0.4231258869171143,
      "num_input_tokens_seen": 18911440,
      "step": 595,
      "train_runtime": 1560.4894,
      "train_tokens_per_second": 12118.916
    },
    {
      "epoch": 0.04039859951521681,
      "grad_norm": 0.9833512331323415,
      "learning_rate": 9.989969599640698e-06,
      "loss": 0.44031991958618166,
      "num_input_tokens_seen": 19071632,
      "step": 600,
      "train_runtime": 1573.2249,
      "train_tokens_per_second": 12122.636
    },
    {
      "epoch": 0.040735254511176947,
      "grad_norm": 0.9655762524888978,
      "learning_rate": 9.989801505516895e-06,
      "loss": 0.5200843334197998,
      "num_input_tokens_seen": 19235296,
      "step": 605,
      "train_runtime": 1585.712,
      "train_tokens_per_second": 12130.384
    },
    {
      "epoch": 0.041071909507137086,
      "grad_norm": 2.227105087763655,
      "learning_rate": 9.989632016011012e-06,
      "loss": 0.48972220420837403,
      "num_input_tokens_seen": 19394960,
      "step": 610,
      "train_runtime": 1598.957,
      "train_tokens_per_second": 12129.757
    },
    {
      "epoch": 0.041408564503097225,
      "grad_norm": 1.0816966835058763,
      "learning_rate": 9.989461131170448e-06,
      "loss": 0.49865164756774905,
      "num_input_tokens_seen": 19551032,
      "step": 615,
      "train_runtime": 1612.8609,
      "train_tokens_per_second": 12121.958
    },
    {
      "epoch": 0.041745219499057365,
      "grad_norm": 1.0211909000275807,
      "learning_rate": 9.98928885104299e-06,
      "loss": 0.4630419254302979,
      "num_input_tokens_seen": 19707904,
      "step": 620,
      "train_runtime": 1625.3538,
      "train_tokens_per_second": 12125.301
    },
    {
      "epoch": 0.042081874495017504,
      "grad_norm": 0.8870274603164188,
      "learning_rate": 9.989115175676812e-06,
      "loss": 0.41678524017333984,
      "num_input_tokens_seen": 19867840,
      "step": 625,
      "train_runtime": 1638.8019,
      "train_tokens_per_second": 12123.393
    },
    {
      "epoch": 0.042418529490977644,
      "grad_norm": 1.0973938509280696,
      "learning_rate": 9.988940105120483e-06,
      "loss": 0.47776055335998535,
      "num_input_tokens_seen": 20027256,
      "step": 630,
      "train_runtime": 1651.8885,
      "train_tokens_per_second": 12123.855
    },
    {
      "epoch": 0.04275518448693778,
      "grad_norm": 0.9754955473188553,
      "learning_rate": 9.988763639422965e-06,
      "loss": 0.49677300453186035,
      "num_input_tokens_seen": 20187928,
      "step": 635,
      "train_runtime": 1664.3713,
      "train_tokens_per_second": 12129.461
    },
    {
      "epoch": 0.04309183948289792,
      "grad_norm": 1.05710332676753,
      "learning_rate": 9.988585778633604e-06,
      "loss": 0.44564361572265626,
      "num_input_tokens_seen": 20337672,
      "step": 640,
      "train_runtime": 1677.242,
      "train_tokens_per_second": 12125.664
    },
    {
      "epoch": 0.04342849447885807,
      "grad_norm": 1.0755486143060007,
      "learning_rate": 9.988406522802134e-06,
      "loss": 0.4577667236328125,
      "num_input_tokens_seen": 20493384,
      "step": 645,
      "train_runtime": 1690.9599,
      "train_tokens_per_second": 12119.379
    },
    {
      "epoch": 0.04376514947481821,
      "grad_norm": 0.9957485095541426,
      "learning_rate": 9.98822587197869e-06,
      "loss": 0.41981868743896483,
      "num_input_tokens_seen": 20652248,
      "step": 650,
      "train_runtime": 1703.6982,
      "train_tokens_per_second": 12122.011
    },
    {
      "epoch": 0.04410180447077835,
      "grad_norm": 1.097361010079167,
      "learning_rate": 9.988043826213785e-06,
      "loss": 0.4444459915161133,
      "num_input_tokens_seen": 20810184,
      "step": 655,
      "train_runtime": 1716.7975,
      "train_tokens_per_second": 12121.514
    },
    {
      "epoch": 0.04443845946673849,
      "grad_norm": 0.9688622586590958,
      "learning_rate": 9.98786038555833e-06,
      "loss": 0.4006513595581055,
      "num_input_tokens_seen": 20971432,
      "step": 660,
      "train_runtime": 1730.1133,
      "train_tokens_per_second": 12121.421
    },
    {
      "epoch": 0.04477511446269863,
      "grad_norm": 1.065686120976239,
      "learning_rate": 9.987675550063625e-06,
      "loss": 0.4553042411804199,
      "num_input_tokens_seen": 21132776,
      "step": 665,
      "train_runtime": 1743.2392,
      "train_tokens_per_second": 12122.706
    },
    {
      "epoch": 0.04511176945865877,
      "grad_norm": 1.2388907576786294,
      "learning_rate": 9.987489319781355e-06,
      "loss": 0.4581089973449707,
      "num_input_tokens_seen": 21291848,
      "step": 670,
      "train_runtime": 1757.0713,
      "train_tokens_per_second": 12117.805
    },
    {
      "epoch": 0.045448424454618906,
      "grad_norm": 0.9382466004733375,
      "learning_rate": 9.987301694763602e-06,
      "loss": 0.5131166934967041,
      "num_input_tokens_seen": 21449872,
      "step": 675,
      "train_runtime": 1770.1588,
      "train_tokens_per_second": 12117.484
    },
    {
      "epoch": 0.045785079450579046,
      "grad_norm": 1.0397110435257384,
      "learning_rate": 9.987112675062834e-06,
      "loss": 0.4381747245788574,
      "num_input_tokens_seen": 21608688,
      "step": 680,
      "train_runtime": 1782.9453,
      "train_tokens_per_second": 12119.659
    },
    {
      "epoch": 0.046121734446539185,
      "grad_norm": 0.9636033745701335,
      "learning_rate": 9.986922260731908e-06,
      "loss": 0.4304304599761963,
      "num_input_tokens_seen": 21765680,
      "step": 685,
      "train_runtime": 1796.8322,
      "train_tokens_per_second": 12113.363
    },
    {
      "epoch": 0.046458389442499325,
      "grad_norm": 0.8540414275451069,
      "learning_rate": 9.986730451824073e-06,
      "loss": 0.47147550582885744,
      "num_input_tokens_seen": 21923856,
      "step": 690,
      "train_runtime": 1809.5653,
      "train_tokens_per_second": 12115.537
    },
    {
      "epoch": 0.046795044438459464,
      "grad_norm": 0.9872550347858896,
      "learning_rate": 9.986537248392969e-06,
      "loss": 0.47453083992004397,
      "num_input_tokens_seen": 22086424,
      "step": 695,
      "train_runtime": 1823.0195,
      "train_tokens_per_second": 12115.298
    },
    {
      "epoch": 0.047131699434419604,
      "grad_norm": 1.1266168723622052,
      "learning_rate": 9.986342650492626e-06,
      "loss": 0.4527026653289795,
      "num_input_tokens_seen": 22249592,
      "step": 700,
      "train_runtime": 1836.1295,
      "train_tokens_per_second": 12117.659
    },
    {
      "epoch": 0.04746835443037975,
      "grad_norm": 0.9400657818178781,
      "learning_rate": 9.98614665817746e-06,
      "loss": 0.4358218193054199,
      "num_input_tokens_seen": 22404016,
      "step": 705,
      "train_runtime": 1849.5818,
      "train_tokens_per_second": 12113.017
    },
    {
      "epoch": 0.04780500942633989,
      "grad_norm": 1.1061918007274816,
      "learning_rate": 9.98594927150228e-06,
      "loss": 0.4545723915100098,
      "num_input_tokens_seen": 22567856,
      "step": 710,
      "train_runtime": 1862.0175,
      "train_tokens_per_second": 12120.11
    },
    {
      "epoch": 0.04814166442230003,
      "grad_norm": 0.9213558129731466,
      "learning_rate": 9.985750490522286e-06,
      "loss": 0.42123003005981446,
      "num_input_tokens_seen": 22727280,
      "step": 715,
      "train_runtime": 1876.2106,
      "train_tokens_per_second": 12113.395
    },
    {
      "epoch": 0.04847831941826017,
      "grad_norm": 0.9594189723125057,
      "learning_rate": 9.985550315293066e-06,
      "loss": 0.4831345558166504,
      "num_input_tokens_seen": 22891120,
      "step": 720,
      "train_runtime": 1888.6573,
      "train_tokens_per_second": 12120.314
    },
    {
      "epoch": 0.04881497441422031,
      "grad_norm": 1.069807928188195,
      "learning_rate": 9.985348745870598e-06,
      "loss": 0.47385759353637696,
      "num_input_tokens_seen": 23051680,
      "step": 725,
      "train_runtime": 1901.7329,
      "train_tokens_per_second": 12121.408
    },
    {
      "epoch": 0.04915162941018045,
      "grad_norm": 0.8975497520665994,
      "learning_rate": 9.98514578231125e-06,
      "loss": 0.4108592510223389,
      "num_input_tokens_seen": 23213160,
      "step": 730,
      "train_runtime": 1914.5418,
      "train_tokens_per_second": 12124.656
    },
    {
      "epoch": 0.04948828440614059,
      "grad_norm": 0.9493558709229178,
      "learning_rate": 9.984941424671781e-06,
      "loss": 0.4824410915374756,
      "num_input_tokens_seen": 23374792,
      "step": 735,
      "train_runtime": 1928.375,
      "train_tokens_per_second": 12121.497
    },
    {
      "epoch": 0.049824939402100726,
      "grad_norm": 1.0344641837261108,
      "learning_rate": 9.984735673009337e-06,
      "loss": 0.4443234443664551,
      "num_input_tokens_seen": 23535176,
      "step": 740,
      "train_runtime": 1941.4685,
      "train_tokens_per_second": 12122.358
    },
    {
      "epoch": 0.050161594398060866,
      "grad_norm": 0.9580131202241611,
      "learning_rate": 9.984528527381458e-06,
      "loss": 0.45436906814575195,
      "num_input_tokens_seen": 23690408,
      "step": 745,
      "train_runtime": 1954.1461,
      "train_tokens_per_second": 12123.151
    },
    {
      "epoch": 0.050498249394021005,
      "grad_norm": 1.0597551372281049,
      "learning_rate": 9.984319987846072e-06,
      "loss": 0.4441669464111328,
      "num_input_tokens_seen": 23849512,
      "step": 750,
      "train_runtime": 1967.5717,
      "train_tokens_per_second": 12121.292
    },
    {
      "epoch": 0.050834904389981145,
      "grad_norm": 1.091777532592783,
      "learning_rate": 9.984110054461495e-06,
      "loss": 0.4507458686828613,
      "num_input_tokens_seen": 24013352,
      "step": 755,
      "train_runtime": 1980.0144,
      "train_tokens_per_second": 12127.867
    },
    {
      "epoch": 0.051171559385941284,
      "grad_norm": 0.8403560730705298,
      "learning_rate": 9.983898727286435e-06,
      "loss": 0.43334531784057617,
      "num_input_tokens_seen": 24174008,
      "step": 760,
      "train_runtime": 1992.5423,
      "train_tokens_per_second": 12132.243
    },
    {
      "epoch": 0.05150821438190143,
      "grad_norm": 1.0556905650410375,
      "learning_rate": 9.983686006379985e-06,
      "loss": 0.47033281326293946,
      "num_input_tokens_seen": 24331120,
      "step": 765,
      "train_runtime": 2005.6323,
      "train_tokens_per_second": 12131.396
    },
    {
      "epoch": 0.05184486937786157,
      "grad_norm": 0.9955114482895413,
      "learning_rate": 9.983471891801638e-06,
      "loss": 0.4417285919189453,
      "num_input_tokens_seen": 24491688,
      "step": 770,
      "train_runtime": 2018.1024,
      "train_tokens_per_second": 12135.998
    },
    {
      "epoch": 0.05218152437382171,
      "grad_norm": 1.0914149014820869,
      "learning_rate": 9.983256383611267e-06,
      "loss": 0.4434773921966553,
      "num_input_tokens_seen": 24647920,
      "step": 775,
      "train_runtime": 2030.5642,
      "train_tokens_per_second": 12138.459
    },
    {
      "epoch": 0.05251817936978185,
      "grad_norm": 1.0353304994801185,
      "learning_rate": 9.983039481869138e-06,
      "loss": 0.43852434158325193,
      "num_input_tokens_seen": 24807688,
      "step": 780,
      "train_runtime": 2043.5833,
      "train_tokens_per_second": 12139.309
    },
    {
      "epoch": 0.05285483436574199,
      "grad_norm": 0.9713045784537268,
      "learning_rate": 9.98282118663591e-06,
      "loss": 0.44553141593933104,
      "num_input_tokens_seen": 24970376,
      "step": 785,
      "train_runtime": 2056.2335,
      "train_tokens_per_second": 12143.745
    },
    {
      "epoch": 0.05319148936170213,
      "grad_norm": 0.9394098983302592,
      "learning_rate": 9.982601497972624e-06,
      "loss": 0.45570807456970214,
      "num_input_tokens_seen": 25129328,
      "step": 790,
      "train_runtime": 2069.1329,
      "train_tokens_per_second": 12144.859
    },
    {
      "epoch": 0.05352814435766227,
      "grad_norm": 1.0032209046748868,
      "learning_rate": 9.982380415940717e-06,
      "loss": 0.4181360721588135,
      "num_input_tokens_seen": 25287752,
      "step": 795,
      "train_runtime": 2083.1836,
      "train_tokens_per_second": 12138.993
    },
    {
      "epoch": 0.05386479935362241,
      "grad_norm": 0.9744314586907671,
      "learning_rate": 9.982157940602015e-06,
      "loss": 0.4843122959136963,
      "num_input_tokens_seen": 25450064,
      "step": 800,
      "train_runtime": 2096.4053,
      "train_tokens_per_second": 12139.859
    },
    {
      "epoch": 0.05420145434958255,
      "grad_norm": 0.9501135820643248,
      "learning_rate": 9.981934072018732e-06,
      "loss": 0.468695068359375,
      "num_input_tokens_seen": 25610400,
      "step": 805,
      "train_runtime": 2108.8645,
      "train_tokens_per_second": 12144.166
    },
    {
      "epoch": 0.054538109345542686,
      "grad_norm": 0.9230825263230974,
      "learning_rate": 9.981708810253472e-06,
      "loss": 0.44299874305725095,
      "num_input_tokens_seen": 25770208,
      "step": 810,
      "train_runtime": 2122.0915,
      "train_tokens_per_second": 12143.778
    },
    {
      "epoch": 0.054874764341502826,
      "grad_norm": 1.068961383822381,
      "learning_rate": 9.981482155369229e-06,
      "loss": 0.46859169006347656,
      "num_input_tokens_seen": 25930424,
      "step": 815,
      "train_runtime": 2134.6437,
      "train_tokens_per_second": 12147.425
    },
    {
      "epoch": 0.055211419337462965,
      "grad_norm": 1.1127926351239978,
      "learning_rate": 9.981254107429385e-06,
      "loss": 0.4283432960510254,
      "num_input_tokens_seen": 26086816,
      "step": 820,
      "train_runtime": 2147.5432,
      "train_tokens_per_second": 12147.284
    },
    {
      "epoch": 0.05554807433342311,
      "grad_norm": 0.9379990425064615,
      "learning_rate": 9.981024666497716e-06,
      "loss": 0.4239189147949219,
      "num_input_tokens_seen": 26246576,
      "step": 825,
      "train_runtime": 2160.7057,
      "train_tokens_per_second": 12147.224
    },
    {
      "epoch": 0.05588472932938325,
      "grad_norm": 0.9832051634385591,
      "learning_rate": 9.98079383263838e-06,
      "loss": 0.4533884048461914,
      "num_input_tokens_seen": 26399456,
      "step": 830,
      "train_runtime": 2173.7939,
      "train_tokens_per_second": 12144.415
    },
    {
      "epoch": 0.05622138432534339,
      "grad_norm": 1.0348712759741439,
      "learning_rate": 9.980561605915932e-06,
      "loss": 0.44814014434814453,
      "num_input_tokens_seen": 26561712,
      "step": 835,
      "train_runtime": 2188.2651,
      "train_tokens_per_second": 12138.252
    },
    {
      "epoch": 0.05655803932130353,
      "grad_norm": 0.9937143809000505,
      "learning_rate": 9.980327986395316e-06,
      "loss": 0.45450782775878906,
      "num_input_tokens_seen": 26719000,
      "step": 840,
      "train_runtime": 2201.5501,
      "train_tokens_per_second": 12136.449
    },
    {
      "epoch": 0.05689469431726367,
      "grad_norm": 0.878069446594541,
      "learning_rate": 9.980092974141855e-06,
      "loss": 0.44374475479125974,
      "num_input_tokens_seen": 26879984,
      "step": 845,
      "train_runtime": 2214.5802,
      "train_tokens_per_second": 12137.733
    },
    {
      "epoch": 0.05723134931322381,
      "grad_norm": 0.981586328607755,
      "learning_rate": 9.979856569221276e-06,
      "loss": 0.4550945281982422,
      "num_input_tokens_seen": 27039752,
      "step": 850,
      "train_runtime": 2227.5365,
      "train_tokens_per_second": 12138.859
    },
    {
      "epoch": 0.05756800430918395,
      "grad_norm": 0.9578561598243582,
      "learning_rate": 9.979618771699688e-06,
      "loss": 0.48041372299194335,
      "num_input_tokens_seen": 27202888,
      "step": 855,
      "train_runtime": 2241.1941,
      "train_tokens_per_second": 12137.676
    },
    {
      "epoch": 0.05790465930514409,
      "grad_norm": 1.0554037548126747,
      "learning_rate": 9.979379581643588e-06,
      "loss": 0.42430791854858396,
      "num_input_tokens_seen": 27361456,
      "step": 860,
      "train_runtime": 2254.4805,
      "train_tokens_per_second": 12136.479
    },
    {
      "epoch": 0.05824131430110423,
      "grad_norm": 0.9964914745895169,
      "learning_rate": 9.979138999119866e-06,
      "loss": 0.4808645248413086,
      "num_input_tokens_seen": 27520968,
      "step": 865,
      "train_runtime": 2267.3816,
      "train_tokens_per_second": 12137.775
    },
    {
      "epoch": 0.05857796929706437,
      "grad_norm": 0.9998883552859759,
      "learning_rate": 9.978897024195801e-06,
      "loss": 0.446726131439209,
      "num_input_tokens_seen": 27684808,
      "step": 870,
      "train_runtime": 2279.831,
      "train_tokens_per_second": 12143.36
    },
    {
      "epoch": 0.058914624293024506,
      "grad_norm": 0.9213470897090208,
      "learning_rate": 9.978653656939058e-06,
      "loss": 0.4312605857849121,
      "num_input_tokens_seen": 27847040,
      "step": 875,
      "train_runtime": 2294.1163,
      "train_tokens_per_second": 12138.46
    },
    {
      "epoch": 0.059251279288984646,
      "grad_norm": 1.2113903884256734,
      "learning_rate": 9.978408897417699e-06,
      "loss": 0.4486079692840576,
      "num_input_tokens_seen": 28007872,
      "step": 880,
      "train_runtime": 2306.8486,
      "train_tokens_per_second": 12141.183
    },
    {
      "epoch": 0.059587934284944785,
      "grad_norm": 0.9378337150853849,
      "learning_rate": 9.978162745700165e-06,
      "loss": 0.4522806167602539,
      "num_input_tokens_seen": 28168824,
      "step": 885,
      "train_runtime": 2319.9217,
      "train_tokens_per_second": 12142.144
    },
    {
      "epoch": 0.05992458928090493,
      "grad_norm": 0.9045748055251455,
      "learning_rate": 9.977915201855293e-06,
      "loss": 0.4393168449401855,
      "num_input_tokens_seen": 28329240,
      "step": 890,
      "train_runtime": 2333.4008,
      "train_tokens_per_second": 12140.752
    },
    {
      "epoch": 0.06026124427686507,
      "grad_norm": 0.9751095599503165,
      "learning_rate": 9.977666265952309e-06,
      "loss": 0.456513786315918,
      "num_input_tokens_seen": 28488464,
      "step": 895,
      "train_runtime": 2346.3956,
      "train_tokens_per_second": 12141.373
    },
    {
      "epoch": 0.06059789927282521,
      "grad_norm": 0.9753716871374147,
      "learning_rate": 9.977415938060823e-06,
      "loss": 0.4530789852142334,
      "num_input_tokens_seen": 28646600,
      "step": 900,
      "train_runtime": 2359.6604,
      "train_tokens_per_second": 12140.137
    },
    {
      "epoch": 0.06093455426878535,
      "grad_norm": 0.9588995853125265,
      "learning_rate": 9.977164218250846e-06,
      "loss": 0.44556493759155275,
      "num_input_tokens_seen": 28805688,
      "step": 905,
      "train_runtime": 2372.5925,
      "train_tokens_per_second": 12141.018
    },
    {
      "epoch": 0.06127120926474549,
      "grad_norm": 0.9668287099742489,
      "learning_rate": 9.976911106592765e-06,
      "loss": 0.4143209457397461,
      "num_input_tokens_seen": 28961112,
      "step": 910,
      "train_runtime": 2385.8131,
      "train_tokens_per_second": 12138.886
    },
    {
      "epoch": 0.06160786426070563,
      "grad_norm": 0.9914823534100267,
      "learning_rate": 9.976656603157363e-06,
      "loss": 0.4627963066101074,
      "num_input_tokens_seen": 29121760,
      "step": 915,
      "train_runtime": 2399.6101,
      "train_tokens_per_second": 12136.038
    },
    {
      "epoch": 0.06194451925666577,
      "grad_norm": 0.9814944984261621,
      "learning_rate": 9.976400708015811e-06,
      "loss": 0.4357560157775879,
      "num_input_tokens_seen": 29279904,
      "step": 920,
      "train_runtime": 2413.1364,
      "train_tokens_per_second": 12133.547
    },
    {
      "epoch": 0.06228117425262591,
      "grad_norm": 0.9717162351659326,
      "learning_rate": 9.97614342123967e-06,
      "loss": 0.4263772487640381,
      "num_input_tokens_seen": 29442280,
      "step": 925,
      "train_runtime": 2425.8257,
      "train_tokens_per_second": 12137.014
    },
    {
      "epoch": 0.06261782924858605,
      "grad_norm": 0.9536663678940152,
      "learning_rate": 9.97588474290089e-06,
      "loss": 0.4196614265441895,
      "num_input_tokens_seen": 29601608,
      "step": 930,
      "train_runtime": 2439.3365,
      "train_tokens_per_second": 12135.106
    },
    {
      "epoch": 0.0629544842445462,
      "grad_norm": 1.1042378197565843,
      "learning_rate": 9.975624673071806e-06,
      "loss": 0.4309551239013672,
      "num_input_tokens_seen": 29760856,
      "step": 935,
      "train_runtime": 2452.7937,
      "train_tokens_per_second": 12133.452
    },
    {
      "epoch": 0.06329113924050633,
      "grad_norm": 0.9721153951582121,
      "learning_rate": 9.97536321182515e-06,
      "loss": 0.4095101833343506,
      "num_input_tokens_seen": 29914728,
      "step": 940,
      "train_runtime": 2465.8468,
      "train_tokens_per_second": 12131.625
    },
    {
      "epoch": 0.06362779423646647,
      "grad_norm": 0.964650368114325,
      "learning_rate": 9.975100359234037e-06,
      "loss": 0.47099714279174804,
      "num_input_tokens_seen": 30076816,
      "step": 945,
      "train_runtime": 2479.1451,
      "train_tokens_per_second": 12131.93
    },
    {
      "epoch": 0.06396444923242661,
      "grad_norm": 0.9596190554790406,
      "learning_rate": 9.974836115371972e-06,
      "loss": 0.45456132888793943,
      "num_input_tokens_seen": 30239696,
      "step": 950,
      "train_runtime": 2492.7665,
      "train_tokens_per_second": 12130.978
    },
    {
      "epoch": 0.06430110422838675,
      "grad_norm": 0.9238515492144911,
      "learning_rate": 9.974570480312852e-06,
      "loss": 0.43305373191833496,
      "num_input_tokens_seen": 30403536,
      "step": 955,
      "train_runtime": 2505.2275,
      "train_tokens_per_second": 12136.038
    },
    {
      "epoch": 0.06463775922434689,
      "grad_norm": 0.8960003111608499,
      "learning_rate": 9.974303454130962e-06,
      "loss": 0.45706024169921877,
      "num_input_tokens_seen": 30557376,
      "step": 960,
      "train_runtime": 2519.1244,
      "train_tokens_per_second": 12130.158
    },
    {
      "epoch": 0.06497441422030703,
      "grad_norm": 0.9160566129873569,
      "learning_rate": 9.97403503690097e-06,
      "loss": 0.4423353672027588,
      "num_input_tokens_seen": 30717464,
      "step": 965,
      "train_runtime": 2531.8805,
      "train_tokens_per_second": 12132.272
    },
    {
      "epoch": 0.06531106921626717,
      "grad_norm": 0.9556030749245238,
      "learning_rate": 9.973765228697942e-06,
      "loss": 0.41098475456237793,
      "num_input_tokens_seen": 30876392,
      "step": 970,
      "train_runtime": 2545.4042,
      "train_tokens_per_second": 12130.251
    },
    {
      "epoch": 0.06564772421222731,
      "grad_norm": 0.9421151616624401,
      "learning_rate": 9.973494029597327e-06,
      "loss": 0.4436800956726074,
      "num_input_tokens_seen": 31028488,
      "step": 975,
      "train_runtime": 2558.5511,
      "train_tokens_per_second": 12127.367
    },
    {
      "epoch": 0.06598437920818745,
      "grad_norm": 1.1836654392332189,
      "learning_rate": 9.973221439674968e-06,
      "loss": 0.4125687599182129,
      "num_input_tokens_seen": 31186536,
      "step": 980,
      "train_runtime": 2571.8573,
      "train_tokens_per_second": 12126.075
    },
    {
      "epoch": 0.06632103420414759,
      "grad_norm": 0.9154423214731736,
      "learning_rate": 9.972947459007089e-06,
      "loss": 0.450439977645874,
      "num_input_tokens_seen": 31343544,
      "step": 985,
      "train_runtime": 2585.0281,
      "train_tokens_per_second": 12125.03
    },
    {
      "epoch": 0.06665768920010773,
      "grad_norm": 0.9835095618826193,
      "learning_rate": 9.972672087670313e-06,
      "loss": 0.44240775108337405,
      "num_input_tokens_seen": 31503376,
      "step": 990,
      "train_runtime": 2597.7517,
      "train_tokens_per_second": 12127.17
    },
    {
      "epoch": 0.06699434419606787,
      "grad_norm": 0.8920790454204629,
      "learning_rate": 9.972395325741644e-06,
      "loss": 0.4211949348449707,
      "num_input_tokens_seen": 31662736,
      "step": 995,
      "train_runtime": 2610.5234,
      "train_tokens_per_second": 12128.884
    },
    {
      "epoch": 0.06733099919202801,
      "grad_norm": 0.8739919784346051,
      "learning_rate": 9.972117173298476e-06,
      "loss": 0.3989120960235596,
      "num_input_tokens_seen": 31821072,
      "step": 1000,
      "train_runtime": 2624.7352,
      "train_tokens_per_second": 12123.536
    },
    {
      "epoch": 0.06766765418798815,
      "grad_norm": 0.9426689708047346,
      "learning_rate": 9.971837630418597e-06,
      "loss": 0.42686920166015624,
      "num_input_tokens_seen": 31980040,
      "step": 1005,
      "train_runtime": 2638.3383,
      "train_tokens_per_second": 12121.281
    },
    {
      "epoch": 0.06800430918394829,
      "grad_norm": 0.9261265568747744,
      "learning_rate": 9.971556697180179e-06,
      "loss": 0.45174551010131836,
      "num_input_tokens_seen": 32135704,
      "step": 1010,
      "train_runtime": 2651.1069,
      "train_tokens_per_second": 12121.618
    },
    {
      "epoch": 0.06834096417990843,
      "grad_norm": 0.9533090303087755,
      "learning_rate": 9.971274373661784e-06,
      "loss": 0.44016642570495607,
      "num_input_tokens_seen": 32293272,
      "step": 1015,
      "train_runtime": 2663.6674,
      "train_tokens_per_second": 12123.613
    },
    {
      "epoch": 0.06867761917586857,
      "grad_norm": 0.9262311080176922,
      "learning_rate": 9.97099065994236e-06,
      "loss": 0.41562590599060056,
      "num_input_tokens_seen": 32455576,
      "step": 1020,
      "train_runtime": 2676.1481,
      "train_tokens_per_second": 12127.721
    },
    {
      "epoch": 0.0690142741718287,
      "grad_norm": 0.8953125099202213,
      "learning_rate": 9.970705556101253e-06,
      "loss": 0.45358028411865237,
      "num_input_tokens_seen": 32611264,
      "step": 1025,
      "train_runtime": 2690.3807,
      "train_tokens_per_second": 12121.431
    },
    {
      "epoch": 0.06935092916778884,
      "grad_norm": 0.9163240638824789,
      "learning_rate": 9.970419062218186e-06,
      "loss": 0.46768550872802733,
      "num_input_tokens_seen": 32774528,
      "step": 1030,
      "train_runtime": 2703.239,
      "train_tokens_per_second": 12124.169
    },
    {
      "epoch": 0.06968758416374898,
      "grad_norm": 0.9603827773348432,
      "learning_rate": 9.970131178373276e-06,
      "loss": 0.48101062774658204,
      "num_input_tokens_seen": 32934904,
      "step": 1035,
      "train_runtime": 2715.744,
      "train_tokens_per_second": 12127.396
    },
    {
      "epoch": 0.07002423915970914,
      "grad_norm": 1.0282896743824106,
      "learning_rate": 9.969841904647033e-06,
      "loss": 0.48900203704833983,
      "num_input_tokens_seen": 33092888,
      "step": 1040,
      "train_runtime": 2728.7943,
      "train_tokens_per_second": 12127.293
    },
    {
      "epoch": 0.07036089415566928,
      "grad_norm": 0.9804303127303902,
      "learning_rate": 9.969551241120349e-06,
      "loss": 0.442124080657959,
      "num_input_tokens_seen": 33254368,
      "step": 1045,
      "train_runtime": 2742.0081,
      "train_tokens_per_second": 12127.743
    },
    {
      "epoch": 0.07069754915162942,
      "grad_norm": 0.9257510383264234,
      "learning_rate": 9.969259187874507e-06,
      "loss": 0.43998470306396487,
      "num_input_tokens_seen": 33415336,
      "step": 1050,
      "train_runtime": 2754.6516,
      "train_tokens_per_second": 12130.513
    },
    {
      "epoch": 0.07103420414758956,
      "grad_norm": 0.9155175873229721,
      "learning_rate": 9.968965744991178e-06,
      "loss": 0.4427349090576172,
      "num_input_tokens_seen": 33576136,
      "step": 1055,
      "train_runtime": 2768.1845,
      "train_tokens_per_second": 12129.298
    },
    {
      "epoch": 0.0713708591435497,
      "grad_norm": 1.016727740243101,
      "learning_rate": 9.968670912552422e-06,
      "loss": 0.42871675491333006,
      "num_input_tokens_seen": 33735992,
      "step": 1060,
      "train_runtime": 2780.9264,
      "train_tokens_per_second": 12131.206
    },
    {
      "epoch": 0.07170751413950983,
      "grad_norm": 0.9472032533692066,
      "learning_rate": 9.968374690640692e-06,
      "loss": 0.42758769989013673,
      "num_input_tokens_seen": 33897520,
      "step": 1065,
      "train_runtime": 2793.4592,
      "train_tokens_per_second": 12134.603
    },
    {
      "epoch": 0.07204416913546997,
      "grad_norm": 0.9520158989514792,
      "learning_rate": 9.968077079338822e-06,
      "loss": 0.4488554000854492,
      "num_input_tokens_seen": 34057720,
      "step": 1070,
      "train_runtime": 2806.3094,
      "train_tokens_per_second": 12136.124
    },
    {
      "epoch": 0.07238082413143011,
      "grad_norm": 0.9911935639040145,
      "learning_rate": 9.967778078730038e-06,
      "loss": 0.4689912796020508,
      "num_input_tokens_seen": 34220392,
      "step": 1075,
      "train_runtime": 2819.3589,
      "train_tokens_per_second": 12137.65
    },
    {
      "epoch": 0.07271747912739025,
      "grad_norm": 0.9386121465327312,
      "learning_rate": 9.967477688897957e-06,
      "loss": 0.4770233631134033,
      "num_input_tokens_seen": 34380368,
      "step": 1080,
      "train_runtime": 2832.8794,
      "train_tokens_per_second": 12136.192
    },
    {
      "epoch": 0.07305413412335039,
      "grad_norm": 0.849002130220033,
      "learning_rate": 9.967175909926578e-06,
      "loss": 0.45032968521118166,
      "num_input_tokens_seen": 34541920,
      "step": 1085,
      "train_runtime": 2845.9238,
      "train_tokens_per_second": 12137.331
    },
    {
      "epoch": 0.07339078911931053,
      "grad_norm": 1.0708298307957156,
      "learning_rate": 9.966872741900296e-06,
      "loss": 0.4828330516815186,
      "num_input_tokens_seen": 34705120,
      "step": 1090,
      "train_runtime": 2858.8246,
      "train_tokens_per_second": 12139.647
    },
    {
      "epoch": 0.07372744411527067,
      "grad_norm": 1.0832687212391385,
      "learning_rate": 9.96656818490389e-06,
      "loss": 0.46305160522460936,
      "num_input_tokens_seen": 34866424,
      "step": 1095,
      "train_runtime": 2871.2932,
      "train_tokens_per_second": 12143.108
    },
    {
      "epoch": 0.07406409911123081,
      "grad_norm": 0.9632528346713082,
      "learning_rate": 9.966262239022529e-06,
      "loss": 0.48505916595458987,
      "num_input_tokens_seen": 35026944,
      "step": 1100,
      "train_runtime": 2884.2426,
      "train_tokens_per_second": 12144.243
    },
    {
      "epoch": 0.07440075410719095,
      "grad_norm": 0.9924514881032961,
      "learning_rate": 9.965954904341768e-06,
      "loss": 0.4741023540496826,
      "num_input_tokens_seen": 35186824,
      "step": 1105,
      "train_runtime": 2898.0253,
      "train_tokens_per_second": 12141.655
    },
    {
      "epoch": 0.07473740910315109,
      "grad_norm": 0.901432988676257,
      "learning_rate": 9.965646180947555e-06,
      "loss": 0.44626588821411134,
      "num_input_tokens_seen": 35344536,
      "step": 1110,
      "train_runtime": 2910.6383,
      "train_tokens_per_second": 12143.225
    },
    {
      "epoch": 0.07507406409911123,
      "grad_norm": 0.9166866632852102,
      "learning_rate": 9.96533606892622e-06,
      "loss": 0.42901973724365233,
      "num_input_tokens_seen": 35502296,
      "step": 1115,
      "train_runtime": 2924.5765,
      "train_tokens_per_second": 12139.295
    },
    {
      "epoch": 0.07541071909507137,
      "grad_norm": 0.9226116187763157,
      "learning_rate": 9.965024568364487e-06,
      "loss": 0.4359426975250244,
      "num_input_tokens_seen": 35665888,
      "step": 1120,
      "train_runtime": 2937.62,
      "train_tokens_per_second": 12141.083
    },
    {
      "epoch": 0.07574737409103151,
      "grad_norm": 0.9440127468004684,
      "learning_rate": 9.964711679349467e-06,
      "loss": 0.42251172065734866,
      "num_input_tokens_seen": 35820432,
      "step": 1125,
      "train_runtime": 2950.7812,
      "train_tokens_per_second": 12139.305
    },
    {
      "epoch": 0.07608402908699165,
      "grad_norm": 1.3678719819459766,
      "learning_rate": 9.964397401968656e-06,
      "loss": 0.4953521728515625,
      "num_input_tokens_seen": 35978544,
      "step": 1130,
      "train_runtime": 2963.2708,
      "train_tokens_per_second": 12141.497
    },
    {
      "epoch": 0.07642068408295179,
      "grad_norm": 0.9608408454067201,
      "learning_rate": 9.964081736309942e-06,
      "loss": 0.43913826942443845,
      "num_input_tokens_seen": 36142384,
      "step": 1135,
      "train_runtime": 2975.718,
      "train_tokens_per_second": 12145.769
    },
    {
      "epoch": 0.07675733907891193,
      "grad_norm": 0.898926228112129,
      "learning_rate": 9.963764682461599e-06,
      "loss": 0.4480710029602051,
      "num_input_tokens_seen": 36304896,
      "step": 1140,
      "train_runtime": 2988.4618,
      "train_tokens_per_second": 12148.356
    },
    {
      "epoch": 0.07709399407487207,
      "grad_norm": 0.9776204961335204,
      "learning_rate": 9.96344624051229e-06,
      "loss": 0.44857540130615237,
      "num_input_tokens_seen": 36464624,
      "step": 1145,
      "train_runtime": 3001.9843,
      "train_tokens_per_second": 12146.84
    },
    {
      "epoch": 0.0774306490708322,
      "grad_norm": 1.0214306211301467,
      "learning_rate": 9.96312641055107e-06,
      "loss": 0.424755859375,
      "num_input_tokens_seen": 36628000,
      "step": 1150,
      "train_runtime": 3015.0435,
      "train_tokens_per_second": 12148.415
    },
    {
      "epoch": 0.07776730406679234,
      "grad_norm": 1.1870289328200219,
      "learning_rate": 9.962805192667375e-06,
      "loss": 0.41829962730407716,
      "num_input_tokens_seen": 36786728,
      "step": 1155,
      "train_runtime": 3028.3516,
      "train_tokens_per_second": 12147.443
    },
    {
      "epoch": 0.07810395906275248,
      "grad_norm": 1.105027574907607,
      "learning_rate": 9.962482586951033e-06,
      "loss": 0.4151276111602783,
      "num_input_tokens_seen": 36948808,
      "step": 1160,
      "train_runtime": 3041.647,
      "train_tokens_per_second": 12147.632
    },
    {
      "epoch": 0.07844061405871264,
      "grad_norm": 0.8984616758301404,
      "learning_rate": 9.962158593492258e-06,
      "loss": 0.4334708213806152,
      "num_input_tokens_seen": 37108496,
      "step": 1165,
      "train_runtime": 3055.107,
      "train_tokens_per_second": 12146.382
    },
    {
      "epoch": 0.07877726905467278,
      "grad_norm": 0.8738845454844043,
      "learning_rate": 9.961833212381657e-06,
      "loss": 0.44759311676025393,
      "num_input_tokens_seen": 37270392,
      "step": 1170,
      "train_runtime": 3068.9872,
      "train_tokens_per_second": 12144.199
    },
    {
      "epoch": 0.07911392405063292,
      "grad_norm": 0.9223430513688491,
      "learning_rate": 9.961506443710221e-06,
      "loss": 0.44571504592895506,
      "num_input_tokens_seen": 37433024,
      "step": 1175,
      "train_runtime": 3082.4156,
      "train_tokens_per_second": 12144.055
    },
    {
      "epoch": 0.07945057904659306,
      "grad_norm": 1.108925270054749,
      "learning_rate": 9.961178287569326e-06,
      "loss": 0.40459270477294923,
      "num_input_tokens_seen": 37594584,
      "step": 1180,
      "train_runtime": 3094.9914,
      "train_tokens_per_second": 12146.911
    },
    {
      "epoch": 0.0797872340425532,
      "grad_norm": 0.9619238247568774,
      "learning_rate": 9.960848744050747e-06,
      "loss": 0.45792160034179685,
      "num_input_tokens_seen": 37752984,
      "step": 1185,
      "train_runtime": 3109.8913,
      "train_tokens_per_second": 12139.647
    },
    {
      "epoch": 0.08012388903851334,
      "grad_norm": 0.8310051890549981,
      "learning_rate": 9.960517813246633e-06,
      "loss": 0.4673145771026611,
      "num_input_tokens_seen": 37910904,
      "step": 1190,
      "train_runtime": 3123.4113,
      "train_tokens_per_second": 12137.66
    },
    {
      "epoch": 0.08046054403447347,
      "grad_norm": 0.9912816890677105,
      "learning_rate": 9.96018549524953e-06,
      "loss": 0.44765472412109375,
      "num_input_tokens_seen": 38069952,
      "step": 1195,
      "train_runtime": 3136.9484,
      "train_tokens_per_second": 12135.983
    },
    {
      "epoch": 0.08079719903043361,
      "grad_norm": 1.245467796411856,
      "learning_rate": 9.959851790152371e-06,
      "loss": 0.4483974456787109,
      "num_input_tokens_seen": 38230168,
      "step": 1200,
      "train_runtime": 3149.8925,
      "train_tokens_per_second": 12136.975
    },
    {
      "epoch": 0.08113385402639375,
      "grad_norm": 0.8609337799542169,
      "learning_rate": 9.959516698048475e-06,
      "loss": 0.40086779594421384,
      "num_input_tokens_seen": 38386248,
      "step": 1205,
      "train_runtime": 3163.7275,
      "train_tokens_per_second": 12133.234
    },
    {
      "epoch": 0.08147050902235389,
      "grad_norm": 0.9318743561328942,
      "learning_rate": 9.959180219031551e-06,
      "loss": 0.38164477348327636,
      "num_input_tokens_seen": 38545360,
      "step": 1210,
      "train_runtime": 3176.404,
      "train_tokens_per_second": 12134.905
    },
    {
      "epoch": 0.08180716401831403,
      "grad_norm": 0.8889836854898727,
      "learning_rate": 9.958842353195688e-06,
      "loss": 0.4082005977630615,
      "num_input_tokens_seen": 38706448,
      "step": 1215,
      "train_runtime": 3189.3509,
      "train_tokens_per_second": 12136.152
    },
    {
      "epoch": 0.08214381901427417,
      "grad_norm": 0.950575810030566,
      "learning_rate": 9.958503100635377e-06,
      "loss": 0.45647764205932617,
      "num_input_tokens_seen": 38859696,
      "step": 1220,
      "train_runtime": 3202.0646,
      "train_tokens_per_second": 12135.825
    },
    {
      "epoch": 0.08248047401023431,
      "grad_norm": 0.9161106694220389,
      "learning_rate": 9.958162461445484e-06,
      "loss": 0.46908087730407716,
      "num_input_tokens_seen": 39020472,
      "step": 1225,
      "train_runtime": 3215.0335,
      "train_tokens_per_second": 12136.879
    },
    {
      "epoch": 0.08281712900619445,
      "grad_norm": 0.9224080666472065,
      "learning_rate": 9.957820435721271e-06,
      "loss": 0.4472689628601074,
      "num_input_tokens_seen": 39178424,
      "step": 1230,
      "train_runtime": 3228.0044,
      "train_tokens_per_second": 12137.042
    },
    {
      "epoch": 0.08315378400215459,
      "grad_norm": 0.9837165995216243,
      "learning_rate": 9.95747702355838e-06,
      "loss": 0.4689658164978027,
      "num_input_tokens_seen": 39339920,
      "step": 1235,
      "train_runtime": 3241.1748,
      "train_tokens_per_second": 12137.549
    },
    {
      "epoch": 0.08349043899811473,
      "grad_norm": 0.8353197310709026,
      "learning_rate": 9.957132225052848e-06,
      "loss": 0.46484966278076173,
      "num_input_tokens_seen": 39500072,
      "step": 1240,
      "train_runtime": 3254.4365,
      "train_tokens_per_second": 12137.3
    },
    {
      "epoch": 0.08382709399407487,
      "grad_norm": 0.9366691451614378,
      "learning_rate": 9.956786040301098e-06,
      "loss": 0.43509821891784667,
      "num_input_tokens_seen": 39659832,
      "step": 1245,
      "train_runtime": 3268.1416,
      "train_tokens_per_second": 12135.286
    },
    {
      "epoch": 0.08416374899003501,
      "grad_norm": 0.9994105110699071,
      "learning_rate": 9.956438469399936e-06,
      "loss": 0.4476628303527832,
      "num_input_tokens_seen": 39817576,
      "step": 1250,
      "train_runtime": 3280.8605,
      "train_tokens_per_second": 12136.321
    },
    {
      "epoch": 0.08450040398599515,
      "grad_norm": 1.0899840849477886,
      "learning_rate": 9.956089512446562e-06,
      "loss": 0.43466830253601074,
      "num_input_tokens_seen": 39973840,
      "step": 1255,
      "train_runtime": 3294.8454,
      "train_tokens_per_second": 12132.235
    },
    {
      "epoch": 0.08483705898195529,
      "grad_norm": 0.9782638411396614,
      "learning_rate": 9.955739169538559e-06,
      "loss": 0.4437400817871094,
      "num_input_tokens_seen": 40132360,
      "step": 1260,
      "train_runtime": 3308.4365,
      "train_tokens_per_second": 12130.31
    },
    {
      "epoch": 0.08517371397791543,
      "grad_norm": 1.073873825751969,
      "learning_rate": 9.955387440773902e-06,
      "loss": 0.438496732711792,
      "num_input_tokens_seen": 40292856,
      "step": 1265,
      "train_runtime": 3322.4584,
      "train_tokens_per_second": 12127.422
    },
    {
      "epoch": 0.08551036897387557,
      "grad_norm": 1.062730821617072,
      "learning_rate": 9.955034326250947e-06,
      "loss": 0.40947647094726564,
      "num_input_tokens_seen": 40455504,
      "step": 1270,
      "train_runtime": 3335.9083,
      "train_tokens_per_second": 12127.283
    },
    {
      "epoch": 0.0858470239698357,
      "grad_norm": 0.8666865316986838,
      "learning_rate": 9.954679826068444e-06,
      "loss": 0.41518335342407225,
      "num_input_tokens_seen": 40609864,
      "step": 1275,
      "train_runtime": 3348.5211,
      "train_tokens_per_second": 12127.701
    },
    {
      "epoch": 0.08618367896579585,
      "grad_norm": 0.8954599093534942,
      "learning_rate": 9.954323940325526e-06,
      "loss": 0.42731266021728515,
      "num_input_tokens_seen": 40767416,
      "step": 1280,
      "train_runtime": 3361.3846,
      "train_tokens_per_second": 12128.162
    },
    {
      "epoch": 0.086520333961756,
      "grad_norm": 0.8951000012290237,
      "learning_rate": 9.953966669121715e-06,
      "loss": 0.44504098892211913,
      "num_input_tokens_seen": 40931256,
      "step": 1285,
      "train_runtime": 3373.8252,
      "train_tokens_per_second": 12132.003
    },
    {
      "epoch": 0.08685698895771614,
      "grad_norm": 0.958524125965118,
      "learning_rate": 9.953608012556924e-06,
      "loss": 0.4569901466369629,
      "num_input_tokens_seen": 41093872,
      "step": 1290,
      "train_runtime": 3387.2465,
      "train_tokens_per_second": 12131.94
    },
    {
      "epoch": 0.08719364395367628,
      "grad_norm": 0.8299090977720209,
      "learning_rate": 9.953247970731449e-06,
      "loss": 0.450609016418457,
      "num_input_tokens_seen": 41253288,
      "step": 1295,
      "train_runtime": 3400.6827,
      "train_tokens_per_second": 12130.884
    },
    {
      "epoch": 0.08753029894963642,
      "grad_norm": 0.9978150904199796,
      "learning_rate": 9.952886543745973e-06,
      "loss": 0.45717573165893555,
      "num_input_tokens_seen": 41413936,
      "step": 1300,
      "train_runtime": 3414.7123,
      "train_tokens_per_second": 12128.089
    },
    {
      "epoch": 0.08786695394559656,
      "grad_norm": 1.0544884402830437,
      "learning_rate": 9.952523731701566e-06,
      "loss": 0.4507907867431641,
      "num_input_tokens_seen": 41577576,
      "step": 1305,
      "train_runtime": 3427.1987,
      "train_tokens_per_second": 12131.65
    },
    {
      "epoch": 0.0882036089415567,
      "grad_norm": 0.8823892227065908,
      "learning_rate": 9.952159534699693e-06,
      "loss": 0.4093925476074219,
      "num_input_tokens_seen": 41732360,
      "step": 1310,
      "train_runtime": 3440.2939,
      "train_tokens_per_second": 12130.464
    },
    {
      "epoch": 0.08854026393751684,
      "grad_norm": 1.0182163971375215,
      "learning_rate": 9.951793952842197e-06,
      "loss": 0.4836570739746094,
      "num_input_tokens_seen": 41887800,
      "step": 1315,
      "train_runtime": 3453.9625,
      "train_tokens_per_second": 12127.462
    },
    {
      "epoch": 0.08887691893347698,
      "grad_norm": 0.9258509674609388,
      "learning_rate": 9.95142698623131e-06,
      "loss": 0.4167964458465576,
      "num_input_tokens_seen": 42043408,
      "step": 1320,
      "train_runtime": 3468.2289,
      "train_tokens_per_second": 12122.443
    },
    {
      "epoch": 0.08921357392943711,
      "grad_norm": 1.0047184965677176,
      "learning_rate": 9.951058634969657e-06,
      "loss": 0.44746012687683107,
      "num_input_tokens_seen": 42200128,
      "step": 1325,
      "train_runtime": 3481.5576,
      "train_tokens_per_second": 12121.048
    },
    {
      "epoch": 0.08955022892539725,
      "grad_norm": 0.953025597735266,
      "learning_rate": 9.950688899160244e-06,
      "loss": 0.4588298797607422,
      "num_input_tokens_seen": 42362480,
      "step": 1330,
      "train_runtime": 3494.227,
      "train_tokens_per_second": 12123.563
    },
    {
      "epoch": 0.0898868839213574,
      "grad_norm": 0.9174157514853671,
      "learning_rate": 9.950317778906469e-06,
      "loss": 0.4360856056213379,
      "num_input_tokens_seen": 42525872,
      "step": 1335,
      "train_runtime": 3507.2399,
      "train_tokens_per_second": 12125.168
    },
    {
      "epoch": 0.09022353891731753,
      "grad_norm": 0.8669923017517235,
      "learning_rate": 9.949945274312109e-06,
      "loss": 0.4161503314971924,
      "num_input_tokens_seen": 42689472,
      "step": 1340,
      "train_runtime": 3520.2982,
      "train_tokens_per_second": 12126.664
    },
    {
      "epoch": 0.09056019391327767,
      "grad_norm": 1.0627728227791875,
      "learning_rate": 9.94957138548134e-06,
      "loss": 0.4660347938537598,
      "num_input_tokens_seen": 42840376,
      "step": 1345,
      "train_runtime": 3533.6444,
      "train_tokens_per_second": 12123.567
    },
    {
      "epoch": 0.09089684890923781,
      "grad_norm": 1.0413034989700771,
      "learning_rate": 9.949196112518715e-06,
      "loss": 0.42073640823364256,
      "num_input_tokens_seen": 43000168,
      "step": 1350,
      "train_runtime": 3546.1296,
      "train_tokens_per_second": 12125.944
    },
    {
      "epoch": 0.09123350390519795,
      "grad_norm": 0.9926759290790045,
      "learning_rate": 9.94881945552918e-06,
      "loss": 0.45658254623413086,
      "num_input_tokens_seen": 43160680,
      "step": 1355,
      "train_runtime": 3558.6026,
      "train_tokens_per_second": 12128.547
    },
    {
      "epoch": 0.09157015890115809,
      "grad_norm": 1.1331792792018747,
      "learning_rate": 9.948441414618064e-06,
      "loss": 0.4689138889312744,
      "num_input_tokens_seen": 43319456,
      "step": 1360,
      "train_runtime": 3571.6871,
      "train_tokens_per_second": 12128.57
    },
    {
      "epoch": 0.09190681389711823,
      "grad_norm": 0.9748317820985101,
      "learning_rate": 9.948061989891085e-06,
      "loss": 0.4370171070098877,
      "num_input_tokens_seen": 43481296,
      "step": 1365,
      "train_runtime": 3585.0845,
      "train_tokens_per_second": 12128.388
    },
    {
      "epoch": 0.09224346889307837,
      "grad_norm": 0.9857529772608139,
      "learning_rate": 9.94768118145435e-06,
      "loss": 0.4425950527191162,
      "num_input_tokens_seen": 43639760,
      "step": 1370,
      "train_runtime": 3599.0995,
      "train_tokens_per_second": 12125.189
    },
    {
      "epoch": 0.09258012388903851,
      "grad_norm": 0.8818333263856521,
      "learning_rate": 9.947298989414349e-06,
      "loss": 0.4210357666015625,
      "num_input_tokens_seen": 43801224,
      "step": 1375,
      "train_runtime": 3611.6488,
      "train_tokens_per_second": 12127.764
    },
    {
      "epoch": 0.09291677888499865,
      "grad_norm": 0.9852834539317858,
      "learning_rate": 9.946915413877963e-06,
      "loss": 0.4292550563812256,
      "num_input_tokens_seen": 43965032,
      "step": 1380,
      "train_runtime": 3624.7725,
      "train_tokens_per_second": 12129.046
    },
    {
      "epoch": 0.09325343388095879,
      "grad_norm": 0.9306783826618251,
      "learning_rate": 9.946530454952454e-06,
      "loss": 0.45383524894714355,
      "num_input_tokens_seen": 44128872,
      "step": 1385,
      "train_runtime": 3637.2176,
      "train_tokens_per_second": 12132.591
    },
    {
      "epoch": 0.09359008887691893,
      "grad_norm": 0.9581511629288435,
      "learning_rate": 9.946144112745477e-06,
      "loss": 0.4641427040100098,
      "num_input_tokens_seen": 44290256,
      "step": 1390,
      "train_runtime": 3649.7337,
      "train_tokens_per_second": 12135.202
    },
    {
      "epoch": 0.09392674387287907,
      "grad_norm": 1.1591141466046666,
      "learning_rate": 9.94575638736507e-06,
      "loss": 0.4267313003540039,
      "num_input_tokens_seen": 44451016,
      "step": 1395,
      "train_runtime": 3662.6274,
      "train_tokens_per_second": 12136.374
    },
    {
      "epoch": 0.09426339886883921,
      "grad_norm": 1.0467058730294923,
      "learning_rate": 9.945367278919662e-06,
      "loss": 0.46062068939208983,
      "num_input_tokens_seen": 44611696,
      "step": 1400,
      "train_runtime": 3676.3758,
      "train_tokens_per_second": 12134.694
    },
    {
      "epoch": 0.09460005386479936,
      "grad_norm": 1.0240264460043018,
      "learning_rate": 9.944976787518064e-06,
      "loss": 0.450745964050293,
      "num_input_tokens_seen": 44770592,
      "step": 1405,
      "train_runtime": 3689.9272,
      "train_tokens_per_second": 12133.191
    },
    {
      "epoch": 0.0949367088607595,
      "grad_norm": 0.9071400522366527,
      "learning_rate": 9.944584913269474e-06,
      "loss": 0.43937244415283205,
      "num_input_tokens_seen": 44934288,
      "step": 1410,
      "train_runtime": 3703.0161,
      "train_tokens_per_second": 12134.511
    },
    {
      "epoch": 0.09527336385671964,
      "grad_norm": 0.83937585899858,
      "learning_rate": 9.944191656283481e-06,
      "loss": 0.4137714862823486,
      "num_input_tokens_seen": 45095936,
      "step": 1415,
      "train_runtime": 3716.2965,
      "train_tokens_per_second": 12134.644
    },
    {
      "epoch": 0.09561001885267978,
      "grad_norm": 1.0868900519848876,
      "learning_rate": 9.943797016670059e-06,
      "loss": 0.43173813819885254,
      "num_input_tokens_seen": 45253448,
      "step": 1420,
      "train_runtime": 3730.3934,
      "train_tokens_per_second": 12131.012
    },
    {
      "epoch": 0.09594667384863992,
      "grad_norm": 0.9753176495018745,
      "learning_rate": 9.943400994539565e-06,
      "loss": 0.43209304809570315,
      "num_input_tokens_seen": 45413944,
      "step": 1425,
      "train_runtime": 3743.8384,
      "train_tokens_per_second": 12130.316
    },
    {
      "epoch": 0.09628332884460006,
      "grad_norm": 0.9392448880944373,
      "learning_rate": 9.943003590002744e-06,
      "loss": 0.44062557220458987,
      "num_input_tokens_seen": 45576136,
      "step": 1430,
      "train_runtime": 3756.4936,
      "train_tokens_per_second": 12132.627
    },
    {
      "epoch": 0.0966199838405602,
      "grad_norm": 0.9854116851896264,
      "learning_rate": 9.942604803170733e-06,
      "loss": 0.44598841667175293,
      "num_input_tokens_seen": 45732544,
      "step": 1435,
      "train_runtime": 3770.057,
      "train_tokens_per_second": 12130.465
    },
    {
      "epoch": 0.09695663883652034,
      "grad_norm": 0.9165607209930458,
      "learning_rate": 9.94220463415505e-06,
      "loss": 0.44730443954467775,
      "num_input_tokens_seen": 45887432,
      "step": 1440,
      "train_runtime": 3782.7157,
      "train_tokens_per_second": 12130.817
    },
    {
      "epoch": 0.09729329383248048,
      "grad_norm": 0.8480367621187166,
      "learning_rate": 9.9418030830676e-06,
      "loss": 0.45311617851257324,
      "num_input_tokens_seen": 46051272,
      "step": 1445,
      "train_runtime": 3795.1625,
      "train_tokens_per_second": 12134.203
    },
    {
      "epoch": 0.09762994882844062,
      "grad_norm": 1.0583837703836312,
      "learning_rate": 9.941400150020676e-06,
      "loss": 0.4435733795166016,
      "num_input_tokens_seen": 46205616,
      "step": 1450,
      "train_runtime": 3808.5648,
      "train_tokens_per_second": 12132.028
    },
    {
      "epoch": 0.09796660382440076,
      "grad_norm": 0.8871343722220218,
      "learning_rate": 9.940995835126957e-06,
      "loss": 0.4543262481689453,
      "num_input_tokens_seen": 46366208,
      "step": 1455,
      "train_runtime": 3821.9036,
      "train_tokens_per_second": 12131.705
    },
    {
      "epoch": 0.0983032588203609,
      "grad_norm": 0.9118323090649372,
      "learning_rate": 9.940590138499508e-06,
      "loss": 0.4204091548919678,
      "num_input_tokens_seen": 46528168,
      "step": 1460,
      "train_runtime": 3835.0607,
      "train_tokens_per_second": 12132.316
    },
    {
      "epoch": 0.09863991381632103,
      "grad_norm": 0.918325004832027,
      "learning_rate": 9.940183060251783e-06,
      "loss": 0.428496265411377,
      "num_input_tokens_seen": 46678840,
      "step": 1465,
      "train_runtime": 3848.1423,
      "train_tokens_per_second": 12130.227
    },
    {
      "epoch": 0.09897656881228117,
      "grad_norm": 0.8356241263106327,
      "learning_rate": 9.939774600497616e-06,
      "loss": 0.40030221939086913,
      "num_input_tokens_seen": 46840008,
      "step": 1470,
      "train_runtime": 3861.5684,
      "train_tokens_per_second": 12129.788
    },
    {
      "epoch": 0.09931322380824131,
      "grad_norm": 0.8633439072735342,
      "learning_rate": 9.939364759351236e-06,
      "loss": 0.44556803703308107,
      "num_input_tokens_seen": 46999400,
      "step": 1475,
      "train_runtime": 3875.3513,
      "train_tokens_per_second": 12127.778
    },
    {
      "epoch": 0.09964987880420145,
      "grad_norm": 0.9241174116140954,
      "learning_rate": 9.938953536927249e-06,
      "loss": 0.4041449546813965,
      "num_input_tokens_seen": 47163112,
      "step": 1480,
      "train_runtime": 3888.4121,
      "train_tokens_per_second": 12129.144
    },
    {
      "epoch": 0.09998653380016159,
      "grad_norm": 0.9158896320984881,
      "learning_rate": 9.938540933340657e-06,
      "loss": 0.45334687232971194,
      "num_input_tokens_seen": 47322520,
      "step": 1485,
      "train_runtime": 3901.8571,
      "train_tokens_per_second": 12128.204
    },
    {
      "epoch": 0.10032318879612173,
      "grad_norm": 0.8162617020873536,
      "learning_rate": 9.938126948706839e-06,
      "loss": 0.43866500854492185,
      "num_input_tokens_seen": 47484520,
      "step": 1490,
      "train_runtime": 3915.1052,
      "train_tokens_per_second": 12128.542
    },
    {
      "epoch": 0.10065984379208187,
      "grad_norm": 0.7596109827826565,
      "learning_rate": 9.937711583141567e-06,
      "loss": 0.42181997299194335,
      "num_input_tokens_seen": 47644776,
      "step": 1495,
      "train_runtime": 3928.329,
      "train_tokens_per_second": 12128.51
    },
    {
      "epoch": 0.10099649878804201,
      "grad_norm": 0.9017738622746162,
      "learning_rate": 9.937294836760995e-06,
      "loss": 0.4284855365753174,
      "num_input_tokens_seen": 47804768,
      "step": 1500,
      "train_runtime": 3942.2453,
      "train_tokens_per_second": 12126.279
    },
    {
      "epoch": 0.10133315378400215,
      "grad_norm": 1.4606469661161126,
      "learning_rate": 9.936876709681668e-06,
      "loss": 0.4764401435852051,
      "num_input_tokens_seen": 47967000,
      "step": 1505,
      "train_runtime": 3955.5768,
      "train_tokens_per_second": 12126.424
    },
    {
      "epoch": 0.10166980877996229,
      "grad_norm": 0.839610105506026,
      "learning_rate": 9.93645720202051e-06,
      "loss": 0.38546133041381836,
      "num_input_tokens_seen": 48126392,
      "step": 1510,
      "train_runtime": 3968.7386,
      "train_tokens_per_second": 12126.37
    },
    {
      "epoch": 0.10200646377592243,
      "grad_norm": 0.9106614882016977,
      "learning_rate": 9.936036313894836e-06,
      "loss": 0.46453375816345216,
      "num_input_tokens_seen": 48283456,
      "step": 1515,
      "train_runtime": 3982.4835,
      "train_tokens_per_second": 12123.956
    },
    {
      "epoch": 0.10234311877188257,
      "grad_norm": 1.0247329171439983,
      "learning_rate": 9.935614045422349e-06,
      "loss": 0.4605582237243652,
      "num_input_tokens_seen": 48443448,
      "step": 1520,
      "train_runtime": 3995.2111,
      "train_tokens_per_second": 12125.379
    },
    {
      "epoch": 0.10267977376784271,
      "grad_norm": 0.8980150113086468,
      "learning_rate": 9.935190396721132e-06,
      "loss": 0.4190380096435547,
      "num_input_tokens_seen": 48603992,
      "step": 1525,
      "train_runtime": 4008.1549,
      "train_tokens_per_second": 12126.276
    },
    {
      "epoch": 0.10301642876380286,
      "grad_norm": 0.9254513639303532,
      "learning_rate": 9.934765367909658e-06,
      "loss": 0.44364585876464846,
      "num_input_tokens_seen": 48763344,
      "step": 1530,
      "train_runtime": 4020.8471,
      "train_tokens_per_second": 12127.629
    },
    {
      "epoch": 0.103353083759763,
      "grad_norm": 0.9852787080419828,
      "learning_rate": 9.934338959106783e-06,
      "loss": 0.4201723575592041,
      "num_input_tokens_seen": 48921168,
      "step": 1535,
      "train_runtime": 4033.5296,
      "train_tokens_per_second": 12128.625
    },
    {
      "epoch": 0.10368973875572314,
      "grad_norm": 1.1623498598175181,
      "learning_rate": 9.933911170431753e-06,
      "loss": 0.46906318664550783,
      "num_input_tokens_seen": 49075448,
      "step": 1540,
      "train_runtime": 4047.5353,
      "train_tokens_per_second": 12124.773
    },
    {
      "epoch": 0.10402639375168328,
      "grad_norm": 1.0049841697626876,
      "learning_rate": 9.933482002004198e-06,
      "loss": 0.4533809185028076,
      "num_input_tokens_seen": 49239008,
      "step": 1545,
      "train_runtime": 4060.5543,
      "train_tokens_per_second": 12126.179
    },
    {
      "epoch": 0.10436304874764342,
      "grad_norm": 1.0829885273211237,
      "learning_rate": 9.933051453944134e-06,
      "loss": 0.4875025749206543,
      "num_input_tokens_seen": 49398224,
      "step": 1550,
      "train_runtime": 4073.2749,
      "train_tokens_per_second": 12127.398
    },
    {
      "epoch": 0.10469970374360356,
      "grad_norm": 0.8862916745420065,
      "learning_rate": 9.932619526371958e-06,
      "loss": 0.39573619365692136,
      "num_input_tokens_seen": 49559168,
      "step": 1555,
      "train_runtime": 4086.1177,
      "train_tokens_per_second": 12128.669
    },
    {
      "epoch": 0.1050363587395637,
      "grad_norm": 0.9136804465008656,
      "learning_rate": 9.932186219408463e-06,
      "loss": 0.4129672050476074,
      "num_input_tokens_seen": 49711104,
      "step": 1560,
      "train_runtime": 4098.7775,
      "train_tokens_per_second": 12128.276
    },
    {
      "epoch": 0.10537301373552384,
      "grad_norm": 1.080397205079806,
      "learning_rate": 9.931751533174819e-06,
      "loss": 0.46036815643310547,
      "num_input_tokens_seen": 49867376,
      "step": 1565,
      "train_runtime": 4112.219,
      "train_tokens_per_second": 12126.634
    },
    {
      "epoch": 0.10570966873148398,
      "grad_norm": 0.9233372326641195,
      "learning_rate": 9.931315467792584e-06,
      "loss": 0.44809575080871583,
      "num_input_tokens_seen": 50031216,
      "step": 1570,
      "train_runtime": 4124.6563,
      "train_tokens_per_second": 12129.79
    },
    {
      "epoch": 0.10604632372744412,
      "grad_norm": 0.9000342578095065,
      "learning_rate": 9.930878023383705e-06,
      "loss": 0.4406759738922119,
      "num_input_tokens_seen": 50190816,
      "step": 1575,
      "train_runtime": 4137.8195,
      "train_tokens_per_second": 12129.774
    },
    {
      "epoch": 0.10638297872340426,
      "grad_norm": 0.8317163547044458,
      "learning_rate": 9.93043920007051e-06,
      "loss": 0.42117891311645506,
      "num_input_tokens_seen": 50352472,
      "step": 1580,
      "train_runtime": 4150.6882,
      "train_tokens_per_second": 12131.114
    },
    {
      "epoch": 0.1067196337193644,
      "grad_norm": 1.1160349310192461,
      "learning_rate": 9.929998997975712e-06,
      "loss": 0.42940635681152345,
      "num_input_tokens_seen": 50515560,
      "step": 1585,
      "train_runtime": 4163.5451,
      "train_tokens_per_second": 12132.824
    },
    {
      "epoch": 0.10705628871532454,
      "grad_norm": 0.9220999821639656,
      "learning_rate": 9.929557417222417e-06,
      "loss": 0.416658353805542,
      "num_input_tokens_seen": 50670216,
      "step": 1590,
      "train_runtime": 4176.866,
      "train_tokens_per_second": 12131.157
    },
    {
      "epoch": 0.10739294371128467,
      "grad_norm": 0.9373099688109692,
      "learning_rate": 9.929114457934109e-06,
      "loss": 0.43518705368041993,
      "num_input_tokens_seen": 50822800,
      "step": 1595,
      "train_runtime": 4189.4316,
      "train_tokens_per_second": 12131.192
    },
    {
      "epoch": 0.10772959870724481,
      "grad_norm": 0.9412657002038296,
      "learning_rate": 9.92867012023466e-06,
      "loss": 0.4593684196472168,
      "num_input_tokens_seen": 50982720,
      "step": 1600,
      "train_runtime": 4202.2286,
      "train_tokens_per_second": 12132.305
    },
    {
      "epoch": 0.10806625370320495,
      "grad_norm": 0.9024611883882057,
      "learning_rate": 9.92822440424833e-06,
      "loss": 0.4697587966918945,
      "num_input_tokens_seen": 51143544,
      "step": 1605,
      "train_runtime": 4214.7171,
      "train_tokens_per_second": 12134.514
    },
    {
      "epoch": 0.1084029086991651,
      "grad_norm": 0.9206575384576616,
      "learning_rate": 9.927777310099759e-06,
      "loss": 0.4916232585906982,
      "num_input_tokens_seen": 51306432,
      "step": 1610,
      "train_runtime": 4227.7571,
      "train_tokens_per_second": 12135.615
    },
    {
      "epoch": 0.10873956369512523,
      "grad_norm": 0.9475097961115961,
      "learning_rate": 9.927328837913976e-06,
      "loss": 0.4308351516723633,
      "num_input_tokens_seen": 51466696,
      "step": 1615,
      "train_runtime": 4240.3507,
      "train_tokens_per_second": 12137.368
    },
    {
      "epoch": 0.10907621869108537,
      "grad_norm": 0.9096331088191733,
      "learning_rate": 9.926878987816397e-06,
      "loss": 0.42225494384765627,
      "num_input_tokens_seen": 51630536,
      "step": 1620,
      "train_runtime": 4252.8043,
      "train_tokens_per_second": 12140.351
    },
    {
      "epoch": 0.10941287368704551,
      "grad_norm": 0.8204446692865083,
      "learning_rate": 9.92642775993282e-06,
      "loss": 0.4369563102722168,
      "num_input_tokens_seen": 51792200,
      "step": 1625,
      "train_runtime": 4265.892,
      "train_tokens_per_second": 12141.001
    },
    {
      "epoch": 0.10974952868300565,
      "grad_norm": 0.8990218935810885,
      "learning_rate": 9.925975154389428e-06,
      "loss": 0.39328005313873293,
      "num_input_tokens_seen": 51948552,
      "step": 1630,
      "train_runtime": 4278.7623,
      "train_tokens_per_second": 12141.023
    },
    {
      "epoch": 0.11008618367896579,
      "grad_norm": 0.9257407765811783,
      "learning_rate": 9.925521171312792e-06,
      "loss": 0.42905502319335936,
      "num_input_tokens_seen": 52105488,
      "step": 1635,
      "train_runtime": 4291.3414,
      "train_tokens_per_second": 12142.005
    },
    {
      "epoch": 0.11042283867492593,
      "grad_norm": 0.9306979963405362,
      "learning_rate": 9.925065810829868e-06,
      "loss": 0.45661077499389646,
      "num_input_tokens_seen": 52260032,
      "step": 1640,
      "train_runtime": 4304.5997,
      "train_tokens_per_second": 12140.509
    },
    {
      "epoch": 0.11075949367088607,
      "grad_norm": 0.9518070246743361,
      "learning_rate": 9.924609073067994e-06,
      "loss": 0.46037540435791013,
      "num_input_tokens_seen": 52420192,
      "step": 1645,
      "train_runtime": 4317.5488,
      "train_tokens_per_second": 12141.193
    },
    {
      "epoch": 0.11109614866684622,
      "grad_norm": 0.9996141438042896,
      "learning_rate": 9.924150958154897e-06,
      "loss": 0.4645636558532715,
      "num_input_tokens_seen": 52583640,
      "step": 1650,
      "train_runtime": 4330.6265,
      "train_tokens_per_second": 12142.271
    },
    {
      "epoch": 0.11143280366280636,
      "grad_norm": 0.9288808966586459,
      "learning_rate": 9.923691466218686e-06,
      "loss": 0.41077795028686526,
      "num_input_tokens_seen": 52744216,
      "step": 1655,
      "train_runtime": 4343.3621,
      "train_tokens_per_second": 12143.638
    },
    {
      "epoch": 0.1117694586587665,
      "grad_norm": 0.9560837089912206,
      "learning_rate": 9.923230597387856e-06,
      "loss": 0.4429797649383545,
      "num_input_tokens_seen": 52908056,
      "step": 1660,
      "train_runtime": 4355.8028,
      "train_tokens_per_second": 12146.568
    },
    {
      "epoch": 0.11210611365472664,
      "grad_norm": 1.0012339857977564,
      "learning_rate": 9.922768351791289e-06,
      "loss": 0.46070384979248047,
      "num_input_tokens_seen": 53064256,
      "step": 1665,
      "train_runtime": 4368.5474,
      "train_tokens_per_second": 12146.888
    },
    {
      "epoch": 0.11244276865068678,
      "grad_norm": 0.907849566596143,
      "learning_rate": 9.92230472955825e-06,
      "loss": 0.4141829490661621,
      "num_input_tokens_seen": 53227248,
      "step": 1670,
      "train_runtime": 4382.1861,
      "train_tokens_per_second": 12146.277
    },
    {
      "epoch": 0.11277942364664692,
      "grad_norm": 0.8460392997155691,
      "learning_rate": 9.92183973081839e-06,
      "loss": 0.40268354415893554,
      "num_input_tokens_seen": 53390448,
      "step": 1675,
      "train_runtime": 4395.3076,
      "train_tokens_per_second": 12147.147
    },
    {
      "epoch": 0.11311607864260706,
      "grad_norm": 0.8495464006739524,
      "learning_rate": 9.921373355701743e-06,
      "loss": 0.4864508152008057,
      "num_input_tokens_seen": 53551912,
      "step": 1680,
      "train_runtime": 4408.1744,
      "train_tokens_per_second": 12148.32
    },
    {
      "epoch": 0.1134527336385672,
      "grad_norm": 0.86896584972378,
      "learning_rate": 9.920905604338732e-06,
      "loss": 0.46452083587646487,
      "num_input_tokens_seen": 53715528,
      "step": 1685,
      "train_runtime": 4420.9546,
      "train_tokens_per_second": 12150.21
    },
    {
      "epoch": 0.11378938863452734,
      "grad_norm": 0.9097096798343015,
      "learning_rate": 9.920436476860158e-06,
      "loss": 0.4091813087463379,
      "num_input_tokens_seen": 53873288,
      "step": 1690,
      "train_runtime": 4434.5402,
      "train_tokens_per_second": 12148.562
    },
    {
      "epoch": 0.11412604363048748,
      "grad_norm": 0.8584497153256407,
      "learning_rate": 9.919965973397214e-06,
      "loss": 0.4077142715454102,
      "num_input_tokens_seen": 54031912,
      "step": 1695,
      "train_runtime": 4448.0087,
      "train_tokens_per_second": 12147.438
    },
    {
      "epoch": 0.11446269862644762,
      "grad_norm": 0.9078107756345885,
      "learning_rate": 9.919494094081475e-06,
      "loss": 0.4653289794921875,
      "num_input_tokens_seen": 54189080,
      "step": 1700,
      "train_runtime": 4461.9538,
      "train_tokens_per_second": 12144.698
    },
    {
      "epoch": 0.11479935362240776,
      "grad_norm": 0.9511292995129742,
      "learning_rate": 9.919020839044899e-06,
      "loss": 0.4436480522155762,
      "num_input_tokens_seen": 54352920,
      "step": 1705,
      "train_runtime": 4474.3818,
      "train_tokens_per_second": 12147.582
    },
    {
      "epoch": 0.1151360086183679,
      "grad_norm": 0.7937590042259438,
      "learning_rate": 9.918546208419832e-06,
      "loss": 0.4208201885223389,
      "num_input_tokens_seen": 54512176,
      "step": 1710,
      "train_runtime": 4487.0913,
      "train_tokens_per_second": 12148.667
    },
    {
      "epoch": 0.11547266361432804,
      "grad_norm": 0.8928356534020111,
      "learning_rate": 9.918070202339e-06,
      "loss": 0.4348869323730469,
      "num_input_tokens_seen": 54674880,
      "step": 1715,
      "train_runtime": 4500.0807,
      "train_tokens_per_second": 12149.755
    },
    {
      "epoch": 0.11580931861028818,
      "grad_norm": 0.9292342724588977,
      "learning_rate": 9.91759282093552e-06,
      "loss": 0.4313792705535889,
      "num_input_tokens_seen": 54832480,
      "step": 1720,
      "train_runtime": 4513.3919,
      "train_tokens_per_second": 12148.841
    },
    {
      "epoch": 0.11614597360624832,
      "grad_norm": 0.8640350706481296,
      "learning_rate": 9.917114064342888e-06,
      "loss": 0.40947756767272947,
      "num_input_tokens_seen": 54989208,
      "step": 1725,
      "train_runtime": 4526.7887,
      "train_tokens_per_second": 12147.509
    },
    {
      "epoch": 0.11648262860220845,
      "grad_norm": 1.1359038702136577,
      "learning_rate": 9.916633932694988e-06,
      "loss": 0.4014447212219238,
      "num_input_tokens_seen": 55150064,
      "step": 1730,
      "train_runtime": 4539.8194,
      "train_tokens_per_second": 12148.074
    },
    {
      "epoch": 0.1168192835981686,
      "grad_norm": 0.9063518913514674,
      "learning_rate": 9.916152426126086e-06,
      "loss": 0.38015270233154297,
      "num_input_tokens_seen": 55310872,
      "step": 1735,
      "train_runtime": 4553.8494,
      "train_tokens_per_second": 12145.96
    },
    {
      "epoch": 0.11715593859412873,
      "grad_norm": 0.9687998748062824,
      "learning_rate": 9.915669544770837e-06,
      "loss": 0.4296428680419922,
      "num_input_tokens_seen": 55474240,
      "step": 1740,
      "train_runtime": 4566.3211,
      "train_tokens_per_second": 12148.563
    },
    {
      "epoch": 0.11749259359008887,
      "grad_norm": 0.7877490489384283,
      "learning_rate": 9.915185288764272e-06,
      "loss": 0.42476983070373536,
      "num_input_tokens_seen": 55636768,
      "step": 1745,
      "train_runtime": 4579.0625,
      "train_tokens_per_second": 12150.253
    },
    {
      "epoch": 0.11782924858604901,
      "grad_norm": 0.847654496280336,
      "learning_rate": 9.914699658241815e-06,
      "loss": 0.42523956298828125,
      "num_input_tokens_seen": 55799656,
      "step": 1750,
      "train_runtime": 4592.5643,
      "train_tokens_per_second": 12150.0
    },
    {
      "epoch": 0.11816590358200915,
      "grad_norm": 0.9957985712545597,
      "learning_rate": 9.91421265333927e-06,
      "loss": 0.44651594161987307,
      "num_input_tokens_seen": 55962608,
      "step": 1755,
      "train_runtime": 4605.3874,
      "train_tokens_per_second": 12151.553
    },
    {
      "epoch": 0.11850255857796929,
      "grad_norm": 0.8694558678305105,
      "learning_rate": 9.913724274192829e-06,
      "loss": 0.4229731559753418,
      "num_input_tokens_seen": 56125320,
      "step": 1760,
      "train_runtime": 4618.1847,
      "train_tokens_per_second": 12153.113
    },
    {
      "epoch": 0.11883921357392943,
      "grad_norm": 0.9377605837423263,
      "learning_rate": 9.91323452093906e-06,
      "loss": 0.41498355865478515,
      "num_input_tokens_seen": 56283976,
      "step": 1765,
      "train_runtime": 4631.6,
      "train_tokens_per_second": 12152.167
    },
    {
      "epoch": 0.11917586856988957,
      "grad_norm": 0.94791109184082,
      "learning_rate": 9.912743393714927e-06,
      "loss": 0.3888449192047119,
      "num_input_tokens_seen": 56443152,
      "step": 1770,
      "train_runtime": 4644.2836,
      "train_tokens_per_second": 12153.253
    },
    {
      "epoch": 0.11951252356584972,
      "grad_norm": 0.8685863959576018,
      "learning_rate": 9.912250892657769e-06,
      "loss": 0.41208562850952146,
      "num_input_tokens_seen": 56602848,
      "step": 1775,
      "train_runtime": 4657.3503,
      "train_tokens_per_second": 12153.444
    },
    {
      "epoch": 0.11984917856180986,
      "grad_norm": 0.9020342597516245,
      "learning_rate": 9.911757017905312e-06,
      "loss": 0.42429428100585936,
      "num_input_tokens_seen": 56756320,
      "step": 1780,
      "train_runtime": 4670.8096,
      "train_tokens_per_second": 12151.281
    },
    {
      "epoch": 0.12018583355777,
      "grad_norm": 0.9266008330206336,
      "learning_rate": 9.911261769595668e-06,
      "loss": 0.4274741649627686,
      "num_input_tokens_seen": 56919768,
      "step": 1785,
      "train_runtime": 4683.8192,
      "train_tokens_per_second": 12152.426
    },
    {
      "epoch": 0.12052248855373014,
      "grad_norm": 1.0093198843999878,
      "learning_rate": 9.910765147867332e-06,
      "loss": 0.4207764148712158,
      "num_input_tokens_seen": 57079784,
      "step": 1790,
      "train_runtime": 4697.0317,
      "train_tokens_per_second": 12152.31
    },
    {
      "epoch": 0.12085914354969028,
      "grad_norm": 0.9416409677441909,
      "learning_rate": 9.91026715285918e-06,
      "loss": 0.425150203704834,
      "num_input_tokens_seen": 57239952,
      "step": 1795,
      "train_runtime": 4710.1098,
      "train_tokens_per_second": 12152.573
    },
    {
      "epoch": 0.12119579854565042,
      "grad_norm": 1.02313325709948,
      "learning_rate": 9.909767784710476e-06,
      "loss": 0.4768087387084961,
      "num_input_tokens_seen": 57403792,
      "step": 1800,
      "train_runtime": 4722.5445,
      "train_tokens_per_second": 12155.268
    },
    {
      "epoch": 0.12153245354161056,
      "grad_norm": 0.8894346181418543,
      "learning_rate": 9.909267043560868e-06,
      "loss": 0.46009435653686526,
      "num_input_tokens_seen": 57560992,
      "step": 1805,
      "train_runtime": 4735.1093,
      "train_tokens_per_second": 12156.212
    },
    {
      "epoch": 0.1218691085375707,
      "grad_norm": 0.9078126221919158,
      "learning_rate": 9.908764929550388e-06,
      "loss": 0.43431482315063474,
      "num_input_tokens_seen": 57721744,
      "step": 1810,
      "train_runtime": 4748.2558,
      "train_tokens_per_second": 12156.41
    },
    {
      "epoch": 0.12220576353353084,
      "grad_norm": 0.9240817581316834,
      "learning_rate": 9.908261442819444e-06,
      "loss": 0.4190185070037842,
      "num_input_tokens_seen": 57874976,
      "step": 1815,
      "train_runtime": 4761.5437,
      "train_tokens_per_second": 12154.667
    },
    {
      "epoch": 0.12254241852949098,
      "grad_norm": 0.9186413943945044,
      "learning_rate": 9.907756583508842e-06,
      "loss": 0.4270307540893555,
      "num_input_tokens_seen": 58038264,
      "step": 1820,
      "train_runtime": 4774.0203,
      "train_tokens_per_second": 12157.105
    },
    {
      "epoch": 0.12287907352545112,
      "grad_norm": 0.853608751016872,
      "learning_rate": 9.907250351759761e-06,
      "loss": 0.4223791122436523,
      "num_input_tokens_seen": 58199144,
      "step": 1825,
      "train_runtime": 4787.4318,
      "train_tokens_per_second": 12156.652
    },
    {
      "epoch": 0.12321572852141126,
      "grad_norm": 0.9741325446978811,
      "learning_rate": 9.906742747713766e-06,
      "loss": 0.4422037124633789,
      "num_input_tokens_seen": 58357976,
      "step": 1830,
      "train_runtime": 4801.0386,
      "train_tokens_per_second": 12155.282
    },
    {
      "epoch": 0.1235523835173714,
      "grad_norm": 0.9856021144089,
      "learning_rate": 9.906233771512808e-06,
      "loss": 0.4520071506500244,
      "num_input_tokens_seen": 58515072,
      "step": 1835,
      "train_runtime": 4814.2615,
      "train_tokens_per_second": 12154.527
    },
    {
      "epoch": 0.12388903851333154,
      "grad_norm": 1.081238427906188,
      "learning_rate": 9.905723423299222e-06,
      "loss": 0.47678585052490235,
      "num_input_tokens_seen": 58676736,
      "step": 1840,
      "train_runtime": 4826.7508,
      "train_tokens_per_second": 12156.57
    },
    {
      "epoch": 0.12422569350929168,
      "grad_norm": 0.9362505076125667,
      "learning_rate": 9.905211703215723e-06,
      "loss": 0.40844020843505857,
      "num_input_tokens_seen": 58831800,
      "step": 1845,
      "train_runtime": 4839.5987,
      "train_tokens_per_second": 12156.338
    },
    {
      "epoch": 0.12456234850525182,
      "grad_norm": 0.8264260910013779,
      "learning_rate": 9.904698611405412e-06,
      "loss": 0.4227871894836426,
      "num_input_tokens_seen": 58991928,
      "step": 1850,
      "train_runtime": 4853.1645,
      "train_tokens_per_second": 12155.353
    },
    {
      "epoch": 0.12489900350121196,
      "grad_norm": 0.7919523068279856,
      "learning_rate": 9.904184148011774e-06,
      "loss": 0.43866868019104005,
      "num_input_tokens_seen": 59151208,
      "step": 1855,
      "train_runtime": 4866.1326,
      "train_tokens_per_second": 12155.692
    },
    {
      "epoch": 0.1252356584971721,
      "grad_norm": 1.0239417593489215,
      "learning_rate": 9.903668313178681e-06,
      "loss": 0.46463823318481445,
      "num_input_tokens_seen": 59313416,
      "step": 1860,
      "train_runtime": 4878.7524,
      "train_tokens_per_second": 12157.497
    },
    {
      "epoch": 0.12557231349313225,
      "grad_norm": 0.9699631652297127,
      "learning_rate": 9.903151107050377e-06,
      "loss": 0.40071907043457033,
      "num_input_tokens_seen": 59467808,
      "step": 1865,
      "train_runtime": 4892.2397,
      "train_tokens_per_second": 12155.538
    },
    {
      "epoch": 0.1259089684890924,
      "grad_norm": 0.918565763654552,
      "learning_rate": 9.902632529771502e-06,
      "loss": 0.4531282424926758,
      "num_input_tokens_seen": 59628216,
      "step": 1870,
      "train_runtime": 4905.2225,
      "train_tokens_per_second": 12156.068
    },
    {
      "epoch": 0.12624562348505253,
      "grad_norm": 1.08492557807277,
      "learning_rate": 9.902112581487074e-06,
      "loss": 0.43126940727233887,
      "num_input_tokens_seen": 59789424,
      "step": 1875,
      "train_runtime": 4919.234,
      "train_tokens_per_second": 12154.214
    },
    {
      "epoch": 0.12658227848101267,
      "grad_norm": 0.9330675985787636,
      "learning_rate": 9.901591262342494e-06,
      "loss": 0.45319561958312987,
      "num_input_tokens_seen": 59947864,
      "step": 1880,
      "train_runtime": 4932.3163,
      "train_tokens_per_second": 12154.1
    },
    {
      "epoch": 0.1269189334769728,
      "grad_norm": 1.0144851121417797,
      "learning_rate": 9.901068572483548e-06,
      "loss": 0.41886653900146487,
      "num_input_tokens_seen": 60109312,
      "step": 1885,
      "train_runtime": 4945.7008,
      "train_tokens_per_second": 12153.851
    },
    {
      "epoch": 0.12725558847293295,
      "grad_norm": 0.9503891138886303,
      "learning_rate": 9.900544512056402e-06,
      "loss": 0.4605095863342285,
      "num_input_tokens_seen": 60271256,
      "step": 1890,
      "train_runtime": 4958.9237,
      "train_tokens_per_second": 12154.1
    },
    {
      "epoch": 0.12759224346889309,
      "grad_norm": 0.855031792864856,
      "learning_rate": 9.900019081207612e-06,
      "loss": 0.4420969009399414,
      "num_input_tokens_seen": 60434864,
      "step": 1895,
      "train_runtime": 4972.0377,
      "train_tokens_per_second": 12154.949
    },
    {
      "epoch": 0.12792889846485322,
      "grad_norm": 0.9396584523608502,
      "learning_rate": 9.89949228008411e-06,
      "loss": 0.42201433181762693,
      "num_input_tokens_seen": 60598240,
      "step": 1900,
      "train_runtime": 4985.1037,
      "train_tokens_per_second": 12155.864
    },
    {
      "epoch": 0.12826555346081336,
      "grad_norm": 0.8849544336878005,
      "learning_rate": 9.898964108833216e-06,
      "loss": 0.439554500579834,
      "num_input_tokens_seen": 60759632,
      "step": 1905,
      "train_runtime": 4998.2545,
      "train_tokens_per_second": 12156.17
    },
    {
      "epoch": 0.1286022084567735,
      "grad_norm": 0.889455508342976,
      "learning_rate": 9.898434567602629e-06,
      "loss": 0.4066291809082031,
      "num_input_tokens_seen": 60920720,
      "step": 1910,
      "train_runtime": 5011.1512,
      "train_tokens_per_second": 12157.031
    },
    {
      "epoch": 0.12893886345273364,
      "grad_norm": 1.0063744423310308,
      "learning_rate": 9.897903656540437e-06,
      "loss": 0.4452242374420166,
      "num_input_tokens_seen": 61079144,
      "step": 1915,
      "train_runtime": 5023.7899,
      "train_tokens_per_second": 12157.981
    },
    {
      "epoch": 0.12927551844869378,
      "grad_norm": 1.0542159138203344,
      "learning_rate": 9.897371375795103e-06,
      "loss": 0.45699567794799806,
      "num_input_tokens_seen": 61237432,
      "step": 1920,
      "train_runtime": 5036.9101,
      "train_tokens_per_second": 12157.738
    },
    {
      "epoch": 0.12961217344465392,
      "grad_norm": 0.914664701355832,
      "learning_rate": 9.896837725515484e-06,
      "loss": 0.4431015491485596,
      "num_input_tokens_seen": 61393024,
      "step": 1925,
      "train_runtime": 5049.7862,
      "train_tokens_per_second": 12157.549
    },
    {
      "epoch": 0.12994882844061406,
      "grad_norm": 0.7777937706476692,
      "learning_rate": 9.89630270585081e-06,
      "loss": 0.3898467063903809,
      "num_input_tokens_seen": 61553304,
      "step": 1930,
      "train_runtime": 5063.9994,
      "train_tokens_per_second": 12155.077
    },
    {
      "epoch": 0.1302854834365742,
      "grad_norm": 0.9528636579416194,
      "learning_rate": 9.895766316950694e-06,
      "loss": 0.4428374767303467,
      "num_input_tokens_seen": 61710824,
      "step": 1935,
      "train_runtime": 5077.017,
      "train_tokens_per_second": 12154.937
    },
    {
      "epoch": 0.13062213843253434,
      "grad_norm": 0.8688429945161484,
      "learning_rate": 9.89522855896514e-06,
      "loss": 0.4184412002563477,
      "num_input_tokens_seen": 61872784,
      "step": 1940,
      "train_runtime": 5089.6153,
      "train_tokens_per_second": 12156.672
    },
    {
      "epoch": 0.13095879342849448,
      "grad_norm": 0.8882981258541316,
      "learning_rate": 9.89468943204453e-06,
      "loss": 0.42222070693969727,
      "num_input_tokens_seen": 62034264,
      "step": 1945,
      "train_runtime": 5102.7622,
      "train_tokens_per_second": 12156.997
    },
    {
      "epoch": 0.13129544842445462,
      "grad_norm": 0.9682131919957763,
      "learning_rate": 9.89414893633963e-06,
      "loss": 0.4139063835144043,
      "num_input_tokens_seen": 62188736,
      "step": 1950,
      "train_runtime": 5115.4607,
      "train_tokens_per_second": 12157.016
    },
    {
      "epoch": 0.13163210342041476,
      "grad_norm": 0.949298813494029,
      "learning_rate": 9.893607072001584e-06,
      "loss": 0.4268019199371338,
      "num_input_tokens_seen": 62352576,
      "step": 1955,
      "train_runtime": 5127.9193,
      "train_tokens_per_second": 12159.43
    },
    {
      "epoch": 0.1319687584163749,
      "grad_norm": 0.8291848995043224,
      "learning_rate": 9.893063839181925e-06,
      "loss": 0.4225578308105469,
      "num_input_tokens_seen": 62513376,
      "step": 1960,
      "train_runtime": 5140.417,
      "train_tokens_per_second": 12161.149
    },
    {
      "epoch": 0.13230541341233504,
      "grad_norm": 0.8979947424718404,
      "learning_rate": 9.892519238032566e-06,
      "loss": 0.4502850532531738,
      "num_input_tokens_seen": 62674000,
      "step": 1965,
      "train_runtime": 5152.9267,
      "train_tokens_per_second": 12162.797
    },
    {
      "epoch": 0.13264206840829518,
      "grad_norm": 0.917298699883059,
      "learning_rate": 9.891973268705804e-06,
      "loss": 0.43758282661437986,
      "num_input_tokens_seen": 62832344,
      "step": 1970,
      "train_runtime": 5166.6163,
      "train_tokens_per_second": 12161.217
    },
    {
      "epoch": 0.13297872340425532,
      "grad_norm": 0.7773452536847344,
      "learning_rate": 9.891425931354316e-06,
      "loss": 0.4187131404876709,
      "num_input_tokens_seen": 62992416,
      "step": 1975,
      "train_runtime": 5179.3362,
      "train_tokens_per_second": 12162.257
    },
    {
      "epoch": 0.13331537840021546,
      "grad_norm": 1.0238820241724995,
      "learning_rate": 9.890877226131165e-06,
      "loss": 0.43040995597839354,
      "num_input_tokens_seen": 63156088,
      "step": 1980,
      "train_runtime": 5192.4586,
      "train_tokens_per_second": 12163.041
    },
    {
      "epoch": 0.1336520333961756,
      "grad_norm": 0.9235638745612735,
      "learning_rate": 9.890327153189793e-06,
      "loss": 0.4169161319732666,
      "num_input_tokens_seen": 63317336,
      "step": 1985,
      "train_runtime": 5205.0035,
      "train_tokens_per_second": 12164.706
    },
    {
      "epoch": 0.13398868839213574,
      "grad_norm": 0.9909414973730858,
      "learning_rate": 9.889775712684026e-06,
      "loss": 0.42840237617492677,
      "num_input_tokens_seen": 63480248,
      "step": 1990,
      "train_runtime": 5218.107,
      "train_tokens_per_second": 12165.379
    },
    {
      "epoch": 0.13432534338809587,
      "grad_norm": 1.1076612141519002,
      "learning_rate": 9.889222904768074e-06,
      "loss": 0.4068309783935547,
      "num_input_tokens_seen": 63641160,
      "step": 1995,
      "train_runtime": 5231.2841,
      "train_tokens_per_second": 12165.495
    },
    {
      "epoch": 0.13466199838405601,
      "grad_norm": 1.0720905515219408,
      "learning_rate": 9.888668729596528e-06,
      "loss": 0.4225417137145996,
      "num_input_tokens_seen": 63798584,
      "step": 2000,
      "train_runtime": 5244.604,
      "train_tokens_per_second": 12164.614
    },
    {
      "epoch": 0.13499865338001615,
      "grad_norm": 0.9384464485062592,
      "learning_rate": 9.88811318732436e-06,
      "loss": 0.4285943031311035,
      "num_input_tokens_seen": 63952560,
      "step": 2005,
      "train_runtime": 5257.3901,
      "train_tokens_per_second": 12164.317
    },
    {
      "epoch": 0.1353353083759763,
      "grad_norm": 1.0757053762684194,
      "learning_rate": 9.887556278106927e-06,
      "loss": 0.43933773040771484,
      "num_input_tokens_seen": 64114424,
      "step": 2010,
      "train_runtime": 5270.6599,
      "train_tokens_per_second": 12164.402
    },
    {
      "epoch": 0.13567196337193643,
      "grad_norm": 0.95817810959479,
      "learning_rate": 9.886998002099964e-06,
      "loss": 0.43896913528442383,
      "num_input_tokens_seen": 64277520,
      "step": 2015,
      "train_runtime": 5283.6079,
      "train_tokens_per_second": 12165.46
    },
    {
      "epoch": 0.13600861836789657,
      "grad_norm": 0.9545132118654873,
      "learning_rate": 9.886438359459595e-06,
      "loss": 0.45852227210998536,
      "num_input_tokens_seen": 64436808,
      "step": 2020,
      "train_runtime": 5296.8625,
      "train_tokens_per_second": 12165.09
    },
    {
      "epoch": 0.1363452733638567,
      "grad_norm": 0.8832538179927079,
      "learning_rate": 9.885877350342319e-06,
      "loss": 0.4092560768127441,
      "num_input_tokens_seen": 64597360,
      "step": 2025,
      "train_runtime": 5309.885,
      "train_tokens_per_second": 12165.491
    },
    {
      "epoch": 0.13668192835981685,
      "grad_norm": 0.834871105055598,
      "learning_rate": 9.885314974905024e-06,
      "loss": 0.39460463523864747,
      "num_input_tokens_seen": 64761200,
      "step": 2030,
      "train_runtime": 5322.3483,
      "train_tokens_per_second": 12167.787
    },
    {
      "epoch": 0.137018583355777,
      "grad_norm": 0.9915336773494975,
      "learning_rate": 9.884751233304973e-06,
      "loss": 0.4243456840515137,
      "num_input_tokens_seen": 64914432,
      "step": 2035,
      "train_runtime": 5335.5537,
      "train_tokens_per_second": 12166.391
    },
    {
      "epoch": 0.13735523835173713,
      "grad_norm": 0.9056028477037573,
      "learning_rate": 9.884186125699816e-06,
      "loss": 0.4550370693206787,
      "num_input_tokens_seen": 65073232,
      "step": 2040,
      "train_runtime": 5348.0538,
      "train_tokens_per_second": 12167.647
    },
    {
      "epoch": 0.13769189334769727,
      "grad_norm": 1.0062509762778098,
      "learning_rate": 9.883619652247584e-06,
      "loss": 0.4427315711975098,
      "num_input_tokens_seen": 65236072,
      "step": 2045,
      "train_runtime": 5362.2891,
      "train_tokens_per_second": 12165.713
    },
    {
      "epoch": 0.1380285483436574,
      "grad_norm": 0.7808483009573155,
      "learning_rate": 9.883051813106687e-06,
      "loss": 0.37587838172912597,
      "num_input_tokens_seen": 65397080,
      "step": 2050,
      "train_runtime": 5375.969,
      "train_tokens_per_second": 12164.705
    },
    {
      "epoch": 0.13836520333961755,
      "grad_norm": 0.99011518278659,
      "learning_rate": 9.882482608435924e-06,
      "loss": 0.4543206214904785,
      "num_input_tokens_seen": 65558984,
      "step": 2055,
      "train_runtime": 5389.3222,
      "train_tokens_per_second": 12164.606
    },
    {
      "epoch": 0.1387018583355777,
      "grad_norm": 0.9788486450877367,
      "learning_rate": 9.881912038394468e-06,
      "loss": 0.43818678855896,
      "num_input_tokens_seen": 65718448,
      "step": 2060,
      "train_runtime": 5402.1701,
      "train_tokens_per_second": 12165.194
    },
    {
      "epoch": 0.13903851333153783,
      "grad_norm": 0.8491765071717267,
      "learning_rate": 9.881340103141878e-06,
      "loss": 0.450365686416626,
      "num_input_tokens_seen": 65880032,
      "step": 2065,
      "train_runtime": 5414.7268,
      "train_tokens_per_second": 12166.825
    },
    {
      "epoch": 0.13937516832749797,
      "grad_norm": 0.8495415772711788,
      "learning_rate": 9.880766802838092e-06,
      "loss": 0.40735154151916503,
      "num_input_tokens_seen": 66040328,
      "step": 2070,
      "train_runtime": 5427.2214,
      "train_tokens_per_second": 12168.35
    },
    {
      "epoch": 0.1397118233234581,
      "grad_norm": 0.8098099702853278,
      "learning_rate": 9.880192137643434e-06,
      "loss": 0.39875073432922364,
      "num_input_tokens_seen": 66198856,
      "step": 2075,
      "train_runtime": 5440.6209,
      "train_tokens_per_second": 12167.518
    },
    {
      "epoch": 0.14004847831941827,
      "grad_norm": 1.0095396615032801,
      "learning_rate": 9.879616107718608e-06,
      "loss": 0.41994829177856446,
      "num_input_tokens_seen": 66355264,
      "step": 2080,
      "train_runtime": 5454.1025,
      "train_tokens_per_second": 12166.12
    },
    {
      "epoch": 0.1403851333153784,
      "grad_norm": 0.8761389171509144,
      "learning_rate": 9.879038713224695e-06,
      "loss": 0.43233041763305663,
      "num_input_tokens_seen": 66519080,
      "step": 2085,
      "train_runtime": 5467.2099,
      "train_tokens_per_second": 12166.915
    },
    {
      "epoch": 0.14072178831133855,
      "grad_norm": 0.8347151191842973,
      "learning_rate": 9.878459954323165e-06,
      "loss": 0.4214962005615234,
      "num_input_tokens_seen": 66679536,
      "step": 2090,
      "train_runtime": 5480.0942,
      "train_tokens_per_second": 12167.589
    },
    {
      "epoch": 0.1410584433072987,
      "grad_norm": 0.9102580199880363,
      "learning_rate": 9.877879831175865e-06,
      "loss": 0.40982770919799805,
      "num_input_tokens_seen": 66838848,
      "step": 2095,
      "train_runtime": 5492.7321,
      "train_tokens_per_second": 12168.598
    },
    {
      "epoch": 0.14139509830325883,
      "grad_norm": 0.7947832663386954,
      "learning_rate": 9.877298343945025e-06,
      "loss": 0.4155694007873535,
      "num_input_tokens_seen": 67001488,
      "step": 2100,
      "train_runtime": 5506.2047,
      "train_tokens_per_second": 12168.361
    },
    {
      "epoch": 0.14173175329921897,
      "grad_norm": 0.8056645514749221,
      "learning_rate": 9.876715492793256e-06,
      "loss": 0.41213302612304686,
      "num_input_tokens_seen": 67162024,
      "step": 2105,
      "train_runtime": 5519.9743,
      "train_tokens_per_second": 12167.09
    },
    {
      "epoch": 0.1420684082951791,
      "grad_norm": 1.0016942058541025,
      "learning_rate": 9.87613127788355e-06,
      "loss": 0.38554019927978517,
      "num_input_tokens_seen": 67316344,
      "step": 2110,
      "train_runtime": 5532.6493,
      "train_tokens_per_second": 12167.108
    },
    {
      "epoch": 0.14240506329113925,
      "grad_norm": 1.0181137444524948,
      "learning_rate": 9.87554569937928e-06,
      "loss": 0.48628959655761717,
      "num_input_tokens_seen": 67480184,
      "step": 2115,
      "train_runtime": 5545.0919,
      "train_tokens_per_second": 12169.353
    },
    {
      "epoch": 0.1427417182870994,
      "grad_norm": 0.986771189199152,
      "learning_rate": 9.874958757444203e-06,
      "loss": 0.4479678153991699,
      "num_input_tokens_seen": 67644024,
      "step": 2120,
      "train_runtime": 5557.5597,
      "train_tokens_per_second": 12171.533
    },
    {
      "epoch": 0.14307837328305953,
      "grad_norm": 0.9172627224791394,
      "learning_rate": 9.874370452242454e-06,
      "loss": 0.4456611633300781,
      "num_input_tokens_seen": 67806360,
      "step": 2125,
      "train_runtime": 5570.1901,
      "train_tokens_per_second": 12173.078
    },
    {
      "epoch": 0.14341502827901967,
      "grad_norm": 0.9169203833420295,
      "learning_rate": 9.873780783938553e-06,
      "loss": 0.46516051292419436,
      "num_input_tokens_seen": 67966688,
      "step": 2130,
      "train_runtime": 5583.0612,
      "train_tokens_per_second": 12173.732
    },
    {
      "epoch": 0.1437516832749798,
      "grad_norm": 0.825432726696393,
      "learning_rate": 9.873189752697396e-06,
      "loss": 0.3940959215164185,
      "num_input_tokens_seen": 68126832,
      "step": 2135,
      "train_runtime": 5596.261,
      "train_tokens_per_second": 12173.634
    },
    {
      "epoch": 0.14408833827093995,
      "grad_norm": 0.869345593681209,
      "learning_rate": 9.872597358684265e-06,
      "loss": 0.45507278442382815,
      "num_input_tokens_seen": 68281776,
      "step": 2140,
      "train_runtime": 5609.5935,
      "train_tokens_per_second": 12172.322
    },
    {
      "epoch": 0.1444249932669001,
      "grad_norm": 0.9128839986312763,
      "learning_rate": 9.872003602064816e-06,
      "loss": 0.42637219429016116,
      "num_input_tokens_seen": 68443200,
      "step": 2145,
      "train_runtime": 5622.7866,
      "train_tokens_per_second": 12172.47
    },
    {
      "epoch": 0.14476164826286023,
      "grad_norm": 1.5946973137189007,
      "learning_rate": 9.8714084830051e-06,
      "loss": 0.4010769844055176,
      "num_input_tokens_seen": 68605992,
      "step": 2150,
      "train_runtime": 5636.282,
      "train_tokens_per_second": 12172.207
    },
    {
      "epoch": 0.14509830325882037,
      "grad_norm": 0.8619865349758978,
      "learning_rate": 9.870812001671533e-06,
      "loss": 0.4232743740081787,
      "num_input_tokens_seen": 68766816,
      "step": 2155,
      "train_runtime": 5649.7052,
      "train_tokens_per_second": 12171.753
    },
    {
      "epoch": 0.1454349582547805,
      "grad_norm": 1.0450211873307693,
      "learning_rate": 9.870214158230922e-06,
      "loss": 0.4479037284851074,
      "num_input_tokens_seen": 68926608,
      "step": 2160,
      "train_runtime": 5662.5827,
      "train_tokens_per_second": 12172.292
    },
    {
      "epoch": 0.14577161325074064,
      "grad_norm": 0.8433864802768252,
      "learning_rate": 9.869614952850454e-06,
      "loss": 0.4208380699157715,
      "num_input_tokens_seen": 69087056,
      "step": 2165,
      "train_runtime": 5675.9909,
      "train_tokens_per_second": 12171.805
    },
    {
      "epoch": 0.14610826824670078,
      "grad_norm": 0.9697911217534166,
      "learning_rate": 9.86901438569769e-06,
      "loss": 0.45280089378356936,
      "num_input_tokens_seen": 69246632,
      "step": 2170,
      "train_runtime": 5688.7466,
      "train_tokens_per_second": 12172.564
    },
    {
      "epoch": 0.14644492324266092,
      "grad_norm": 0.9135087367546444,
      "learning_rate": 9.86841245694058e-06,
      "loss": 0.4174619197845459,
      "num_input_tokens_seen": 69400416,
      "step": 2175,
      "train_runtime": 5702.6752,
      "train_tokens_per_second": 12169.8
    },
    {
      "epoch": 0.14678157823862106,
      "grad_norm": 0.9724829776782624,
      "learning_rate": 9.86780916674745e-06,
      "loss": 0.4612767219543457,
      "num_input_tokens_seen": 69554912,
      "step": 2180,
      "train_runtime": 5716.8379,
      "train_tokens_per_second": 12166.675
    },
    {
      "epoch": 0.1471182332345812,
      "grad_norm": 0.8728710316147338,
      "learning_rate": 9.86720451528701e-06,
      "loss": 0.41460409164428713,
      "num_input_tokens_seen": 69711744,
      "step": 2185,
      "train_runtime": 5730.3482,
      "train_tokens_per_second": 12165.359
    },
    {
      "epoch": 0.14745488823054134,
      "grad_norm": 0.8781031333623331,
      "learning_rate": 9.866598502728348e-06,
      "loss": 0.42920470237731934,
      "num_input_tokens_seen": 69867784,
      "step": 2190,
      "train_runtime": 5744.164,
      "train_tokens_per_second": 12163.264
    },
    {
      "epoch": 0.14779154322650148,
      "grad_norm": 0.9045139793419407,
      "learning_rate": 9.865991129240931e-06,
      "loss": 0.44884295463562013,
      "num_input_tokens_seen": 70030376,
      "step": 2195,
      "train_runtime": 5758.4003,
      "train_tokens_per_second": 12161.429
    },
    {
      "epoch": 0.14812819822246162,
      "grad_norm": 0.958863730674743,
      "learning_rate": 9.865382394994614e-06,
      "loss": 0.42292466163635256,
      "num_input_tokens_seen": 70189512,
      "step": 2200,
      "train_runtime": 5772.4201,
      "train_tokens_per_second": 12159.46
    },
    {
      "epoch": 0.14846485321842176,
      "grad_norm": 0.7648537293414875,
      "learning_rate": 9.864772300159622e-06,
      "loss": 0.40708608627319337,
      "num_input_tokens_seen": 70348864,
      "step": 2205,
      "train_runtime": 5785.619,
      "train_tokens_per_second": 12159.263
    },
    {
      "epoch": 0.1488015082143819,
      "grad_norm": 1.1608650859727252,
      "learning_rate": 9.864160844906571e-06,
      "loss": 0.4426380157470703,
      "num_input_tokens_seen": 70505376,
      "step": 2210,
      "train_runtime": 5798.6779,
      "train_tokens_per_second": 12158.871
    },
    {
      "epoch": 0.14913816321034204,
      "grad_norm": 0.9109203404235952,
      "learning_rate": 9.863548029406449e-06,
      "loss": 0.4217958450317383,
      "num_input_tokens_seen": 70665688,
      "step": 2215,
      "train_runtime": 5811.9703,
      "train_tokens_per_second": 12158.646
    },
    {
      "epoch": 0.14947481820630218,
      "grad_norm": 0.8430512469870374,
      "learning_rate": 9.86293385383063e-06,
      "loss": 0.40331134796142576,
      "num_input_tokens_seen": 70825128,
      "step": 2220,
      "train_runtime": 5825.3252,
      "train_tokens_per_second": 12158.141
    },
    {
      "epoch": 0.14981147320226232,
      "grad_norm": 1.008019018116167,
      "learning_rate": 9.862318318350865e-06,
      "loss": 0.4379251956939697,
      "num_input_tokens_seen": 70984600,
      "step": 2225,
      "train_runtime": 5838.3079,
      "train_tokens_per_second": 12158.42
    },
    {
      "epoch": 0.15014812819822246,
      "grad_norm": 0.9655406579116413,
      "learning_rate": 9.861701423139287e-06,
      "loss": 0.3935262203216553,
      "num_input_tokens_seen": 71148136,
      "step": 2230,
      "train_runtime": 5851.4308,
      "train_tokens_per_second": 12159.101
    },
    {
      "epoch": 0.1504847831941826,
      "grad_norm": 0.9093511787917998,
      "learning_rate": 9.861083168368411e-06,
      "loss": 0.42951068878173826,
      "num_input_tokens_seen": 71311368,
      "step": 2235,
      "train_runtime": 5865.1801,
      "train_tokens_per_second": 12158.428
    },
    {
      "epoch": 0.15082143819014274,
      "grad_norm": 0.9401396276354811,
      "learning_rate": 9.860463554211125e-06,
      "loss": 0.3876911401748657,
      "num_input_tokens_seen": 71473080,
      "step": 2240,
      "train_runtime": 5878.2512,
      "train_tokens_per_second": 12158.902
    },
    {
      "epoch": 0.15115809318610288,
      "grad_norm": 0.9564800891090146,
      "learning_rate": 9.859842580840706e-06,
      "loss": 0.47974791526794436,
      "num_input_tokens_seen": 71633144,
      "step": 2245,
      "train_runtime": 5890.7142,
      "train_tokens_per_second": 12160.35
    },
    {
      "epoch": 0.15149474818206302,
      "grad_norm": 0.8089120994707854,
      "learning_rate": 9.859220248430806e-06,
      "loss": 0.42336549758911135,
      "num_input_tokens_seen": 71796984,
      "step": 2250,
      "train_runtime": 5903.1665,
      "train_tokens_per_second": 12162.453
    },
    {
      "epoch": 0.15183140317802316,
      "grad_norm": 0.7748896563225325,
      "learning_rate": 9.85859655715546e-06,
      "loss": 0.4047839164733887,
      "num_input_tokens_seen": 71959704,
      "step": 2255,
      "train_runtime": 5915.6264,
      "train_tokens_per_second": 12164.342
    },
    {
      "epoch": 0.1521680581739833,
      "grad_norm": 0.9576851563232459,
      "learning_rate": 9.857971507189077e-06,
      "loss": 0.39892871379852296,
      "num_input_tokens_seen": 72121744,
      "step": 2260,
      "train_runtime": 5929.0855,
      "train_tokens_per_second": 12164.059
    },
    {
      "epoch": 0.15250471316994343,
      "grad_norm": 0.9744963176053137,
      "learning_rate": 9.857345098706455e-06,
      "loss": 0.43654985427856446,
      "num_input_tokens_seen": 72284992,
      "step": 2265,
      "train_runtime": 5942.0473,
      "train_tokens_per_second": 12164.998
    },
    {
      "epoch": 0.15284136816590357,
      "grad_norm": 0.9424102984854913,
      "learning_rate": 9.856717331882765e-06,
      "loss": 0.41117372512817385,
      "num_input_tokens_seen": 72439104,
      "step": 2270,
      "train_runtime": 5955.4614,
      "train_tokens_per_second": 12163.475
    },
    {
      "epoch": 0.1531780231618637,
      "grad_norm": 0.7862135004212158,
      "learning_rate": 9.85608820689356e-06,
      "loss": 0.40647225379943847,
      "num_input_tokens_seen": 72598448,
      "step": 2275,
      "train_runtime": 5968.2423,
      "train_tokens_per_second": 12164.126
    },
    {
      "epoch": 0.15351467815782385,
      "grad_norm": 0.8981793910298355,
      "learning_rate": 9.855457723914772e-06,
      "loss": 0.41314330101013186,
      "num_input_tokens_seen": 72755712,
      "step": 2280,
      "train_runtime": 5981.6016,
      "train_tokens_per_second": 12163.249
    },
    {
      "epoch": 0.153851333153784,
      "grad_norm": 0.9094111779304819,
      "learning_rate": 9.854825883122713e-06,
      "loss": 0.39554009437561033,
      "num_input_tokens_seen": 72916040,
      "step": 2285,
      "train_runtime": 5994.8711,
      "train_tokens_per_second": 12163.071
    },
    {
      "epoch": 0.15418798814974413,
      "grad_norm": 0.8669042633507054,
      "learning_rate": 9.85419268469408e-06,
      "loss": 0.3989769697189331,
      "num_input_tokens_seen": 73073200,
      "step": 2290,
      "train_runtime": 6009.0924,
      "train_tokens_per_second": 12160.439
    },
    {
      "epoch": 0.15452464314570427,
      "grad_norm": 0.8198038272205199,
      "learning_rate": 9.85355812880594e-06,
      "loss": 0.3911727428436279,
      "num_input_tokens_seen": 73232400,
      "step": 2295,
      "train_runtime": 6022.5474,
      "train_tokens_per_second": 12159.705
    },
    {
      "epoch": 0.1548612981416644,
      "grad_norm": 0.8637946234496597,
      "learning_rate": 9.852922215635747e-06,
      "loss": 0.3806226968765259,
      "num_input_tokens_seen": 73390896,
      "step": 2300,
      "train_runtime": 6035.8007,
      "train_tokens_per_second": 12159.264
    },
    {
      "epoch": 0.15519795313762455,
      "grad_norm": 1.040980399846696,
      "learning_rate": 9.85228494536133e-06,
      "loss": 0.44347715377807617,
      "num_input_tokens_seen": 73549800,
      "step": 2305,
      "train_runtime": 6049.1517,
      "train_tokens_per_second": 12158.697
    },
    {
      "epoch": 0.1555346081335847,
      "grad_norm": 0.9415171415056507,
      "learning_rate": 9.851646318160901e-06,
      "loss": 0.4267573833465576,
      "num_input_tokens_seen": 73710168,
      "step": 2310,
      "train_runtime": 6062.413,
      "train_tokens_per_second": 12158.553
    },
    {
      "epoch": 0.15587126312954483,
      "grad_norm": 0.8401530802692957,
      "learning_rate": 9.85100633421305e-06,
      "loss": 0.39694502353668215,
      "num_input_tokens_seen": 73868280,
      "step": 2315,
      "train_runtime": 6075.6392,
      "train_tokens_per_second": 12158.108
    },
    {
      "epoch": 0.15620791812550497,
      "grad_norm": 0.93708089274018,
      "learning_rate": 9.850364993696749e-06,
      "loss": 0.43337383270263674,
      "num_input_tokens_seen": 74028280,
      "step": 2320,
      "train_runtime": 6088.7386,
      "train_tokens_per_second": 12158.229
    },
    {
      "epoch": 0.15654457312146514,
      "grad_norm": 0.9432320297946613,
      "learning_rate": 9.849722296791339e-06,
      "loss": 0.4253643035888672,
      "num_input_tokens_seen": 74187608,
      "step": 2325,
      "train_runtime": 6102.8667,
      "train_tokens_per_second": 12156.19
    },
    {
      "epoch": 0.15688122811742528,
      "grad_norm": 0.9514214571023777,
      "learning_rate": 9.849078243676558e-06,
      "loss": 0.44068102836608886,
      "num_input_tokens_seen": 74343432,
      "step": 2330,
      "train_runtime": 6115.8292,
      "train_tokens_per_second": 12155.904
    },
    {
      "epoch": 0.15721788311338541,
      "grad_norm": 0.8444722942485434,
      "learning_rate": 9.848432834532506e-06,
      "loss": 0.39463765621185304,
      "num_input_tokens_seen": 74503128,
      "step": 2335,
      "train_runtime": 6128.4435,
      "train_tokens_per_second": 12156.941
    },
    {
      "epoch": 0.15755453810934555,
      "grad_norm": 0.9842436989673532,
      "learning_rate": 9.847786069539673e-06,
      "loss": 0.443955135345459,
      "num_input_tokens_seen": 74665328,
      "step": 2340,
      "train_runtime": 6141.4003,
      "train_tokens_per_second": 12157.704
    },
    {
      "epoch": 0.1578911931053057,
      "grad_norm": 0.8763206059627097,
      "learning_rate": 9.847137948878926e-06,
      "loss": 0.4358677387237549,
      "num_input_tokens_seen": 74824624,
      "step": 2345,
      "train_runtime": 6154.7616,
      "train_tokens_per_second": 12157.193
    },
    {
      "epoch": 0.15822784810126583,
      "grad_norm": 0.9931358669314813,
      "learning_rate": 9.846488472731505e-06,
      "loss": 0.4636721134185791,
      "num_input_tokens_seen": 74987320,
      "step": 2350,
      "train_runtime": 6167.4837,
      "train_tokens_per_second": 12158.495
    },
    {
      "epoch": 0.15856450309722597,
      "grad_norm": 0.8697944936866674,
      "learning_rate": 9.845837641279038e-06,
      "loss": 0.4378199577331543,
      "num_input_tokens_seen": 75148664,
      "step": 2355,
      "train_runtime": 6180.3608,
      "train_tokens_per_second": 12159.268
    },
    {
      "epoch": 0.1589011580931861,
      "grad_norm": 1.0143506592640161,
      "learning_rate": 9.845185454703528e-06,
      "loss": 0.4419397830963135,
      "num_input_tokens_seen": 75309896,
      "step": 2360,
      "train_runtime": 6193.4918,
      "train_tokens_per_second": 12159.521
    },
    {
      "epoch": 0.15923781308914625,
      "grad_norm": 0.8430647178312407,
      "learning_rate": 9.844531913187356e-06,
      "loss": 0.3914985418319702,
      "num_input_tokens_seen": 75469432,
      "step": 2365,
      "train_runtime": 6206.8694,
      "train_tokens_per_second": 12159.017
    },
    {
      "epoch": 0.1595744680851064,
      "grad_norm": 1.0483327683762995,
      "learning_rate": 9.84387701691328e-06,
      "loss": 0.4408010482788086,
      "num_input_tokens_seen": 75624264,
      "step": 2370,
      "train_runtime": 6220.8391,
      "train_tokens_per_second": 12156.602
    },
    {
      "epoch": 0.15991112308106653,
      "grad_norm": 1.0875999522455426,
      "learning_rate": 9.843220766064443e-06,
      "loss": 0.4410550117492676,
      "num_input_tokens_seen": 75775016,
      "step": 2375,
      "train_runtime": 6233.8454,
      "train_tokens_per_second": 12155.421
    },
    {
      "epoch": 0.16024777807702667,
      "grad_norm": 0.9938095872966253,
      "learning_rate": 9.842563160824364e-06,
      "loss": 0.45755348205566404,
      "num_input_tokens_seen": 75929952,
      "step": 2380,
      "train_runtime": 6246.8067,
      "train_tokens_per_second": 12155.003
    },
    {
      "epoch": 0.1605844330729868,
      "grad_norm": 0.9224873550407295,
      "learning_rate": 9.841904201376939e-06,
      "loss": 0.44507179260253904,
      "num_input_tokens_seen": 76091720,
      "step": 2385,
      "train_runtime": 6259.4067,
      "train_tokens_per_second": 12156.379
    },
    {
      "epoch": 0.16092108806894695,
      "grad_norm": 0.8954708178473121,
      "learning_rate": 9.841243887906443e-06,
      "loss": 0.3931128740310669,
      "num_input_tokens_seen": 76248824,
      "step": 2390,
      "train_runtime": 6272.2133,
      "train_tokens_per_second": 12156.606
    },
    {
      "epoch": 0.1612577430649071,
      "grad_norm": 0.9715207985878648,
      "learning_rate": 9.840582220597532e-06,
      "loss": 0.41417856216430665,
      "num_input_tokens_seen": 76409640,
      "step": 2395,
      "train_runtime": 6284.8091,
      "train_tokens_per_second": 12157.83
    },
    {
      "epoch": 0.16159439806086723,
      "grad_norm": 1.0263942745724974,
      "learning_rate": 9.839919199635238e-06,
      "loss": 0.44920849800109863,
      "num_input_tokens_seen": 76568248,
      "step": 2400,
      "train_runtime": 6297.7759,
      "train_tokens_per_second": 12157.982
    },
    {
      "epoch": 0.16193105305682737,
      "grad_norm": 0.8954264246418393,
      "learning_rate": 9.839254825204973e-06,
      "loss": 0.4416805744171143,
      "num_input_tokens_seen": 76732088,
      "step": 2405,
      "train_runtime": 6310.208,
      "train_tokens_per_second": 12159.993
    },
    {
      "epoch": 0.1622677080527875,
      "grad_norm": 0.8590077363557056,
      "learning_rate": 9.838589097492527e-06,
      "loss": 0.42504053115844725,
      "num_input_tokens_seen": 76893152,
      "step": 2410,
      "train_runtime": 6323.2115,
      "train_tokens_per_second": 12160.459
    },
    {
      "epoch": 0.16260436304874765,
      "grad_norm": 1.6131964591850911,
      "learning_rate": 9.83792201668407e-06,
      "loss": 0.4367863655090332,
      "num_input_tokens_seen": 77054616,
      "step": 2415,
      "train_runtime": 6335.7627,
      "train_tokens_per_second": 12161.853
    },
    {
      "epoch": 0.16294101804470779,
      "grad_norm": 0.9932610149244813,
      "learning_rate": 9.837253582966146e-06,
      "loss": 0.43808488845825194,
      "num_input_tokens_seen": 77215984,
      "step": 2420,
      "train_runtime": 6348.4083,
      "train_tokens_per_second": 12163.046
    },
    {
      "epoch": 0.16327767304066793,
      "grad_norm": 0.8794695278434497,
      "learning_rate": 9.836583796525686e-06,
      "loss": 0.40805864334106445,
      "num_input_tokens_seen": 77370960,
      "step": 2425,
      "train_runtime": 6361.9206,
      "train_tokens_per_second": 12161.573
    },
    {
      "epoch": 0.16361432803662807,
      "grad_norm": 0.8687875334455333,
      "learning_rate": 9.835912657549986e-06,
      "loss": 0.4591360092163086,
      "num_input_tokens_seen": 77534032,
      "step": 2430,
      "train_runtime": 6375.4291,
      "train_tokens_per_second": 12161.382
    },
    {
      "epoch": 0.1639509830325882,
      "grad_norm": 1.2016065936588156,
      "learning_rate": 9.835240166226734e-06,
      "loss": 0.397727370262146,
      "num_input_tokens_seen": 77688728,
      "step": 2435,
      "train_runtime": 6388.0355,
      "train_tokens_per_second": 12161.599
    },
    {
      "epoch": 0.16428763802854834,
      "grad_norm": 0.8731767631359908,
      "learning_rate": 9.834566322743987e-06,
      "loss": 0.4353171825408936,
      "num_input_tokens_seen": 77847832,
      "step": 2440,
      "train_runtime": 6401.7612,
      "train_tokens_per_second": 12160.377
    },
    {
      "epoch": 0.16462429302450848,
      "grad_norm": 0.8438337804367104,
      "learning_rate": 9.833891127290186e-06,
      "loss": 0.43749122619628905,
      "num_input_tokens_seen": 78011296,
      "step": 2445,
      "train_runtime": 6414.7845,
      "train_tokens_per_second": 12161.172
    },
    {
      "epoch": 0.16496094802046862,
      "grad_norm": 0.8870021679922767,
      "learning_rate": 9.833214580054145e-06,
      "loss": 0.4270503997802734,
      "num_input_tokens_seen": 78172288,
      "step": 2450,
      "train_runtime": 6428.2269,
      "train_tokens_per_second": 12160.785
    },
    {
      "epoch": 0.16529760301642876,
      "grad_norm": 0.9237164801991081,
      "learning_rate": 9.832536681225058e-06,
      "loss": 0.4602365493774414,
      "num_input_tokens_seen": 78332160,
      "step": 2455,
      "train_runtime": 6440.9361,
      "train_tokens_per_second": 12161.611
    },
    {
      "epoch": 0.1656342580123889,
      "grad_norm": 0.9214469997052653,
      "learning_rate": 9.831857430992497e-06,
      "loss": 0.40315423011779783,
      "num_input_tokens_seen": 78493104,
      "step": 2460,
      "train_runtime": 6455.0697,
      "train_tokens_per_second": 12159.916
    },
    {
      "epoch": 0.16597091300834904,
      "grad_norm": 0.8056048888909143,
      "learning_rate": 9.831176829546416e-06,
      "loss": 0.42696781158447267,
      "num_input_tokens_seen": 78655152,
      "step": 2465,
      "train_runtime": 6468.3366,
      "train_tokens_per_second": 12160.028
    },
    {
      "epoch": 0.16630756800430918,
      "grad_norm": 1.0553250985583589,
      "learning_rate": 9.830494877077137e-06,
      "loss": 0.4728592872619629,
      "num_input_tokens_seen": 78816776,
      "step": 2470,
      "train_runtime": 6480.9069,
      "train_tokens_per_second": 12161.381
    },
    {
      "epoch": 0.16664422300026932,
      "grad_norm": 0.9689257334196517,
      "learning_rate": 9.82981157377537e-06,
      "loss": 0.4275308609008789,
      "num_input_tokens_seen": 78976120,
      "step": 2475,
      "train_runtime": 6493.9932,
      "train_tokens_per_second": 12161.411
    },
    {
      "epoch": 0.16698087799622946,
      "grad_norm": 0.8285231772392113,
      "learning_rate": 9.829126919832198e-06,
      "loss": 0.3852450132369995,
      "num_input_tokens_seen": 79135080,
      "step": 2480,
      "train_runtime": 6507.2063,
      "train_tokens_per_second": 12161.145
    },
    {
      "epoch": 0.1673175329921896,
      "grad_norm": 0.8967928469902711,
      "learning_rate": 9.82844091543908e-06,
      "loss": 0.4357333183288574,
      "num_input_tokens_seen": 79298848,
      "step": 2485,
      "train_runtime": 6520.3067,
      "train_tokens_per_second": 12161.828
    },
    {
      "epoch": 0.16765418798814974,
      "grad_norm": 0.9143571418251295,
      "learning_rate": 9.82775356078786e-06,
      "loss": 0.42016186714172366,
      "num_input_tokens_seen": 79456992,
      "step": 2490,
      "train_runtime": 6533.9499,
      "train_tokens_per_second": 12160.637
    },
    {
      "epoch": 0.16799084298410988,
      "grad_norm": 0.9118578208778196,
      "learning_rate": 9.82706485607075e-06,
      "loss": 0.38487706184387205,
      "num_input_tokens_seen": 79611344,
      "step": 2495,
      "train_runtime": 6547.0016,
      "train_tokens_per_second": 12159.97
    },
    {
      "epoch": 0.16832749798007002,
      "grad_norm": 0.8450782831368749,
      "learning_rate": 9.826374801480346e-06,
      "loss": 0.410227632522583,
      "num_input_tokens_seen": 79770824,
      "step": 2500,
      "train_runtime": 6560.7522,
      "train_tokens_per_second": 12158.792
    },
    {
      "epoch": 0.16866415297603016,
      "grad_norm": 0.8375845343278367,
      "learning_rate": 9.825683397209617e-06,
      "loss": 0.4221829414367676,
      "num_input_tokens_seen": 79920104,
      "step": 2505,
      "train_runtime": 6573.4772,
      "train_tokens_per_second": 12157.965
    },
    {
      "epoch": 0.1690008079719903,
      "grad_norm": 0.7837675798267905,
      "learning_rate": 9.824990643451915e-06,
      "loss": 0.4117011547088623,
      "num_input_tokens_seen": 80081520,
      "step": 2510,
      "train_runtime": 6586.5265,
      "train_tokens_per_second": 12158.384
    },
    {
      "epoch": 0.16933746296795044,
      "grad_norm": 0.8697124256039522,
      "learning_rate": 9.824296540400965e-06,
      "loss": 0.40143237113952634,
      "num_input_tokens_seen": 80244744,
      "step": 2515,
      "train_runtime": 6599.4131,
      "train_tokens_per_second": 12159.376
    },
    {
      "epoch": 0.16967411796391058,
      "grad_norm": 1.0611311966583283,
      "learning_rate": 9.823601088250872e-06,
      "loss": 0.4585402488708496,
      "num_input_tokens_seen": 80398424,
      "step": 2520,
      "train_runtime": 6612.251,
      "train_tokens_per_second": 12159.01
    },
    {
      "epoch": 0.17001077295987072,
      "grad_norm": 0.8557854915028446,
      "learning_rate": 9.822904287196114e-06,
      "loss": 0.43123817443847656,
      "num_input_tokens_seen": 80557272,
      "step": 2525,
      "train_runtime": 6625.5651,
      "train_tokens_per_second": 12158.551
    },
    {
      "epoch": 0.17034742795583085,
      "grad_norm": 0.8191619605994133,
      "learning_rate": 9.822206137431553e-06,
      "loss": 0.40825510025024414,
      "num_input_tokens_seen": 80718240,
      "step": 2530,
      "train_runtime": 6638.029,
      "train_tokens_per_second": 12159.971
    },
    {
      "epoch": 0.170684082951791,
      "grad_norm": 0.8126959627832543,
      "learning_rate": 9.821506639152421e-06,
      "loss": 0.4190530300140381,
      "num_input_tokens_seen": 80879840,
      "step": 2535,
      "train_runtime": 6651.1124,
      "train_tokens_per_second": 12160.348
    },
    {
      "epoch": 0.17102073794775113,
      "grad_norm": 0.8496155390347591,
      "learning_rate": 9.820805792554334e-06,
      "loss": 0.4158648490905762,
      "num_input_tokens_seen": 81041736,
      "step": 2540,
      "train_runtime": 6664.2902,
      "train_tokens_per_second": 12160.595
    },
    {
      "epoch": 0.17135739294371127,
      "grad_norm": 0.9451309622270121,
      "learning_rate": 9.820103597833278e-06,
      "loss": 0.4214468002319336,
      "num_input_tokens_seen": 81201000,
      "step": 2545,
      "train_runtime": 6677.4235,
      "train_tokens_per_second": 12160.529
    },
    {
      "epoch": 0.1716940479396714,
      "grad_norm": 0.8402392769387538,
      "learning_rate": 9.819400055185622e-06,
      "loss": 0.4187011241912842,
      "num_input_tokens_seen": 81361200,
      "step": 2550,
      "train_runtime": 6690.8951,
      "train_tokens_per_second": 12159.987
    },
    {
      "epoch": 0.17203070293563155,
      "grad_norm": 0.9037007349024135,
      "learning_rate": 9.818695164808108e-06,
      "loss": 0.41239018440246583,
      "num_input_tokens_seen": 81523800,
      "step": 2555,
      "train_runtime": 6703.5938,
      "train_tokens_per_second": 12161.208
    },
    {
      "epoch": 0.1723673579315917,
      "grad_norm": 0.9297174818427115,
      "learning_rate": 9.817988926897856e-06,
      "loss": 0.39138197898864746,
      "num_input_tokens_seen": 81682504,
      "step": 2560,
      "train_runtime": 6716.8172,
      "train_tokens_per_second": 12160.894
    },
    {
      "epoch": 0.17270401292755186,
      "grad_norm": 0.9393954949772274,
      "learning_rate": 9.817281341652367e-06,
      "loss": 0.42818775177001955,
      "num_input_tokens_seen": 81844192,
      "step": 2565,
      "train_runtime": 6729.3381,
      "train_tokens_per_second": 12162.294
    },
    {
      "epoch": 0.173040667923512,
      "grad_norm": 0.8832508392667486,
      "learning_rate": 9.816572409269508e-06,
      "loss": 0.402614688873291,
      "num_input_tokens_seen": 82006440,
      "step": 2570,
      "train_runtime": 6742.1024,
      "train_tokens_per_second": 12163.334
    },
    {
      "epoch": 0.17337732291947214,
      "grad_norm": 0.8466478124703766,
      "learning_rate": 9.815862129947537e-06,
      "loss": 0.42305798530578614,
      "num_input_tokens_seen": 82166424,
      "step": 2575,
      "train_runtime": 6754.9399,
      "train_tokens_per_second": 12163.902
    },
    {
      "epoch": 0.17371397791543228,
      "grad_norm": 0.774763775922723,
      "learning_rate": 9.815150503885078e-06,
      "loss": 0.367329740524292,
      "num_input_tokens_seen": 82328816,
      "step": 2580,
      "train_runtime": 6768.293,
      "train_tokens_per_second": 12163.897
    },
    {
      "epoch": 0.17405063291139242,
      "grad_norm": 0.9261869754299975,
      "learning_rate": 9.814437531281133e-06,
      "loss": 0.40413622856140136,
      "num_input_tokens_seen": 82491080,
      "step": 2585,
      "train_runtime": 6781.6228,
      "train_tokens_per_second": 12163.915
    },
    {
      "epoch": 0.17438728790735256,
      "grad_norm": 1.0138226450468546,
      "learning_rate": 9.813723212335085e-06,
      "loss": 0.45269336700439455,
      "num_input_tokens_seen": 82643744,
      "step": 2590,
      "train_runtime": 6794.8755,
      "train_tokens_per_second": 12162.658
    },
    {
      "epoch": 0.1747239429033127,
      "grad_norm": 0.8426136420315739,
      "learning_rate": 9.813007547246691e-06,
      "loss": 0.4036527633666992,
      "num_input_tokens_seen": 82805256,
      "step": 2595,
      "train_runtime": 6807.6723,
      "train_tokens_per_second": 12163.52
    },
    {
      "epoch": 0.17506059789927284,
      "grad_norm": 0.8414014497399007,
      "learning_rate": 9.812290536216083e-06,
      "loss": 0.4595314025878906,
      "num_input_tokens_seen": 82958360,
      "step": 2600,
      "train_runtime": 6821.2966,
      "train_tokens_per_second": 12161.67
    },
    {
      "epoch": 0.17539725289523297,
      "grad_norm": 0.7955425707874572,
      "learning_rate": 9.811572179443773e-06,
      "loss": 0.38980703353881835,
      "num_input_tokens_seen": 83119088,
      "step": 2605,
      "train_runtime": 6833.7817,
      "train_tokens_per_second": 12162.971
    },
    {
      "epoch": 0.17573390789119311,
      "grad_norm": 0.8183931476197165,
      "learning_rate": 9.810852477130645e-06,
      "loss": 0.41729736328125,
      "num_input_tokens_seen": 83274536,
      "step": 2610,
      "train_runtime": 6847.748,
      "train_tokens_per_second": 12160.865
    },
    {
      "epoch": 0.17607056288715325,
      "grad_norm": 0.8895853968317164,
      "learning_rate": 9.810131429477964e-06,
      "loss": 0.39148788452148436,
      "num_input_tokens_seen": 83435416,
      "step": 2615,
      "train_runtime": 6860.1969,
      "train_tokens_per_second": 12162.248
    },
    {
      "epoch": 0.1764072178831134,
      "grad_norm": 0.8620651172704589,
      "learning_rate": 9.809409036687364e-06,
      "loss": 0.4088418006896973,
      "num_input_tokens_seen": 83597400,
      "step": 2620,
      "train_runtime": 6873.4452,
      "train_tokens_per_second": 12162.372
    },
    {
      "epoch": 0.17674387287907353,
      "grad_norm": 0.8246484917953475,
      "learning_rate": 9.808685298960867e-06,
      "loss": 0.4190103054046631,
      "num_input_tokens_seen": 83758200,
      "step": 2625,
      "train_runtime": 6886.2817,
      "train_tokens_per_second": 12163.052
    },
    {
      "epoch": 0.17708052787503367,
      "grad_norm": 0.8588629656167998,
      "learning_rate": 9.807960216500858e-06,
      "loss": 0.42969322204589844,
      "num_input_tokens_seen": 83915552,
      "step": 2630,
      "train_runtime": 6899.6483,
      "train_tokens_per_second": 12162.294
    },
    {
      "epoch": 0.1774171828709938,
      "grad_norm": 1.0135195974125502,
      "learning_rate": 9.807233789510106e-06,
      "loss": 0.37296199798583984,
      "num_input_tokens_seen": 84076096,
      "step": 2635,
      "train_runtime": 6913.9177,
      "train_tokens_per_second": 12160.413
    },
    {
      "epoch": 0.17775383786695395,
      "grad_norm": 0.7809124707439674,
      "learning_rate": 9.806506018191754e-06,
      "loss": 0.38485772609710694,
      "num_input_tokens_seen": 84236056,
      "step": 2640,
      "train_runtime": 6926.647,
      "train_tokens_per_second": 12161.159
    },
    {
      "epoch": 0.1780904928629141,
      "grad_norm": 1.0144407983937282,
      "learning_rate": 9.80577690274932e-06,
      "loss": 0.4194784164428711,
      "num_input_tokens_seen": 84399400,
      "step": 2645,
      "train_runtime": 6939.7689,
      "train_tokens_per_second": 12161.702
    },
    {
      "epoch": 0.17842714785887423,
      "grad_norm": 0.7521181520011776,
      "learning_rate": 9.8050464433867e-06,
      "loss": 0.3938760757446289,
      "num_input_tokens_seen": 84559960,
      "step": 2650,
      "train_runtime": 6952.834,
      "train_tokens_per_second": 12161.941
    },
    {
      "epoch": 0.17876380285483437,
      "grad_norm": 0.9689786862195398,
      "learning_rate": 9.804314640308165e-06,
      "loss": 0.44164037704467773,
      "num_input_tokens_seen": 84723800,
      "step": 2655,
      "train_runtime": 6965.2711,
      "train_tokens_per_second": 12163.748
    },
    {
      "epoch": 0.1791004578507945,
      "grad_norm": 0.9156543326978483,
      "learning_rate": 9.803581493718361e-06,
      "loss": 0.4237776756286621,
      "num_input_tokens_seen": 84883616,
      "step": 2660,
      "train_runtime": 6978.1955,
      "train_tokens_per_second": 12164.121
    },
    {
      "epoch": 0.17943711284675465,
      "grad_norm": 0.8221596882586261,
      "learning_rate": 9.80284700382231e-06,
      "loss": 0.4029380798339844,
      "num_input_tokens_seen": 85047456,
      "step": 2665,
      "train_runtime": 6990.6242,
      "train_tokens_per_second": 12165.932
    },
    {
      "epoch": 0.1797737678427148,
      "grad_norm": 0.7769300670443012,
      "learning_rate": 9.802111170825408e-06,
      "loss": 0.42000017166137693,
      "num_input_tokens_seen": 85210840,
      "step": 2670,
      "train_runtime": 7003.6331,
      "train_tokens_per_second": 12166.662
    },
    {
      "epoch": 0.18011042283867493,
      "grad_norm": 0.8866332542495312,
      "learning_rate": 9.80137399493343e-06,
      "loss": 0.4232648849487305,
      "num_input_tokens_seen": 85368704,
      "step": 2675,
      "train_runtime": 7017.1544,
      "train_tokens_per_second": 12165.715
    },
    {
      "epoch": 0.18044707783463507,
      "grad_norm": 0.9409106245388923,
      "learning_rate": 9.800635476352525e-06,
      "loss": 0.38126511573791505,
      "num_input_tokens_seen": 85523832,
      "step": 2680,
      "train_runtime": 7030.4943,
      "train_tokens_per_second": 12164.697
    },
    {
      "epoch": 0.1807837328305952,
      "grad_norm": 1.0123542346764463,
      "learning_rate": 9.799895615289218e-06,
      "loss": 0.3863433837890625,
      "num_input_tokens_seen": 85681544,
      "step": 2685,
      "train_runtime": 7043.2736,
      "train_tokens_per_second": 12165.017
    },
    {
      "epoch": 0.18112038782655535,
      "grad_norm": 1.098626435506201,
      "learning_rate": 9.799154411950407e-06,
      "loss": 0.45793566703796384,
      "num_input_tokens_seen": 85844584,
      "step": 2690,
      "train_runtime": 7056.1259,
      "train_tokens_per_second": 12165.966
    },
    {
      "epoch": 0.18145704282251549,
      "grad_norm": 1.1325905133131922,
      "learning_rate": 9.79841186654337e-06,
      "loss": 0.4152644157409668,
      "num_input_tokens_seen": 86002320,
      "step": 2695,
      "train_runtime": 7069.3229,
      "train_tokens_per_second": 12165.567
    },
    {
      "epoch": 0.18179369781847562,
      "grad_norm": 0.9374756267248823,
      "learning_rate": 9.797667979275752e-06,
      "loss": 0.44652814865112306,
      "num_input_tokens_seen": 86157600,
      "step": 2700,
      "train_runtime": 7082.855,
      "train_tokens_per_second": 12164.247
    },
    {
      "epoch": 0.18213035281443576,
      "grad_norm": 0.8491406094262439,
      "learning_rate": 9.796922750355584e-06,
      "loss": 0.4005905628204346,
      "num_input_tokens_seen": 86311024,
      "step": 2705,
      "train_runtime": 7095.5694,
      "train_tokens_per_second": 12164.073
    },
    {
      "epoch": 0.1824670078103959,
      "grad_norm": 0.8990044402579392,
      "learning_rate": 9.796176179991266e-06,
      "loss": 0.425658130645752,
      "num_input_tokens_seen": 86471544,
      "step": 2710,
      "train_runtime": 7108.7293,
      "train_tokens_per_second": 12164.135
    },
    {
      "epoch": 0.18280366280635604,
      "grad_norm": 0.8891891913189444,
      "learning_rate": 9.795428268391572e-06,
      "loss": 0.403977108001709,
      "num_input_tokens_seen": 86630224,
      "step": 2715,
      "train_runtime": 7122.4635,
      "train_tokens_per_second": 12162.958
    },
    {
      "epoch": 0.18314031780231618,
      "grad_norm": 0.9094741656220893,
      "learning_rate": 9.794679015765655e-06,
      "loss": 0.4087190628051758,
      "num_input_tokens_seen": 86788728,
      "step": 2720,
      "train_runtime": 7135.5668,
      "train_tokens_per_second": 12162.836
    },
    {
      "epoch": 0.18347697279827632,
      "grad_norm": 1.0351580409636236,
      "learning_rate": 9.793928422323041e-06,
      "loss": 0.4177706718444824,
      "num_input_tokens_seen": 86950832,
      "step": 2725,
      "train_runtime": 7148.8797,
      "train_tokens_per_second": 12162.861
    },
    {
      "epoch": 0.18381362779423646,
      "grad_norm": 1.0331667856852635,
      "learning_rate": 9.793176488273629e-06,
      "loss": 0.41898436546325685,
      "num_input_tokens_seen": 87105184,
      "step": 2730,
      "train_runtime": 7161.7664,
      "train_tokens_per_second": 12162.528
    },
    {
      "epoch": 0.1841502827901966,
      "grad_norm": 0.8513843479720603,
      "learning_rate": 9.792423213827697e-06,
      "loss": 0.3745435237884521,
      "num_input_tokens_seen": 87264928,
      "step": 2735,
      "train_runtime": 7174.6237,
      "train_tokens_per_second": 12162.997
    },
    {
      "epoch": 0.18448693778615674,
      "grad_norm": 0.8666803968531231,
      "learning_rate": 9.791668599195897e-06,
      "loss": 0.3980299711227417,
      "num_input_tokens_seen": 87424552,
      "step": 2740,
      "train_runtime": 7187.7765,
      "train_tokens_per_second": 12162.948
    },
    {
      "epoch": 0.18482359278211688,
      "grad_norm": 0.8030918471890202,
      "learning_rate": 9.790912644589248e-06,
      "loss": 0.4387831211090088,
      "num_input_tokens_seen": 87580968,
      "step": 2745,
      "train_runtime": 7201.8702,
      "train_tokens_per_second": 12160.865
    },
    {
      "epoch": 0.18516024777807702,
      "grad_norm": 0.9206122330144707,
      "learning_rate": 9.790155350219159e-06,
      "loss": 0.417657470703125,
      "num_input_tokens_seen": 87739344,
      "step": 2750,
      "train_runtime": 7214.3685,
      "train_tokens_per_second": 12161.75
    },
    {
      "epoch": 0.18549690277403716,
      "grad_norm": 0.8551440661183929,
      "learning_rate": 9.789396716297399e-06,
      "loss": 0.4504976272583008,
      "num_input_tokens_seen": 87902840,
      "step": 2755,
      "train_runtime": 7227.3929,
      "train_tokens_per_second": 12162.455
    },
    {
      "epoch": 0.1858335577699973,
      "grad_norm": 0.9362762822669367,
      "learning_rate": 9.788636743036122e-06,
      "loss": 0.3734339475631714,
      "num_input_tokens_seen": 88060096,
      "step": 2760,
      "train_runtime": 7240.6836,
      "train_tokens_per_second": 12161.848
    },
    {
      "epoch": 0.18617021276595744,
      "grad_norm": 0.8145318910541296,
      "learning_rate": 9.787875430647847e-06,
      "loss": 0.3930145263671875,
      "num_input_tokens_seen": 88222640,
      "step": 2765,
      "train_runtime": 7253.4231,
      "train_tokens_per_second": 12162.897
    },
    {
      "epoch": 0.18650686776191758,
      "grad_norm": 0.7981761472815465,
      "learning_rate": 9.787112779345477e-06,
      "loss": 0.4006065845489502,
      "num_input_tokens_seen": 88379192,
      "step": 2770,
      "train_runtime": 7267.1114,
      "train_tokens_per_second": 12161.53
    },
    {
      "epoch": 0.18684352275787772,
      "grad_norm": 0.9598533194834584,
      "learning_rate": 9.786348789342283e-06,
      "loss": 0.4450218200683594,
      "num_input_tokens_seen": 88536504,
      "step": 2775,
      "train_runtime": 7280.1281,
      "train_tokens_per_second": 12161.394
    },
    {
      "epoch": 0.18718017775383786,
      "grad_norm": 0.8716847862241821,
      "learning_rate": 9.785583460851912e-06,
      "loss": 0.39981999397277834,
      "num_input_tokens_seen": 88697776,
      "step": 2780,
      "train_runtime": 7293.0778,
      "train_tokens_per_second": 12161.913
    },
    {
      "epoch": 0.187516832749798,
      "grad_norm": 0.910674818087515,
      "learning_rate": 9.784816794088386e-06,
      "loss": 0.4131113052368164,
      "num_input_tokens_seen": 88859520,
      "step": 2785,
      "train_runtime": 7305.6749,
      "train_tokens_per_second": 12163.082
    },
    {
      "epoch": 0.18785348774575814,
      "grad_norm": 0.9720515826750926,
      "learning_rate": 9.784048789266101e-06,
      "loss": 0.4105926513671875,
      "num_input_tokens_seen": 89013912,
      "step": 2790,
      "train_runtime": 7318.59,
      "train_tokens_per_second": 12162.713
    },
    {
      "epoch": 0.18819014274171827,
      "grad_norm": 0.8711404511229854,
      "learning_rate": 9.78327944659983e-06,
      "loss": 0.43026247024536135,
      "num_input_tokens_seen": 89172584,
      "step": 2795,
      "train_runtime": 7332.1093,
      "train_tokens_per_second": 12161.928
    },
    {
      "epoch": 0.18852679773767841,
      "grad_norm": 0.8164065363789803,
      "learning_rate": 9.78250876630471e-06,
      "loss": 0.41289725303649905,
      "num_input_tokens_seen": 89326984,
      "step": 2800,
      "train_runtime": 7345.5873,
      "train_tokens_per_second": 12160.632
    },
    {
      "epoch": 0.18886345273363855,
      "grad_norm": 1.0155942984958568,
      "learning_rate": 9.781736748596267e-06,
      "loss": 0.4130856513977051,
      "num_input_tokens_seen": 89482336,
      "step": 2805,
      "train_runtime": 7358.7224,
      "train_tokens_per_second": 12160.037
    },
    {
      "epoch": 0.18920010772959872,
      "grad_norm": 0.8422345644716879,
      "learning_rate": 9.78096339369039e-06,
      "loss": 0.3859609842300415,
      "num_input_tokens_seen": 89638376,
      "step": 2810,
      "train_runtime": 7372.1145,
      "train_tokens_per_second": 12159.113
    },
    {
      "epoch": 0.18953676272555886,
      "grad_norm": 0.8106747269763328,
      "learning_rate": 9.780188701803345e-06,
      "loss": 0.44630231857299807,
      "num_input_tokens_seen": 89802168,
      "step": 2815,
      "train_runtime": 7385.2602,
      "train_tokens_per_second": 12159.649
    },
    {
      "epoch": 0.189873417721519,
      "grad_norm": 0.872900128190699,
      "learning_rate": 9.779412673151772e-06,
      "loss": 0.41963987350463866,
      "num_input_tokens_seen": 89963536,
      "step": 2820,
      "train_runtime": 7397.801,
      "train_tokens_per_second": 12160.848
    },
    {
      "epoch": 0.19021007271747914,
      "grad_norm": 0.8565924751370928,
      "learning_rate": 9.778635307952686e-06,
      "loss": 0.4156802177429199,
      "num_input_tokens_seen": 90122712,
      "step": 2825,
      "train_runtime": 7410.4509,
      "train_tokens_per_second": 12161.569
    },
    {
      "epoch": 0.19054672771343928,
      "grad_norm": 1.0698030465139041,
      "learning_rate": 9.777856606423472e-06,
      "loss": 0.41915321350097656,
      "num_input_tokens_seen": 90286104,
      "step": 2830,
      "train_runtime": 7424.9422,
      "train_tokens_per_second": 12159.839
    },
    {
      "epoch": 0.19088338270939942,
      "grad_norm": 0.8986591027777142,
      "learning_rate": 9.777076568781897e-06,
      "loss": 0.4459648132324219,
      "num_input_tokens_seen": 90440088,
      "step": 2835,
      "train_runtime": 7438.485,
      "train_tokens_per_second": 12158.402
    },
    {
      "epoch": 0.19122003770535956,
      "grad_norm": 0.8342790040470093,
      "learning_rate": 9.77629519524609e-06,
      "loss": 0.37772140502929685,
      "num_input_tokens_seen": 90595824,
      "step": 2840,
      "train_runtime": 7451.5547,
      "train_tokens_per_second": 12157.976
    },
    {
      "epoch": 0.1915566927013197,
      "grad_norm": 0.8408752721654029,
      "learning_rate": 9.775512486034564e-06,
      "loss": 0.4278468132019043,
      "num_input_tokens_seen": 90754960,
      "step": 2845,
      "train_runtime": 7464.817,
      "train_tokens_per_second": 12157.694
    },
    {
      "epoch": 0.19189334769727984,
      "grad_norm": 0.8521883494579162,
      "learning_rate": 9.774728441366197e-06,
      "loss": 0.41966590881347654,
      "num_input_tokens_seen": 90910112,
      "step": 2850,
      "train_runtime": 7478.3399,
      "train_tokens_per_second": 12156.456
    },
    {
      "epoch": 0.19223000269323998,
      "grad_norm": 0.8642989378062204,
      "learning_rate": 9.773943061460249e-06,
      "loss": 0.390500807762146,
      "num_input_tokens_seen": 91067432,
      "step": 2855,
      "train_runtime": 7491.5905,
      "train_tokens_per_second": 12155.954
    },
    {
      "epoch": 0.19256665768920012,
      "grad_norm": 0.894117584925718,
      "learning_rate": 9.773156346536345e-06,
      "loss": 0.43290910720825193,
      "num_input_tokens_seen": 91230056,
      "step": 2860,
      "train_runtime": 7504.9646,
      "train_tokens_per_second": 12155.961
    },
    {
      "epoch": 0.19290331268516026,
      "grad_norm": 0.8518180966932317,
      "learning_rate": 9.772368296814489e-06,
      "loss": 0.38295803070068357,
      "num_input_tokens_seen": 91388624,
      "step": 2865,
      "train_runtime": 7518.0106,
      "train_tokens_per_second": 12155.958
    },
    {
      "epoch": 0.1932399676811204,
      "grad_norm": 0.7887625782981932,
      "learning_rate": 9.771578912515058e-06,
      "loss": 0.45216002464294436,
      "num_input_tokens_seen": 91546448,
      "step": 2870,
      "train_runtime": 7530.8069,
      "train_tokens_per_second": 12156.26
    },
    {
      "epoch": 0.19357662267708053,
      "grad_norm": 0.9211482501729458,
      "learning_rate": 9.7707881938588e-06,
      "loss": 0.4019335746765137,
      "num_input_tokens_seen": 91707400,
      "step": 2875,
      "train_runtime": 7544.5763,
      "train_tokens_per_second": 12155.408
    },
    {
      "epoch": 0.19391327767304067,
      "grad_norm": 0.8826251271691166,
      "learning_rate": 9.769996141066836e-06,
      "loss": 0.4338651657104492,
      "num_input_tokens_seen": 91869920,
      "step": 2880,
      "train_runtime": 7557.0555,
      "train_tokens_per_second": 12156.841
    },
    {
      "epoch": 0.1942499326690008,
      "grad_norm": 0.8697792140617978,
      "learning_rate": 9.76920275436066e-06,
      "loss": 0.4350432395935059,
      "num_input_tokens_seen": 92033560,
      "step": 2885,
      "train_runtime": 7570.127,
      "train_tokens_per_second": 12157.466
    },
    {
      "epoch": 0.19458658766496095,
      "grad_norm": 1.0903094359374617,
      "learning_rate": 9.768408033962145e-06,
      "loss": 0.446491003036499,
      "num_input_tokens_seen": 92195696,
      "step": 2890,
      "train_runtime": 7583.0794,
      "train_tokens_per_second": 12158.081
    },
    {
      "epoch": 0.1949232426609211,
      "grad_norm": 0.9335200253214792,
      "learning_rate": 9.767611980093525e-06,
      "loss": 0.4211887836456299,
      "num_input_tokens_seen": 92358584,
      "step": 2895,
      "train_runtime": 7596.5506,
      "train_tokens_per_second": 12157.963
    },
    {
      "epoch": 0.19525989765688123,
      "grad_norm": 0.8276915732212518,
      "learning_rate": 9.76681459297742e-06,
      "loss": 0.4165052890777588,
      "num_input_tokens_seen": 92518424,
      "step": 2900,
      "train_runtime": 7609.3436,
      "train_tokens_per_second": 12158.529
    },
    {
      "epoch": 0.19559655265284137,
      "grad_norm": 1.01759440098603,
      "learning_rate": 9.766015872836814e-06,
      "loss": 0.47567176818847656,
      "num_input_tokens_seen": 92678896,
      "step": 2905,
      "train_runtime": 7621.8279,
      "train_tokens_per_second": 12159.668
    },
    {
      "epoch": 0.1959332076488015,
      "grad_norm": 1.1625047414651473,
      "learning_rate": 9.765215819895066e-06,
      "loss": 0.43273239135742186,
      "num_input_tokens_seen": 92839112,
      "step": 2910,
      "train_runtime": 7634.3086,
      "train_tokens_per_second": 12160.775
    },
    {
      "epoch": 0.19626986264476165,
      "grad_norm": 0.8158629948489308,
      "learning_rate": 9.76441443437591e-06,
      "loss": 0.393531060218811,
      "num_input_tokens_seen": 93001648,
      "step": 2915,
      "train_runtime": 7647.6992,
      "train_tokens_per_second": 12160.736
    },
    {
      "epoch": 0.1966065176407218,
      "grad_norm": 0.7954184432681562,
      "learning_rate": 9.76361171650345e-06,
      "loss": 0.4203947067260742,
      "num_input_tokens_seen": 93162648,
      "step": 2920,
      "train_runtime": 7660.9004,
      "train_tokens_per_second": 12160.796
    },
    {
      "epoch": 0.19694317263668193,
      "grad_norm": 0.9471817720249008,
      "learning_rate": 9.762807666502164e-06,
      "loss": 0.42482929229736327,
      "num_input_tokens_seen": 93319936,
      "step": 2925,
      "train_runtime": 7673.6178,
      "train_tokens_per_second": 12161.139
    },
    {
      "epoch": 0.19727982763264207,
      "grad_norm": 0.8768205029130177,
      "learning_rate": 9.762002284596898e-06,
      "loss": 0.43389034271240234,
      "num_input_tokens_seen": 93480400,
      "step": 2930,
      "train_runtime": 7687.0679,
      "train_tokens_per_second": 12160.736
    },
    {
      "epoch": 0.1976164826286022,
      "grad_norm": 1.0118364807310998,
      "learning_rate": 9.761195571012882e-06,
      "loss": 0.44169158935546876,
      "num_input_tokens_seen": 93634912,
      "step": 2935,
      "train_runtime": 7700.6411,
      "train_tokens_per_second": 12159.366
    },
    {
      "epoch": 0.19795313762456235,
      "grad_norm": 0.9102576448858622,
      "learning_rate": 9.760387525975705e-06,
      "loss": 0.410272216796875,
      "num_input_tokens_seen": 93796968,
      "step": 2940,
      "train_runtime": 7713.3396,
      "train_tokens_per_second": 12160.358
    },
    {
      "epoch": 0.1982897926205225,
      "grad_norm": 0.7958475773304091,
      "learning_rate": 9.759578149711337e-06,
      "loss": 0.3830035924911499,
      "num_input_tokens_seen": 93960704,
      "step": 2945,
      "train_runtime": 7726.4358,
      "train_tokens_per_second": 12160.938
    },
    {
      "epoch": 0.19862644761648263,
      "grad_norm": 0.8329908107054564,
      "learning_rate": 9.758767442446117e-06,
      "loss": 0.39389288425445557,
      "num_input_tokens_seen": 94123704,
      "step": 2950,
      "train_runtime": 7739.3374,
      "train_tokens_per_second": 12161.726
    },
    {
      "epoch": 0.19896310261244277,
      "grad_norm": 0.8390174649344401,
      "learning_rate": 9.757955404406755e-06,
      "loss": 0.37879266738891604,
      "num_input_tokens_seen": 94286048,
      "step": 2955,
      "train_runtime": 7753.4735,
      "train_tokens_per_second": 12160.491
    },
    {
      "epoch": 0.1992997576084029,
      "grad_norm": 0.8677565205797986,
      "learning_rate": 9.757142035820337e-06,
      "loss": 0.3828702688217163,
      "num_input_tokens_seen": 94437776,
      "step": 2960,
      "train_runtime": 7766.2468,
      "train_tokens_per_second": 12160.028
    },
    {
      "epoch": 0.19963641260436304,
      "grad_norm": 0.927542779772148,
      "learning_rate": 9.756327336914318e-06,
      "loss": 0.4129635810852051,
      "num_input_tokens_seen": 94594824,
      "step": 2965,
      "train_runtime": 7779.0812,
      "train_tokens_per_second": 12160.154
    },
    {
      "epoch": 0.19997306760032318,
      "grad_norm": 1.1419132753671308,
      "learning_rate": 9.755511307916526e-06,
      "loss": 0.4133798599243164,
      "num_input_tokens_seen": 94753928,
      "step": 2970,
      "train_runtime": 7792.1302,
      "train_tokens_per_second": 12160.208
    },
    {
      "epoch": 0.20030972259628332,
      "grad_norm": 0.9776269219331849,
      "learning_rate": 9.754693949055161e-06,
      "loss": 0.43349714279174806,
      "num_input_tokens_seen": 94914776,
      "step": 2975,
      "train_runtime": 7804.5945,
      "train_tokens_per_second": 12161.398
    },
    {
      "epoch": 0.20064637759224346,
      "grad_norm": 0.854838153204509,
      "learning_rate": 9.753875260558795e-06,
      "loss": 0.4501204490661621,
      "num_input_tokens_seen": 95068304,
      "step": 2980,
      "train_runtime": 7818.2941,
      "train_tokens_per_second": 12159.725
    },
    {
      "epoch": 0.2009830325882036,
      "grad_norm": 0.8760234967586558,
      "learning_rate": 9.75305524265637e-06,
      "loss": 0.40970430374145506,
      "num_input_tokens_seen": 95231064,
      "step": 2985,
      "train_runtime": 7831.6655,
      "train_tokens_per_second": 12159.746
    },
    {
      "epoch": 0.20131968758416374,
      "grad_norm": 0.8818880816417082,
      "learning_rate": 9.752233895577205e-06,
      "loss": 0.4172060012817383,
      "num_input_tokens_seen": 95387144,
      "step": 2990,
      "train_runtime": 7844.9517,
      "train_tokens_per_second": 12159.048
    },
    {
      "epoch": 0.20165634258012388,
      "grad_norm": 0.7218494693494788,
      "learning_rate": 9.751411219550984e-06,
      "loss": 0.4078125,
      "num_input_tokens_seen": 95548240,
      "step": 2995,
      "train_runtime": 7858.6602,
      "train_tokens_per_second": 12158.337
    },
    {
      "epoch": 0.20199299757608402,
      "grad_norm": 1.1790781023404124,
      "learning_rate": 9.750587214807766e-06,
      "loss": 0.4112940788269043,
      "num_input_tokens_seen": 95705288,
      "step": 3000,
      "train_runtime": 7872.4553,
      "train_tokens_per_second": 12156.981
    },
    {
      "epoch": 0.20232965257204416,
      "grad_norm": 0.9196973959618487,
      "learning_rate": 9.749761881577981e-06,
      "loss": 0.4491419792175293,
      "num_input_tokens_seen": 95869128,
      "step": 3005,
      "train_runtime": 7884.8991,
      "train_tokens_per_second": 12158.574
    },
    {
      "epoch": 0.2026663075680043,
      "grad_norm": 0.7973597683978836,
      "learning_rate": 9.748935220092434e-06,
      "loss": 0.40469770431518554,
      "num_input_tokens_seen": 96032536,
      "step": 3010,
      "train_runtime": 7897.9103,
      "train_tokens_per_second": 12159.234
    },
    {
      "epoch": 0.20300296256396444,
      "grad_norm": 0.8995241253842347,
      "learning_rate": 9.748107230582295e-06,
      "loss": 0.39678263664245605,
      "num_input_tokens_seen": 96186080,
      "step": 3015,
      "train_runtime": 7910.6393,
      "train_tokens_per_second": 12159.078
    },
    {
      "epoch": 0.20333961755992458,
      "grad_norm": 0.7974175987572382,
      "learning_rate": 9.747277913279109e-06,
      "loss": 0.4466853141784668,
      "num_input_tokens_seen": 96345216,
      "step": 3020,
      "train_runtime": 7923.8113,
      "train_tokens_per_second": 12158.949
    },
    {
      "epoch": 0.20367627255588472,
      "grad_norm": 0.8304329993373757,
      "learning_rate": 9.746447268414794e-06,
      "loss": 0.436503267288208,
      "num_input_tokens_seen": 96506360,
      "step": 3025,
      "train_runtime": 7936.3638,
      "train_tokens_per_second": 12160.022
    },
    {
      "epoch": 0.20401292755184486,
      "grad_norm": 0.9572821429492109,
      "learning_rate": 9.745615296221635e-06,
      "loss": 0.4237049579620361,
      "num_input_tokens_seen": 96668832,
      "step": 3030,
      "train_runtime": 7949.4524,
      "train_tokens_per_second": 12160.439
    },
    {
      "epoch": 0.204349582547805,
      "grad_norm": 0.7972173189225592,
      "learning_rate": 9.74478199693229e-06,
      "loss": 0.4324819087982178,
      "num_input_tokens_seen": 96832480,
      "step": 3035,
      "train_runtime": 7962.5041,
      "train_tokens_per_second": 12161.059
    },
    {
      "epoch": 0.20468623754376514,
      "grad_norm": 1.336232294002001,
      "learning_rate": 9.743947370779791e-06,
      "loss": 0.42787981033325195,
      "num_input_tokens_seen": 96995840,
      "step": 3040,
      "train_runtime": 7975.4913,
      "train_tokens_per_second": 12161.739
    },
    {
      "epoch": 0.20502289253972528,
      "grad_norm": 0.8748869490603304,
      "learning_rate": 9.743111417997536e-06,
      "loss": 0.4295599937438965,
      "num_input_tokens_seen": 97154256,
      "step": 3045,
      "train_runtime": 7988.0299,
      "train_tokens_per_second": 12162.48
    },
    {
      "epoch": 0.20535954753568542,
      "grad_norm": 0.8920763711941498,
      "learning_rate": 9.742274138819298e-06,
      "loss": 0.4323575496673584,
      "num_input_tokens_seen": 97310736,
      "step": 3050,
      "train_runtime": 8001.5648,
      "train_tokens_per_second": 12161.463
    },
    {
      "epoch": 0.20569620253164558,
      "grad_norm": 1.0095917523925655,
      "learning_rate": 9.74143553347922e-06,
      "loss": 0.4573812484741211,
      "num_input_tokens_seen": 97467776,
      "step": 3055,
      "train_runtime": 8014.1969,
      "train_tokens_per_second": 12161.889
    },
    {
      "epoch": 0.20603285752760572,
      "grad_norm": 0.8332008065946794,
      "learning_rate": 9.740595602211814e-06,
      "loss": 0.39818904399871824,
      "num_input_tokens_seen": 97629216,
      "step": 3060,
      "train_runtime": 8028.6216,
      "train_tokens_per_second": 12160.147
    },
    {
      "epoch": 0.20636951252356586,
      "grad_norm": 0.8005408499866398,
      "learning_rate": 9.739754345251963e-06,
      "loss": 0.37448973655700685,
      "num_input_tokens_seen": 97786176,
      "step": 3065,
      "train_runtime": 8041.9188,
      "train_tokens_per_second": 12159.558
    },
    {
      "epoch": 0.206706167519526,
      "grad_norm": 0.7996545807373647,
      "learning_rate": 9.738911762834924e-06,
      "loss": 0.36400160789489744,
      "num_input_tokens_seen": 97948256,
      "step": 3070,
      "train_runtime": 8054.8616,
      "train_tokens_per_second": 12160.141
    },
    {
      "epoch": 0.20704282251548614,
      "grad_norm": 0.9714099397223024,
      "learning_rate": 9.738067855196323e-06,
      "loss": 0.41321558952331544,
      "num_input_tokens_seen": 98108008,
      "step": 3075,
      "train_runtime": 8068.1005,
      "train_tokens_per_second": 12159.988
    },
    {
      "epoch": 0.20737947751144628,
      "grad_norm": 1.369065615445435,
      "learning_rate": 9.737222622572153e-06,
      "loss": 0.4634650230407715,
      "num_input_tokens_seen": 98266064,
      "step": 3080,
      "train_runtime": 8081.9079,
      "train_tokens_per_second": 12158.771
    },
    {
      "epoch": 0.20771613250740642,
      "grad_norm": 1.120909995185053,
      "learning_rate": 9.736376065198784e-06,
      "loss": 0.43927874565124514,
      "num_input_tokens_seen": 98427848,
      "step": 3085,
      "train_runtime": 8095.4609,
      "train_tokens_per_second": 12158.4
    },
    {
      "epoch": 0.20805278750336656,
      "grad_norm": 0.8086704751867378,
      "learning_rate": 9.735528183312951e-06,
      "loss": 0.4097087860107422,
      "num_input_tokens_seen": 98591688,
      "step": 3090,
      "train_runtime": 8107.8895,
      "train_tokens_per_second": 12159.969
    },
    {
      "epoch": 0.2083894424993267,
      "grad_norm": 0.8915868941281733,
      "learning_rate": 9.734678977151762e-06,
      "loss": 0.38752074241638185,
      "num_input_tokens_seen": 98748168,
      "step": 3095,
      "train_runtime": 8120.4895,
      "train_tokens_per_second": 12160.371
    },
    {
      "epoch": 0.20872609749528684,
      "grad_norm": 0.8231356778578908,
      "learning_rate": 9.733828446952696e-06,
      "loss": 0.4035062789916992,
      "num_input_tokens_seen": 98910264,
      "step": 3100,
      "train_runtime": 8133.6192,
      "train_tokens_per_second": 12160.671
    },
    {
      "epoch": 0.20906275249124698,
      "grad_norm": 0.8821657328268957,
      "learning_rate": 9.732976592953598e-06,
      "loss": 0.40680575370788574,
      "num_input_tokens_seen": 99072408,
      "step": 3105,
      "train_runtime": 8146.8998,
      "train_tokens_per_second": 12160.75
    },
    {
      "epoch": 0.20939940748720712,
      "grad_norm": 0.8565146753565397,
      "learning_rate": 9.73212341539269e-06,
      "loss": 0.41437864303588867,
      "num_input_tokens_seen": 99234840,
      "step": 3110,
      "train_runtime": 8159.3684,
      "train_tokens_per_second": 12162.074
    },
    {
      "epoch": 0.20973606248316726,
      "grad_norm": 0.8276772182350852,
      "learning_rate": 9.731268914508556e-06,
      "loss": 0.39754509925842285,
      "num_input_tokens_seen": 99394624,
      "step": 3115,
      "train_runtime": 8172.875,
      "train_tokens_per_second": 12161.525
    },
    {
      "epoch": 0.2100727174791274,
      "grad_norm": 0.900299837752339,
      "learning_rate": 9.730413090540158e-06,
      "loss": 0.38875527381896974,
      "num_input_tokens_seen": 99556136,
      "step": 3120,
      "train_runtime": 8186.0856,
      "train_tokens_per_second": 12161.629
    },
    {
      "epoch": 0.21040937247508754,
      "grad_norm": 0.836001039487523,
      "learning_rate": 9.729555943726825e-06,
      "loss": 0.40328731536865237,
      "num_input_tokens_seen": 99716376,
      "step": 3125,
      "train_runtime": 8198.5925,
      "train_tokens_per_second": 12162.621
    },
    {
      "epoch": 0.21074602747104768,
      "grad_norm": 0.8509910416434696,
      "learning_rate": 9.728697474308253e-06,
      "loss": 0.41985578536987306,
      "num_input_tokens_seen": 99872328,
      "step": 3130,
      "train_runtime": 8211.3525,
      "train_tokens_per_second": 12162.713
    },
    {
      "epoch": 0.21108268246700782,
      "grad_norm": 0.872107986549131,
      "learning_rate": 9.72783768252451e-06,
      "loss": 0.3964792251586914,
      "num_input_tokens_seen": 100029832,
      "step": 3135,
      "train_runtime": 8224.7236,
      "train_tokens_per_second": 12162.09
    },
    {
      "epoch": 0.21141933746296795,
      "grad_norm": 0.9236085383641873,
      "learning_rate": 9.726976568616036e-06,
      "loss": 0.41443648338317873,
      "num_input_tokens_seen": 100186592,
      "step": 3140,
      "train_runtime": 8238.4528,
      "train_tokens_per_second": 12160.85
    },
    {
      "epoch": 0.2117559924589281,
      "grad_norm": 0.9247419871443403,
      "learning_rate": 9.726114132823638e-06,
      "loss": 0.3887040615081787,
      "num_input_tokens_seen": 100346168,
      "step": 3145,
      "train_runtime": 8252.3725,
      "train_tokens_per_second": 12159.675
    },
    {
      "epoch": 0.21209264745488823,
      "grad_norm": 0.8041125503838713,
      "learning_rate": 9.725250375388493e-06,
      "loss": 0.4268604278564453,
      "num_input_tokens_seen": 100505536,
      "step": 3150,
      "train_runtime": 8266.1973,
      "train_tokens_per_second": 12158.618
    },
    {
      "epoch": 0.21242930245084837,
      "grad_norm": 0.8594485198729911,
      "learning_rate": 9.724385296552147e-06,
      "loss": 0.42397661209106446,
      "num_input_tokens_seen": 100666152,
      "step": 3155,
      "train_runtime": 8279.1036,
      "train_tokens_per_second": 12159.064
    },
    {
      "epoch": 0.2127659574468085,
      "grad_norm": 1.0289391436416353,
      "learning_rate": 9.72351889655652e-06,
      "loss": 0.4171248435974121,
      "num_input_tokens_seen": 100824512,
      "step": 3160,
      "train_runtime": 8291.5714,
      "train_tokens_per_second": 12159.88
    },
    {
      "epoch": 0.21310261244276865,
      "grad_norm": 0.8080902580573331,
      "learning_rate": 9.722651175643895e-06,
      "loss": 0.4044322967529297,
      "num_input_tokens_seen": 100984552,
      "step": 3165,
      "train_runtime": 8305.6097,
      "train_tokens_per_second": 12158.596
    },
    {
      "epoch": 0.2134392674387288,
      "grad_norm": 0.9482368453276383,
      "learning_rate": 9.721782134056928e-06,
      "loss": 0.376081657409668,
      "num_input_tokens_seen": 101147192,
      "step": 3170,
      "train_runtime": 8318.3116,
      "train_tokens_per_second": 12159.582
    },
    {
      "epoch": 0.21377592243468893,
      "grad_norm": 0.9050342589181328,
      "learning_rate": 9.720911772038645e-06,
      "loss": 0.4162557125091553,
      "num_input_tokens_seen": 101310512,
      "step": 3175,
      "train_runtime": 8332.0342,
      "train_tokens_per_second": 12159.157
    },
    {
      "epoch": 0.21411257743064907,
      "grad_norm": 0.8660334310465176,
      "learning_rate": 9.720040089832436e-06,
      "loss": 0.423879337310791,
      "num_input_tokens_seen": 101463920,
      "step": 3180,
      "train_runtime": 8346.1319,
      "train_tokens_per_second": 12156.999
    },
    {
      "epoch": 0.2144492324266092,
      "grad_norm": 0.8811187388129222,
      "learning_rate": 9.719167087682069e-06,
      "loss": 0.40487022399902345,
      "num_input_tokens_seen": 101625256,
      "step": 3185,
      "train_runtime": 8359.7941,
      "train_tokens_per_second": 12156.43
    },
    {
      "epoch": 0.21478588742256935,
      "grad_norm": 0.9001390582702702,
      "learning_rate": 9.718292765831673e-06,
      "loss": 0.4114086151123047,
      "num_input_tokens_seen": 101788184,
      "step": 3190,
      "train_runtime": 8372.6283,
      "train_tokens_per_second": 12157.256
    },
    {
      "epoch": 0.2151225424185295,
      "grad_norm": 0.8835423060335061,
      "learning_rate": 9.71741712452575e-06,
      "loss": 0.43125486373901367,
      "num_input_tokens_seen": 101947752,
      "step": 3195,
      "train_runtime": 8385.2485,
      "train_tokens_per_second": 12157.988
    },
    {
      "epoch": 0.21545919741448963,
      "grad_norm": 0.8119730896502064,
      "learning_rate": 9.716540164009174e-06,
      "loss": 0.4259305953979492,
      "num_input_tokens_seen": 102099480,
      "step": 3200,
      "train_runtime": 8398.405,
      "train_tokens_per_second": 12157.008
    },
    {
      "epoch": 0.21579585241044977,
      "grad_norm": 0.9469764098613102,
      "learning_rate": 9.715661884527178e-06,
      "loss": 0.41490988731384276,
      "num_input_tokens_seen": 102261192,
      "step": 3205,
      "train_runtime": 8411.6318,
      "train_tokens_per_second": 12157.117
    },
    {
      "epoch": 0.2161325074064099,
      "grad_norm": 0.9315977277195763,
      "learning_rate": 9.714782286325374e-06,
      "loss": 0.4179119110107422,
      "num_input_tokens_seen": 102421752,
      "step": 3210,
      "train_runtime": 8424.4462,
      "train_tokens_per_second": 12157.684
    },
    {
      "epoch": 0.21646916240237005,
      "grad_norm": 0.924460885738849,
      "learning_rate": 9.713901369649738e-06,
      "loss": 0.41758270263671876,
      "num_input_tokens_seen": 102583040,
      "step": 3215,
      "train_runtime": 8437.5349,
      "train_tokens_per_second": 12157.94
    },
    {
      "epoch": 0.2168058173983302,
      "grad_norm": 0.8398704839344228,
      "learning_rate": 9.713019134746618e-06,
      "loss": 0.3930691719055176,
      "num_input_tokens_seen": 102746872,
      "step": 3220,
      "train_runtime": 8450.6951,
      "train_tokens_per_second": 12158.393
    },
    {
      "epoch": 0.21714247239429033,
      "grad_norm": 0.8247134694776286,
      "learning_rate": 9.712135581862722e-06,
      "loss": 0.42127256393432616,
      "num_input_tokens_seen": 102906336,
      "step": 3225,
      "train_runtime": 8463.1779,
      "train_tokens_per_second": 12159.302
    },
    {
      "epoch": 0.21747912739025047,
      "grad_norm": 0.8659094086556777,
      "learning_rate": 9.711250711245139e-06,
      "loss": 0.4215373516082764,
      "num_input_tokens_seen": 103067224,
      "step": 3230,
      "train_runtime": 8476.7753,
      "train_tokens_per_second": 12158.777
    },
    {
      "epoch": 0.2178157823862106,
      "grad_norm": 0.7891654542020214,
      "learning_rate": 9.710364523141317e-06,
      "loss": 0.41613125801086426,
      "num_input_tokens_seen": 103230792,
      "step": 3235,
      "train_runtime": 8489.868,
      "train_tokens_per_second": 12159.293
    },
    {
      "epoch": 0.21815243738217074,
      "grad_norm": 0.8540677372377209,
      "learning_rate": 9.709477017799076e-06,
      "loss": 0.44175148010253906,
      "num_input_tokens_seen": 103386888,
      "step": 3240,
      "train_runtime": 8502.9321,
      "train_tokens_per_second": 12158.969
    },
    {
      "epoch": 0.21848909237813088,
      "grad_norm": 0.9257612373175095,
      "learning_rate": 9.708588195466604e-06,
      "loss": 0.4187493324279785,
      "num_input_tokens_seen": 103546144,
      "step": 3245,
      "train_runtime": 8516.97,
      "train_tokens_per_second": 12157.627
    },
    {
      "epoch": 0.21882574737409102,
      "grad_norm": 0.8196498692945057,
      "learning_rate": 9.707698056392458e-06,
      "loss": 0.40938844680786135,
      "num_input_tokens_seen": 103704088,
      "step": 3250,
      "train_runtime": 8530.1842,
      "train_tokens_per_second": 12157.309
    },
    {
      "epoch": 0.21916240237005116,
      "grad_norm": 0.8348783569427588,
      "learning_rate": 9.706806600825562e-06,
      "loss": 0.4203173637390137,
      "num_input_tokens_seen": 103863264,
      "step": 3255,
      "train_runtime": 8543.1459,
      "train_tokens_per_second": 12157.496
    },
    {
      "epoch": 0.2194990573660113,
      "grad_norm": 0.9924397888990996,
      "learning_rate": 9.70591382901521e-06,
      "loss": 0.43382821083068845,
      "num_input_tokens_seen": 104021368,
      "step": 3260,
      "train_runtime": 8555.9376,
      "train_tokens_per_second": 12157.799
    },
    {
      "epoch": 0.21983571236197144,
      "grad_norm": 0.878732565925723,
      "learning_rate": 9.705019741211061e-06,
      "loss": 0.3969125747680664,
      "num_input_tokens_seen": 104180800,
      "step": 3265,
      "train_runtime": 8568.6206,
      "train_tokens_per_second": 12158.41
    },
    {
      "epoch": 0.22017236735793158,
      "grad_norm": 0.8715059083657861,
      "learning_rate": 9.704124337663145e-06,
      "loss": 0.40511722564697267,
      "num_input_tokens_seen": 104342824,
      "step": 3270,
      "train_runtime": 8581.2586,
      "train_tokens_per_second": 12159.385
    },
    {
      "epoch": 0.22050902235389172,
      "grad_norm": 0.9524389427909125,
      "learning_rate": 9.703227618621859e-06,
      "loss": 0.411458683013916,
      "num_input_tokens_seen": 104500776,
      "step": 3275,
      "train_runtime": 8594.1626,
      "train_tokens_per_second": 12159.507
    },
    {
      "epoch": 0.22084567734985186,
      "grad_norm": 0.9009508149880984,
      "learning_rate": 9.702329584337965e-06,
      "loss": 0.42221755981445314,
      "num_input_tokens_seen": 104653048,
      "step": 3280,
      "train_runtime": 8607.6314,
      "train_tokens_per_second": 12158.17
    },
    {
      "epoch": 0.221182332345812,
      "grad_norm": 0.8993845512420526,
      "learning_rate": 9.701430235062599e-06,
      "loss": 0.41233162879943847,
      "num_input_tokens_seen": 104811048,
      "step": 3285,
      "train_runtime": 8621.0373,
      "train_tokens_per_second": 12157.591
    },
    {
      "epoch": 0.22151898734177214,
      "grad_norm": 0.9593772464522136,
      "learning_rate": 9.700529571047259e-06,
      "loss": 0.4482554912567139,
      "num_input_tokens_seen": 104969448,
      "step": 3290,
      "train_runtime": 8634.6689,
      "train_tokens_per_second": 12156.743
    },
    {
      "epoch": 0.22185564233773228,
      "grad_norm": 0.9663087812516342,
      "learning_rate": 9.699627592543814e-06,
      "loss": 0.4106715202331543,
      "num_input_tokens_seen": 105130880,
      "step": 3295,
      "train_runtime": 8648.4886,
      "train_tokens_per_second": 12155.983
    },
    {
      "epoch": 0.22219229733369245,
      "grad_norm": 0.8710129534870628,
      "learning_rate": 9.698724299804498e-06,
      "loss": 0.43678598403930663,
      "num_input_tokens_seen": 105293160,
      "step": 3300,
      "train_runtime": 8661.5605,
      "train_tokens_per_second": 12156.373
    },
    {
      "epoch": 0.22252895232965259,
      "grad_norm": 0.8766988326847366,
      "learning_rate": 9.697819693081915e-06,
      "loss": 0.4082034111022949,
      "num_input_tokens_seen": 105453872,
      "step": 3305,
      "train_runtime": 8674.1095,
      "train_tokens_per_second": 12157.314
    },
    {
      "epoch": 0.22286560732561272,
      "grad_norm": 0.8297196286245672,
      "learning_rate": 9.696913772629034e-06,
      "loss": 0.381276535987854,
      "num_input_tokens_seen": 105613656,
      "step": 3310,
      "train_runtime": 8686.8857,
      "train_tokens_per_second": 12157.827
    },
    {
      "epoch": 0.22320226232157286,
      "grad_norm": 0.8248529690674457,
      "learning_rate": 9.696006538699195e-06,
      "loss": 0.4185637474060059,
      "num_input_tokens_seen": 105771304,
      "step": 3315,
      "train_runtime": 8699.7702,
      "train_tokens_per_second": 12157.942
    },
    {
      "epoch": 0.223538917317533,
      "grad_norm": 0.9144822279409108,
      "learning_rate": 9.695097991546102e-06,
      "loss": 0.4286025047302246,
      "num_input_tokens_seen": 105924784,
      "step": 3320,
      "train_runtime": 8713.7221,
      "train_tokens_per_second": 12156.089
    },
    {
      "epoch": 0.22387557231349314,
      "grad_norm": 0.8582715250020818,
      "learning_rate": 9.694188131423828e-06,
      "loss": 0.3919801712036133,
      "num_input_tokens_seen": 106081192,
      "step": 3325,
      "train_runtime": 8726.4794,
      "train_tokens_per_second": 12156.242
    },
    {
      "epoch": 0.22421222730945328,
      "grad_norm": 0.8454697290206538,
      "learning_rate": 9.693276958586809e-06,
      "loss": 0.4598099708557129,
      "num_input_tokens_seen": 106243008,
      "step": 3330,
      "train_runtime": 8738.9344,
      "train_tokens_per_second": 12157.433
    },
    {
      "epoch": 0.22454888230541342,
      "grad_norm": 0.7607502024337314,
      "learning_rate": 9.692364473289856e-06,
      "loss": 0.389451789855957,
      "num_input_tokens_seen": 106399472,
      "step": 3335,
      "train_runtime": 8751.5885,
      "train_tokens_per_second": 12157.732
    },
    {
      "epoch": 0.22488553730137356,
      "grad_norm": 0.9167999336481737,
      "learning_rate": 9.69145067578814e-06,
      "loss": 0.37474632263183594,
      "num_input_tokens_seen": 106558728,
      "step": 3340,
      "train_runtime": 8764.4007,
      "train_tokens_per_second": 12158.131
    },
    {
      "epoch": 0.2252221922973337,
      "grad_norm": 0.9761161865850733,
      "learning_rate": 9.690535566337205e-06,
      "loss": 0.45883026123046877,
      "num_input_tokens_seen": 106717872,
      "step": 3345,
      "train_runtime": 8777.7964,
      "train_tokens_per_second": 12157.706
    },
    {
      "epoch": 0.22555884729329384,
      "grad_norm": 0.9643433582960327,
      "learning_rate": 9.689619145192955e-06,
      "loss": 0.4292914390563965,
      "num_input_tokens_seen": 106876840,
      "step": 3350,
      "train_runtime": 8790.4212,
      "train_tokens_per_second": 12158.33
    },
    {
      "epoch": 0.22589550228925398,
      "grad_norm": 0.8341523166429599,
      "learning_rate": 9.688701412611663e-06,
      "loss": 0.4082505226135254,
      "num_input_tokens_seen": 107036800,
      "step": 3355,
      "train_runtime": 8802.9062,
      "train_tokens_per_second": 12159.257
    },
    {
      "epoch": 0.22623215728521412,
      "grad_norm": 0.8091868277401202,
      "learning_rate": 9.687782368849973e-06,
      "loss": 0.4042107582092285,
      "num_input_tokens_seen": 107198072,
      "step": 3360,
      "train_runtime": 8815.4307,
      "train_tokens_per_second": 12160.276
    },
    {
      "epoch": 0.22656881228117426,
      "grad_norm": 0.8929999519853563,
      "learning_rate": 9.686862014164893e-06,
      "loss": 0.39135012626647947,
      "num_input_tokens_seen": 107358656,
      "step": 3365,
      "train_runtime": 8827.8963,
      "train_tokens_per_second": 12161.295
    },
    {
      "epoch": 0.2269054672771344,
      "grad_norm": 0.8479553361847132,
      "learning_rate": 9.685940348813795e-06,
      "loss": 0.3882725954055786,
      "num_input_tokens_seen": 107521904,
      "step": 3370,
      "train_runtime": 8840.8074,
      "train_tokens_per_second": 12162.0
    },
    {
      "epoch": 0.22724212227309454,
      "grad_norm": 0.8710866407583131,
      "learning_rate": 9.685017373054421e-06,
      "loss": 0.40677661895751954,
      "num_input_tokens_seen": 107682984,
      "step": 3375,
      "train_runtime": 8853.5816,
      "train_tokens_per_second": 12162.647
    },
    {
      "epoch": 0.22757877726905468,
      "grad_norm": 0.8084350149654155,
      "learning_rate": 9.68409308714488e-06,
      "loss": 0.40839109420776365,
      "num_input_tokens_seen": 107841976,
      "step": 3380,
      "train_runtime": 8866.6224,
      "train_tokens_per_second": 12162.69
    },
    {
      "epoch": 0.22791543226501482,
      "grad_norm": 0.8419818745037261,
      "learning_rate": 9.68316749134364e-06,
      "loss": 0.4161202430725098,
      "num_input_tokens_seen": 108005816,
      "step": 3385,
      "train_runtime": 8879.0525,
      "train_tokens_per_second": 12164.115
    },
    {
      "epoch": 0.22825208726097496,
      "grad_norm": 0.795688788607716,
      "learning_rate": 9.682240585909545e-06,
      "loss": 0.3866273880004883,
      "num_input_tokens_seen": 108163880,
      "step": 3390,
      "train_runtime": 8892.2418,
      "train_tokens_per_second": 12163.848
    },
    {
      "epoch": 0.2285887422569351,
      "grad_norm": 0.9031481040507628,
      "learning_rate": 9.681312371101803e-06,
      "loss": 0.37649009227752683,
      "num_input_tokens_seen": 108326320,
      "step": 3395,
      "train_runtime": 8904.9215,
      "train_tokens_per_second": 12164.77
    },
    {
      "epoch": 0.22892539725289524,
      "grad_norm": 0.8295586705692858,
      "learning_rate": 9.680382847179984e-06,
      "loss": 0.4147799491882324,
      "num_input_tokens_seen": 108487936,
      "step": 3400,
      "train_runtime": 8917.7876,
      "train_tokens_per_second": 12165.342
    },
    {
      "epoch": 0.22926205224885537,
      "grad_norm": 0.9086751151899561,
      "learning_rate": 9.679452014404025e-06,
      "loss": 0.42338104248046876,
      "num_input_tokens_seen": 108646768,
      "step": 3405,
      "train_runtime": 8932.0181,
      "train_tokens_per_second": 12163.742
    },
    {
      "epoch": 0.22959870724481551,
      "grad_norm": 1.2669512598145285,
      "learning_rate": 9.678519873034234e-06,
      "loss": 0.44754681587219236,
      "num_input_tokens_seen": 108809224,
      "step": 3410,
      "train_runtime": 8944.4917,
      "train_tokens_per_second": 12164.942
    },
    {
      "epoch": 0.22993536224077565,
      "grad_norm": 1.0160062497283133,
      "learning_rate": 9.677586423331277e-06,
      "loss": 0.42498102188110354,
      "num_input_tokens_seen": 108969712,
      "step": 3415,
      "train_runtime": 8957.2939,
      "train_tokens_per_second": 12165.472
    },
    {
      "epoch": 0.2302720172367358,
      "grad_norm": 0.7540799760126341,
      "learning_rate": 9.676651665556194e-06,
      "loss": 0.4081888198852539,
      "num_input_tokens_seen": 109133552,
      "step": 3420,
      "train_runtime": 8969.7437,
      "train_tokens_per_second": 12166.853
    },
    {
      "epoch": 0.23060867223269593,
      "grad_norm": 0.9202476266256654,
      "learning_rate": 9.675715599970383e-06,
      "loss": 0.4067695617675781,
      "num_input_tokens_seen": 109291328,
      "step": 3425,
      "train_runtime": 8982.6477,
      "train_tokens_per_second": 12166.939
    },
    {
      "epoch": 0.23094532722865607,
      "grad_norm": 0.9847071425682539,
      "learning_rate": 9.674778226835617e-06,
      "loss": 0.44423513412475585,
      "num_input_tokens_seen": 109444560,
      "step": 3430,
      "train_runtime": 8995.8823,
      "train_tokens_per_second": 12166.073
    },
    {
      "epoch": 0.2312819822246162,
      "grad_norm": 0.8727331062700421,
      "learning_rate": 9.673839546414026e-06,
      "loss": 0.4344576358795166,
      "num_input_tokens_seen": 109599736,
      "step": 3435,
      "train_runtime": 9009.2288,
      "train_tokens_per_second": 12165.274
    },
    {
      "epoch": 0.23161863722057635,
      "grad_norm": 0.8463609953608888,
      "learning_rate": 9.672899558968107e-06,
      "loss": 0.3959089994430542,
      "num_input_tokens_seen": 109757400,
      "step": 3440,
      "train_runtime": 9022.9432,
      "train_tokens_per_second": 12164.257
    },
    {
      "epoch": 0.2319552922165365,
      "grad_norm": 0.8233020078002306,
      "learning_rate": 9.67195826476073e-06,
      "loss": 0.4379754543304443,
      "num_input_tokens_seen": 109919176,
      "step": 3445,
      "train_runtime": 9035.4825,
      "train_tokens_per_second": 12165.28
    },
    {
      "epoch": 0.23229194721249663,
      "grad_norm": 0.8949658507352379,
      "learning_rate": 9.671015664055122e-06,
      "loss": 0.4128607749938965,
      "num_input_tokens_seen": 110079376,
      "step": 3450,
      "train_runtime": 9048.5881,
      "train_tokens_per_second": 12165.365
    },
    {
      "epoch": 0.23262860220845677,
      "grad_norm": 0.8541695601708745,
      "learning_rate": 9.670071757114877e-06,
      "loss": 0.4062610626220703,
      "num_input_tokens_seen": 110240984,
      "step": 3455,
      "train_runtime": 9061.8142,
      "train_tokens_per_second": 12165.443
    },
    {
      "epoch": 0.2329652572044169,
      "grad_norm": 0.8278278168855996,
      "learning_rate": 9.669126544203957e-06,
      "loss": 0.41463732719421387,
      "num_input_tokens_seen": 110391984,
      "step": 3460,
      "train_runtime": 9074.7432,
      "train_tokens_per_second": 12164.75
    },
    {
      "epoch": 0.23330191220037705,
      "grad_norm": 0.8782609857711834,
      "learning_rate": 9.668180025586689e-06,
      "loss": 0.41456012725830077,
      "num_input_tokens_seen": 110553664,
      "step": 3465,
      "train_runtime": 9087.3052,
      "train_tokens_per_second": 12165.726
    },
    {
      "epoch": 0.2336385671963372,
      "grad_norm": 0.7805598450692157,
      "learning_rate": 9.667232201527761e-06,
      "loss": 0.39628050327301023,
      "num_input_tokens_seen": 110706472,
      "step": 3470,
      "train_runtime": 9100.5008,
      "train_tokens_per_second": 12164.877
    },
    {
      "epoch": 0.23397522219229733,
      "grad_norm": 0.8813139375760343,
      "learning_rate": 9.666283072292234e-06,
      "loss": 0.4283869743347168,
      "num_input_tokens_seen": 110863224,
      "step": 3475,
      "train_runtime": 9113.994,
      "train_tokens_per_second": 12164.066
    },
    {
      "epoch": 0.23431187718825747,
      "grad_norm": 0.8232365244583492,
      "learning_rate": 9.665332638145522e-06,
      "loss": 0.4345715999603271,
      "num_input_tokens_seen": 111022704,
      "step": 3480,
      "train_runtime": 9127.1895,
      "train_tokens_per_second": 12163.953
    },
    {
      "epoch": 0.2346485321842176,
      "grad_norm": 0.8941185025264897,
      "learning_rate": 9.664380899353414e-06,
      "loss": 0.43789987564086913,
      "num_input_tokens_seen": 111184200,
      "step": 3485,
      "train_runtime": 9141.2982,
      "train_tokens_per_second": 12162.846
    },
    {
      "epoch": 0.23498518718017775,
      "grad_norm": 0.9214894480781967,
      "learning_rate": 9.663427856182062e-06,
      "loss": 0.416960334777832,
      "num_input_tokens_seen": 111348040,
      "step": 3490,
      "train_runtime": 9153.7253,
      "train_tokens_per_second": 12164.232
    },
    {
      "epoch": 0.23532184217613789,
      "grad_norm": 0.8164186286997915,
      "learning_rate": 9.662473508897981e-06,
      "loss": 0.4084165573120117,
      "num_input_tokens_seen": 111509616,
      "step": 3495,
      "train_runtime": 9166.2758,
      "train_tokens_per_second": 12165.204
    },
    {
      "epoch": 0.23565849717209802,
      "grad_norm": 0.9390022953122675,
      "learning_rate": 9.661517857768052e-06,
      "loss": 0.4370316982269287,
      "num_input_tokens_seen": 111667608,
      "step": 3500,
      "train_runtime": 9180.1453,
      "train_tokens_per_second": 12164.035
    },
    {
      "epoch": 0.23599515216805816,
      "grad_norm": 0.8984087082769315,
      "learning_rate": 9.660560903059517e-06,
      "loss": 0.4216012001037598,
      "num_input_tokens_seen": 111823168,
      "step": 3505,
      "train_runtime": 9192.9066,
      "train_tokens_per_second": 12164.071
    },
    {
      "epoch": 0.2363318071640183,
      "grad_norm": 0.7621452436957418,
      "learning_rate": 9.659602645039988e-06,
      "loss": 0.437092924118042,
      "num_input_tokens_seen": 111985056,
      "step": 3510,
      "train_runtime": 9205.5066,
      "train_tokens_per_second": 12165.007
    },
    {
      "epoch": 0.23666846215997844,
      "grad_norm": 0.7702986013180412,
      "learning_rate": 9.658643083977435e-06,
      "loss": 0.3486215591430664,
      "num_input_tokens_seen": 112142272,
      "step": 3515,
      "train_runtime": 9218.5073,
      "train_tokens_per_second": 12164.906
    },
    {
      "epoch": 0.23700511715593858,
      "grad_norm": 0.8892495321950995,
      "learning_rate": 9.657682220140202e-06,
      "loss": 0.38922462463378904,
      "num_input_tokens_seen": 112297832,
      "step": 3520,
      "train_runtime": 9231.7055,
      "train_tokens_per_second": 12164.365
    },
    {
      "epoch": 0.23734177215189872,
      "grad_norm": 1.0978691917136367,
      "learning_rate": 9.656720053796987e-06,
      "loss": 0.42437143325805665,
      "num_input_tokens_seen": 112456296,
      "step": 3525,
      "train_runtime": 9245.836,
      "train_tokens_per_second": 12162.913
    },
    {
      "epoch": 0.23767842714785886,
      "grad_norm": 1.0892584214851264,
      "learning_rate": 9.655756585216859e-06,
      "loss": 0.3968771457672119,
      "num_input_tokens_seen": 112614048,
      "step": 3530,
      "train_runtime": 9258.5112,
      "train_tokens_per_second": 12163.3
    },
    {
      "epoch": 0.238015082143819,
      "grad_norm": 0.863200664259044,
      "learning_rate": 9.654791814669246e-06,
      "loss": 0.43172311782836914,
      "num_input_tokens_seen": 112773928,
      "step": 3535,
      "train_runtime": 9271.8075,
      "train_tokens_per_second": 12163.101
    },
    {
      "epoch": 0.23835173713977914,
      "grad_norm": 0.7943682018371147,
      "learning_rate": 9.653825742423949e-06,
      "loss": 0.38676478862762453,
      "num_input_tokens_seen": 112932416,
      "step": 3540,
      "train_runtime": 9284.418,
      "train_tokens_per_second": 12163.651
    },
    {
      "epoch": 0.2386883921357393,
      "grad_norm": 0.9045352823000757,
      "learning_rate": 9.652858368751118e-06,
      "loss": 0.39141116142272947,
      "num_input_tokens_seen": 113087808,
      "step": 3545,
      "train_runtime": 9298.3679,
      "train_tokens_per_second": 12162.114
    },
    {
      "epoch": 0.23902504713169945,
      "grad_norm": 3.479202922125915,
      "learning_rate": 9.651889693921284e-06,
      "loss": 0.40999598503112794,
      "num_input_tokens_seen": 113250736,
      "step": 3550,
      "train_runtime": 9311.1924,
      "train_tokens_per_second": 12162.861
    },
    {
      "epoch": 0.2393617021276596,
      "grad_norm": 0.8677941344170815,
      "learning_rate": 9.650919718205331e-06,
      "loss": 0.4603950023651123,
      "num_input_tokens_seen": 113410528,
      "step": 3555,
      "train_runtime": 9324.6788,
      "train_tokens_per_second": 12162.406
    },
    {
      "epoch": 0.23969835712361973,
      "grad_norm": 0.7941907491255483,
      "learning_rate": 9.649948441874509e-06,
      "loss": 0.4159419536590576,
      "num_input_tokens_seen": 113571624,
      "step": 3560,
      "train_runtime": 9337.8768,
      "train_tokens_per_second": 12162.468
    },
    {
      "epoch": 0.24003501211957987,
      "grad_norm": 0.7973463112110178,
      "learning_rate": 9.648975865200434e-06,
      "loss": 0.38080658912658694,
      "num_input_tokens_seen": 113728184,
      "step": 3565,
      "train_runtime": 9350.8519,
      "train_tokens_per_second": 12162.334
    },
    {
      "epoch": 0.24037166711554,
      "grad_norm": 0.7793641951819085,
      "learning_rate": 9.648001988455082e-06,
      "loss": 0.4260855674743652,
      "num_input_tokens_seen": 113890688,
      "step": 3570,
      "train_runtime": 9363.799,
      "train_tokens_per_second": 12162.872
    },
    {
      "epoch": 0.24070832211150014,
      "grad_norm": 0.8236218271067685,
      "learning_rate": 9.647026811910794e-06,
      "loss": 0.41405014991760253,
      "num_input_tokens_seen": 114046720,
      "step": 3575,
      "train_runtime": 9377.1067,
      "train_tokens_per_second": 12162.25
    },
    {
      "epoch": 0.24104497710746028,
      "grad_norm": 0.8099877926881521,
      "learning_rate": 9.646050335840279e-06,
      "loss": 0.45738959312438965,
      "num_input_tokens_seen": 114209520,
      "step": 3580,
      "train_runtime": 9389.9039,
      "train_tokens_per_second": 12163.013
    },
    {
      "epoch": 0.24138163210342042,
      "grad_norm": 0.9054846944082656,
      "learning_rate": 9.6450725605166e-06,
      "loss": 0.4009879112243652,
      "num_input_tokens_seen": 114365000,
      "step": 3585,
      "train_runtime": 9402.8509,
      "train_tokens_per_second": 12162.801
    },
    {
      "epoch": 0.24171828709938056,
      "grad_norm": 0.8331422411123904,
      "learning_rate": 9.644093486213192e-06,
      "loss": 0.43718719482421875,
      "num_input_tokens_seen": 114527896,
      "step": 3590,
      "train_runtime": 9415.9905,
      "train_tokens_per_second": 12163.128
    },
    {
      "epoch": 0.2420549420953407,
      "grad_norm": 1.0163723356338525,
      "learning_rate": 9.64311311320385e-06,
      "loss": 0.4103853225708008,
      "num_input_tokens_seen": 114686832,
      "step": 3595,
      "train_runtime": 9429.7622,
      "train_tokens_per_second": 12162.219
    },
    {
      "epoch": 0.24239159709130084,
      "grad_norm": 0.8185094841374555,
      "learning_rate": 9.642131441762732e-06,
      "loss": 0.3965426445007324,
      "num_input_tokens_seen": 114842480,
      "step": 3600,
      "train_runtime": 9444.0371,
      "train_tokens_per_second": 12160.317
    },
    {
      "epoch": 0.24272825208726098,
      "grad_norm": 0.8976878644890596,
      "learning_rate": 9.641148472164359e-06,
      "loss": 0.3918168067932129,
      "num_input_tokens_seen": 115000304,
      "step": 3605,
      "train_runtime": 9457.5047,
      "train_tokens_per_second": 12159.688
    },
    {
      "epoch": 0.24306490708322112,
      "grad_norm": 0.9339461558763776,
      "learning_rate": 9.640164204683616e-06,
      "loss": 0.4022397994995117,
      "num_input_tokens_seen": 115160808,
      "step": 3610,
      "train_runtime": 9470.3104,
      "train_tokens_per_second": 12160.194
    },
    {
      "epoch": 0.24340156207918126,
      "grad_norm": 0.9776924879153237,
      "learning_rate": 9.639178639595748e-06,
      "loss": 0.4423379898071289,
      "num_input_tokens_seen": 115318208,
      "step": 3615,
      "train_runtime": 9483.0396,
      "train_tokens_per_second": 12160.469
    },
    {
      "epoch": 0.2437382170751414,
      "grad_norm": 0.7941532684789867,
      "learning_rate": 9.638191777176367e-06,
      "loss": 0.3806135892868042,
      "num_input_tokens_seen": 115477480,
      "step": 3620,
      "train_runtime": 9496.5642,
      "train_tokens_per_second": 12159.922
    },
    {
      "epoch": 0.24407487207110154,
      "grad_norm": 0.9379879505500848,
      "learning_rate": 9.637203617701446e-06,
      "loss": 0.4460176467895508,
      "num_input_tokens_seen": 115638088,
      "step": 3625,
      "train_runtime": 9509.6313,
      "train_tokens_per_second": 12160.102
    },
    {
      "epoch": 0.24441152706706168,
      "grad_norm": 0.8337398676521538,
      "learning_rate": 9.63621416144732e-06,
      "loss": 0.4018817901611328,
      "num_input_tokens_seen": 115796992,
      "step": 3630,
      "train_runtime": 9522.715,
      "train_tokens_per_second": 12160.082
    },
    {
      "epoch": 0.24474818206302182,
      "grad_norm": 0.9003889527549371,
      "learning_rate": 9.635223408690688e-06,
      "loss": 0.3894858121871948,
      "num_input_tokens_seen": 115952400,
      "step": 3635,
      "train_runtime": 9536.5119,
      "train_tokens_per_second": 12158.785
    },
    {
      "epoch": 0.24508483705898196,
      "grad_norm": 0.8842498925456987,
      "learning_rate": 9.63423135970861e-06,
      "loss": 0.432171630859375,
      "num_input_tokens_seen": 116112808,
      "step": 3640,
      "train_runtime": 9548.9899,
      "train_tokens_per_second": 12159.695
    },
    {
      "epoch": 0.2454214920549421,
      "grad_norm": 1.3307025961396473,
      "learning_rate": 9.633238014778511e-06,
      "loss": 0.43020148277282716,
      "num_input_tokens_seen": 116274056,
      "step": 3645,
      "train_runtime": 9561.7538,
      "train_tokens_per_second": 12160.327
    },
    {
      "epoch": 0.24575814705090224,
      "grad_norm": 0.9547235782138878,
      "learning_rate": 9.632243374178173e-06,
      "loss": 0.42804861068725586,
      "num_input_tokens_seen": 116431600,
      "step": 3650,
      "train_runtime": 9574.2286,
      "train_tokens_per_second": 12160.938
    },
    {
      "epoch": 0.24609480204686238,
      "grad_norm": 0.8714338720390338,
      "learning_rate": 9.63124743818575e-06,
      "loss": 0.4338388442993164,
      "num_input_tokens_seen": 116594064,
      "step": 3655,
      "train_runtime": 9587.2781,
      "train_tokens_per_second": 12161.331
    },
    {
      "epoch": 0.24643145704282252,
      "grad_norm": 0.8184366745128226,
      "learning_rate": 9.630250207079745e-06,
      "loss": 0.3941220998764038,
      "num_input_tokens_seen": 116751224,
      "step": 3660,
      "train_runtime": 9599.7534,
      "train_tokens_per_second": 12161.898
    },
    {
      "epoch": 0.24676811203878266,
      "grad_norm": 0.765148639704,
      "learning_rate": 9.629251681139038e-06,
      "loss": 0.42835197448730467,
      "num_input_tokens_seen": 116913096,
      "step": 3665,
      "train_runtime": 9612.304,
      "train_tokens_per_second": 12162.859
    },
    {
      "epoch": 0.2471047670347428,
      "grad_norm": 1.1360850339361914,
      "learning_rate": 9.628251860642857e-06,
      "loss": 0.4193881988525391,
      "num_input_tokens_seen": 117073408,
      "step": 3670,
      "train_runtime": 9624.9445,
      "train_tokens_per_second": 12163.541
    },
    {
      "epoch": 0.24744142203070293,
      "grad_norm": 0.7877207003672969,
      "learning_rate": 9.627250745870805e-06,
      "loss": 0.41612563133239744,
      "num_input_tokens_seen": 117230136,
      "step": 3675,
      "train_runtime": 9637.6207,
      "train_tokens_per_second": 12163.805
    },
    {
      "epoch": 0.24777807702666307,
      "grad_norm": 1.0769249370912934,
      "learning_rate": 9.626248337102834e-06,
      "loss": 0.452846097946167,
      "num_input_tokens_seen": 117389368,
      "step": 3680,
      "train_runtime": 9650.0845,
      "train_tokens_per_second": 12164.595
    },
    {
      "epoch": 0.2481147320226232,
      "grad_norm": 0.8804891180806412,
      "learning_rate": 9.625244634619269e-06,
      "loss": 0.4418637275695801,
      "num_input_tokens_seen": 117546008,
      "step": 3685,
      "train_runtime": 9662.8278,
      "train_tokens_per_second": 12164.763
    },
    {
      "epoch": 0.24845138701858335,
      "grad_norm": 0.8784089897442168,
      "learning_rate": 9.62423963870079e-06,
      "loss": 0.4203824996948242,
      "num_input_tokens_seen": 117703800,
      "step": 3690,
      "train_runtime": 9676.6665,
      "train_tokens_per_second": 12163.672
    },
    {
      "epoch": 0.2487880420145435,
      "grad_norm": 0.7519237597934966,
      "learning_rate": 9.623233349628441e-06,
      "loss": 0.3803875923156738,
      "num_input_tokens_seen": 117862392,
      "step": 3695,
      "train_runtime": 9689.7761,
      "train_tokens_per_second": 12163.583
    },
    {
      "epoch": 0.24912469701050363,
      "grad_norm": 0.8432619493847373,
      "learning_rate": 9.62222576768363e-06,
      "loss": 0.39006571769714354,
      "num_input_tokens_seen": 118021752,
      "step": 3700,
      "train_runtime": 9703.3832,
      "train_tokens_per_second": 12162.949
    },
    {
      "epoch": 0.24946135200646377,
      "grad_norm": 0.8922021759723706,
      "learning_rate": 9.62121689314812e-06,
      "loss": 0.45485243797302244,
      "num_input_tokens_seen": 118181152,
      "step": 3705,
      "train_runtime": 9716.7491,
      "train_tokens_per_second": 12162.623
    },
    {
      "epoch": 0.2497980070024239,
      "grad_norm": 0.909038938570329,
      "learning_rate": 9.620206726304045e-06,
      "loss": 0.4149444580078125,
      "num_input_tokens_seen": 118338520,
      "step": 3710,
      "train_runtime": 9730.0455,
      "train_tokens_per_second": 12162.175
    },
    {
      "epoch": 0.2501346619983841,
      "grad_norm": 0.816315175951374,
      "learning_rate": 9.61919526743389e-06,
      "loss": 0.44527549743652345,
      "num_input_tokens_seen": 118499080,
      "step": 3715,
      "train_runtime": 9743.0239,
      "train_tokens_per_second": 12162.454
    },
    {
      "epoch": 0.2504713169943442,
      "grad_norm": 0.8898586216688498,
      "learning_rate": 9.618182516820507e-06,
      "loss": 0.41652116775512693,
      "num_input_tokens_seen": 118658800,
      "step": 3720,
      "train_runtime": 9755.5248,
      "train_tokens_per_second": 12163.241
    },
    {
      "epoch": 0.25080797199030436,
      "grad_norm": 0.9284705137003402,
      "learning_rate": 9.61716847474711e-06,
      "loss": 0.4005583763122559,
      "num_input_tokens_seen": 118816936,
      "step": 3725,
      "train_runtime": 9768.5428,
      "train_tokens_per_second": 12163.22
    },
    {
      "epoch": 0.2511446269862645,
      "grad_norm": 0.9162132854818476,
      "learning_rate": 9.616153141497273e-06,
      "loss": 0.42869110107421876,
      "num_input_tokens_seen": 118977976,
      "step": 3730,
      "train_runtime": 9781.3517,
      "train_tokens_per_second": 12163.756
    },
    {
      "epoch": 0.25148128198222464,
      "grad_norm": 0.9835631763540246,
      "learning_rate": 9.61513651735493e-06,
      "loss": 0.3941942691802979,
      "num_input_tokens_seen": 119136280,
      "step": 3735,
      "train_runtime": 9794.0361,
      "train_tokens_per_second": 12164.166
    },
    {
      "epoch": 0.2518179369781848,
      "grad_norm": 0.9109271168084004,
      "learning_rate": 9.614118602604376e-06,
      "loss": 0.4011667251586914,
      "num_input_tokens_seen": 119294320,
      "step": 3740,
      "train_runtime": 9808.3307,
      "train_tokens_per_second": 12162.551
    },
    {
      "epoch": 0.2521545919741449,
      "grad_norm": 0.8934994154932886,
      "learning_rate": 9.613099397530267e-06,
      "loss": 0.3783116340637207,
      "num_input_tokens_seen": 119457592,
      "step": 3745,
      "train_runtime": 9821.9706,
      "train_tokens_per_second": 12162.284
    },
    {
      "epoch": 0.25249124697010505,
      "grad_norm": 0.9907929313420286,
      "learning_rate": 9.612078902417623e-06,
      "loss": 0.40924620628356934,
      "num_input_tokens_seen": 119618992,
      "step": 3750,
      "train_runtime": 9834.4945,
      "train_tokens_per_second": 12163.207
    },
    {
      "epoch": 0.2528279019660652,
      "grad_norm": 0.8377882888839838,
      "learning_rate": 9.611057117551821e-06,
      "loss": 0.4197675704956055,
      "num_input_tokens_seen": 119773872,
      "step": 3755,
      "train_runtime": 9848.0055,
      "train_tokens_per_second": 12162.247
    },
    {
      "epoch": 0.25316455696202533,
      "grad_norm": 0.8553504629298129,
      "learning_rate": 9.610034043218598e-06,
      "loss": 0.4289577007293701,
      "num_input_tokens_seen": 119937712,
      "step": 3760,
      "train_runtime": 9860.4203,
      "train_tokens_per_second": 12163.55
    },
    {
      "epoch": 0.2535012119579855,
      "grad_norm": 0.9789984046142575,
      "learning_rate": 9.609009679704057e-06,
      "loss": 0.42050838470458984,
      "num_input_tokens_seen": 120098256,
      "step": 3765,
      "train_runtime": 9874.0298,
      "train_tokens_per_second": 12163.044
    },
    {
      "epoch": 0.2538378669539456,
      "grad_norm": 0.8674853758619395,
      "learning_rate": 9.607984027294654e-06,
      "loss": 0.3979710578918457,
      "num_input_tokens_seen": 120255440,
      "step": 3770,
      "train_runtime": 9887.3676,
      "train_tokens_per_second": 12162.534
    },
    {
      "epoch": 0.25417452194990575,
      "grad_norm": 0.816873443414167,
      "learning_rate": 9.606957086277213e-06,
      "loss": 0.40817813873291015,
      "num_input_tokens_seen": 120415344,
      "step": 3775,
      "train_runtime": 9900.8466,
      "train_tokens_per_second": 12162.126
    },
    {
      "epoch": 0.2545111769458659,
      "grad_norm": 0.9594431870087078,
      "learning_rate": 9.605928856938913e-06,
      "loss": 0.41196308135986326,
      "num_input_tokens_seen": 120577640,
      "step": 3780,
      "train_runtime": 9915.6773,
      "train_tokens_per_second": 12160.303
    },
    {
      "epoch": 0.25484783194182603,
      "grad_norm": 0.7960282199265253,
      "learning_rate": 9.604899339567298e-06,
      "loss": 0.37593517303466795,
      "num_input_tokens_seen": 120735752,
      "step": 3785,
      "train_runtime": 9928.2264,
      "train_tokens_per_second": 12160.858
    },
    {
      "epoch": 0.25518448693778617,
      "grad_norm": 0.7962533782097339,
      "learning_rate": 9.603868534450263e-06,
      "loss": 0.36553468704223635,
      "num_input_tokens_seen": 120897616,
      "step": 3790,
      "train_runtime": 9940.6697,
      "train_tokens_per_second": 12161.919
    },
    {
      "epoch": 0.2555211419337463,
      "grad_norm": 0.8421695508225978,
      "learning_rate": 9.602836441876074e-06,
      "loss": 0.3682673692703247,
      "num_input_tokens_seen": 121059784,
      "step": 3795,
      "train_runtime": 9953.6002,
      "train_tokens_per_second": 12162.412
    },
    {
      "epoch": 0.25585779692970645,
      "grad_norm": 1.0912803893952283,
      "learning_rate": 9.601803062133353e-06,
      "loss": 0.41713991165161135,
      "num_input_tokens_seen": 121219216,
      "step": 3800,
      "train_runtime": 9966.083,
      "train_tokens_per_second": 12163.175
    },
    {
      "epoch": 0.2561944519256666,
      "grad_norm": 0.8021159110741525,
      "learning_rate": 9.600768395511078e-06,
      "loss": 0.3794227600097656,
      "num_input_tokens_seen": 121381704,
      "step": 3805,
      "train_runtime": 9978.8149,
      "train_tokens_per_second": 12163.94
    },
    {
      "epoch": 0.25653110692162673,
      "grad_norm": 0.779826167648677,
      "learning_rate": 9.599732442298592e-06,
      "loss": 0.3875307559967041,
      "num_input_tokens_seen": 121542336,
      "step": 3810,
      "train_runtime": 9992.353,
      "train_tokens_per_second": 12163.535
    },
    {
      "epoch": 0.25686776191758687,
      "grad_norm": 0.8825055218863129,
      "learning_rate": 9.598695202785596e-06,
      "loss": 0.40799479484558104,
      "num_input_tokens_seen": 121701792,
      "step": 3815,
      "train_runtime": 10005.9395,
      "train_tokens_per_second": 12162.955
    },
    {
      "epoch": 0.257204416913547,
      "grad_norm": 0.8194261588814046,
      "learning_rate": 9.59765667726215e-06,
      "loss": 0.37521777153015134,
      "num_input_tokens_seen": 121864104,
      "step": 3820,
      "train_runtime": 10019.9204,
      "train_tokens_per_second": 12162.183
    },
    {
      "epoch": 0.25754107190950715,
      "grad_norm": 0.9087654582196822,
      "learning_rate": 9.596616866018677e-06,
      "loss": 0.46813130378723145,
      "num_input_tokens_seen": 122020640,
      "step": 3825,
      "train_runtime": 10033.4693,
      "train_tokens_per_second": 12161.361
    },
    {
      "epoch": 0.2578777269054673,
      "grad_norm": 0.8257266846043245,
      "learning_rate": 9.595575769345955e-06,
      "loss": 0.408492374420166,
      "num_input_tokens_seen": 122171256,
      "step": 3830,
      "train_runtime": 10046.232,
      "train_tokens_per_second": 12160.903
    },
    {
      "epoch": 0.2582143819014274,
      "grad_norm": 0.8801058611550759,
      "learning_rate": 9.594533387535122e-06,
      "loss": 0.4131655693054199,
      "num_input_tokens_seen": 122332304,
      "step": 3835,
      "train_runtime": 10059.3406,
      "train_tokens_per_second": 12161.066
    },
    {
      "epoch": 0.25855103689738757,
      "grad_norm": 0.831723258226366,
      "learning_rate": 9.593489720877678e-06,
      "loss": 0.40161967277526855,
      "num_input_tokens_seen": 122493200,
      "step": 3840,
      "train_runtime": 10073.4046,
      "train_tokens_per_second": 12160.06
    },
    {
      "epoch": 0.2588876918933477,
      "grad_norm": 0.7614200186829221,
      "learning_rate": 9.592444769665482e-06,
      "loss": 0.38043780326843263,
      "num_input_tokens_seen": 122657040,
      "step": 3845,
      "train_runtime": 10085.8317,
      "train_tokens_per_second": 12161.321
    },
    {
      "epoch": 0.25922434688930784,
      "grad_norm": 1.0088565155547762,
      "learning_rate": 9.59139853419075e-06,
      "loss": 0.3997699975967407,
      "num_input_tokens_seen": 122808368,
      "step": 3850,
      "train_runtime": 10099.1824,
      "train_tokens_per_second": 12160.229
    },
    {
      "epoch": 0.259561001885268,
      "grad_norm": 0.8767073880258983,
      "learning_rate": 9.590351014746059e-06,
      "loss": 0.40987696647644045,
      "num_input_tokens_seen": 122968912,
      "step": 3855,
      "train_runtime": 10112.0461,
      "train_tokens_per_second": 12160.636
    },
    {
      "epoch": 0.2598976568812281,
      "grad_norm": 0.8828697336988384,
      "learning_rate": 9.589302211624343e-06,
      "loss": 0.4118494987487793,
      "num_input_tokens_seen": 123127512,
      "step": 3860,
      "train_runtime": 10125.2526,
      "train_tokens_per_second": 12160.439
    },
    {
      "epoch": 0.26023431187718826,
      "grad_norm": 0.8868223300685006,
      "learning_rate": 9.5882521251189e-06,
      "loss": 0.40972070693969725,
      "num_input_tokens_seen": 123287848,
      "step": 3865,
      "train_runtime": 10138.0573,
      "train_tokens_per_second": 12160.895
    },
    {
      "epoch": 0.2605709668731484,
      "grad_norm": 1.2065392684686351,
      "learning_rate": 9.58720075552338e-06,
      "loss": 0.4448376655578613,
      "num_input_tokens_seen": 123447760,
      "step": 3870,
      "train_runtime": 10151.7607,
      "train_tokens_per_second": 12160.231
    },
    {
      "epoch": 0.26090762186910854,
      "grad_norm": 0.9136276056161826,
      "learning_rate": 9.586148103131797e-06,
      "loss": 0.41599197387695314,
      "num_input_tokens_seen": 123602968,
      "step": 3875,
      "train_runtime": 10165.7037,
      "train_tokens_per_second": 12158.821
    },
    {
      "epoch": 0.2612442768650687,
      "grad_norm": 0.8954825767082152,
      "learning_rate": 9.585094168238518e-06,
      "loss": 0.4437089920043945,
      "num_input_tokens_seen": 123760432,
      "step": 3880,
      "train_runtime": 10178.7516,
      "train_tokens_per_second": 12158.704
    },
    {
      "epoch": 0.2615809318610288,
      "grad_norm": 0.8267604883535505,
      "learning_rate": 9.58403895113828e-06,
      "loss": 0.4033670425415039,
      "num_input_tokens_seen": 123919920,
      "step": 3885,
      "train_runtime": 10191.7169,
      "train_tokens_per_second": 12158.886
    },
    {
      "epoch": 0.26191758685698896,
      "grad_norm": 1.0222514064696477,
      "learning_rate": 9.582982452126164e-06,
      "loss": 0.38756608963012695,
      "num_input_tokens_seen": 124077120,
      "step": 3890,
      "train_runtime": 10204.4045,
      "train_tokens_per_second": 12159.173
    },
    {
      "epoch": 0.2622542418529491,
      "grad_norm": 0.8430500795054858,
      "learning_rate": 9.58192467149762e-06,
      "loss": 0.3738124370574951,
      "num_input_tokens_seen": 124230192,
      "step": 3895,
      "train_runtime": 10216.8592,
      "train_tokens_per_second": 12159.333
    },
    {
      "epoch": 0.26259089684890924,
      "grad_norm": 0.8924627372368323,
      "learning_rate": 9.580865609548452e-06,
      "loss": 0.4181983470916748,
      "num_input_tokens_seen": 124380808,
      "step": 3900,
      "train_runtime": 10229.7768,
      "train_tokens_per_second": 12158.702
    },
    {
      "epoch": 0.2629275518448694,
      "grad_norm": 0.7468481866069557,
      "learning_rate": 9.579805266574824e-06,
      "loss": 0.38424651622772216,
      "num_input_tokens_seen": 124542288,
      "step": 3905,
      "train_runtime": 10244.0941,
      "train_tokens_per_second": 12157.472
    },
    {
      "epoch": 0.2632642068408295,
      "grad_norm": 0.8568115894821399,
      "learning_rate": 9.578743642873258e-06,
      "loss": 0.3943238973617554,
      "num_input_tokens_seen": 124697856,
      "step": 3910,
      "train_runtime": 10256.8037,
      "train_tokens_per_second": 12157.575
    },
    {
      "epoch": 0.26360086183678966,
      "grad_norm": 0.8326496620327833,
      "learning_rate": 9.57768073874063e-06,
      "loss": 0.41575775146484373,
      "num_input_tokens_seen": 124858128,
      "step": 3915,
      "train_runtime": 10269.7185,
      "train_tokens_per_second": 12157.892
    },
    {
      "epoch": 0.2639375168327498,
      "grad_norm": 0.8295294469717871,
      "learning_rate": 9.576616554474182e-06,
      "loss": 0.4290400505065918,
      "num_input_tokens_seen": 125011600,
      "step": 3920,
      "train_runtime": 10283.5875,
      "train_tokens_per_second": 12156.419
    },
    {
      "epoch": 0.26427417182870994,
      "grad_norm": 0.8719151139755162,
      "learning_rate": 9.575551090371512e-06,
      "loss": 0.4100490093231201,
      "num_input_tokens_seen": 125167128,
      "step": 3925,
      "train_runtime": 10297.6114,
      "train_tokens_per_second": 12154.967
    },
    {
      "epoch": 0.2646108268246701,
      "grad_norm": 0.8752971220136497,
      "learning_rate": 9.574484346730564e-06,
      "loss": 0.4104010105133057,
      "num_input_tokens_seen": 125326112,
      "step": 3930,
      "train_runtime": 10311.0567,
      "train_tokens_per_second": 12154.536
    },
    {
      "epoch": 0.2649474818206302,
      "grad_norm": 0.9716557279175105,
      "learning_rate": 9.573416323849658e-06,
      "loss": 0.4124293327331543,
      "num_input_tokens_seen": 125489136,
      "step": 3935,
      "train_runtime": 10325.4421,
      "train_tokens_per_second": 12153.391
    },
    {
      "epoch": 0.26528413681659035,
      "grad_norm": 0.8661697653347159,
      "learning_rate": 9.572347022027461e-06,
      "loss": 0.4076035499572754,
      "num_input_tokens_seen": 125650216,
      "step": 3940,
      "train_runtime": 10338.5406,
      "train_tokens_per_second": 12153.574
    },
    {
      "epoch": 0.2656207918125505,
      "grad_norm": 0.7906552471950526,
      "learning_rate": 9.571276441563e-06,
      "loss": 0.40136399269104006,
      "num_input_tokens_seen": 125801104,
      "step": 3945,
      "train_runtime": 10351.3617,
      "train_tokens_per_second": 12153.097
    },
    {
      "epoch": 0.26595744680851063,
      "grad_norm": 0.8572791710316278,
      "learning_rate": 9.570204582755659e-06,
      "loss": 0.404392147064209,
      "num_input_tokens_seen": 125960768,
      "step": 3950,
      "train_runtime": 10364.0359,
      "train_tokens_per_second": 12153.641
    },
    {
      "epoch": 0.2662941018044708,
      "grad_norm": 0.8902696335485862,
      "learning_rate": 9.569131445905178e-06,
      "loss": 0.42226834297180177,
      "num_input_tokens_seen": 126118184,
      "step": 3955,
      "train_runtime": 10377.4825,
      "train_tokens_per_second": 12153.062
    },
    {
      "epoch": 0.2666307568004309,
      "grad_norm": 0.9207282485771399,
      "learning_rate": 9.568057031311659e-06,
      "loss": 0.4274171829223633,
      "num_input_tokens_seen": 126272784,
      "step": 3960,
      "train_runtime": 10390.9154,
      "train_tokens_per_second": 12152.229
    },
    {
      "epoch": 0.26696741179639105,
      "grad_norm": 0.9286636294772894,
      "learning_rate": 9.566981339275557e-06,
      "loss": 0.44168734550476074,
      "num_input_tokens_seen": 126430944,
      "step": 3965,
      "train_runtime": 10405.1173,
      "train_tokens_per_second": 12150.843
    },
    {
      "epoch": 0.2673040667923512,
      "grad_norm": 0.8502211820106121,
      "learning_rate": 9.565904370097687e-06,
      "loss": 0.40860185623168943,
      "num_input_tokens_seen": 126585376,
      "step": 3970,
      "train_runtime": 10419.0017,
      "train_tokens_per_second": 12149.473
    },
    {
      "epoch": 0.26764072178831133,
      "grad_norm": 0.8249241010822921,
      "learning_rate": 9.564826124079219e-06,
      "loss": 0.38613989353179934,
      "num_input_tokens_seen": 126745440,
      "step": 3975,
      "train_runtime": 10431.4835,
      "train_tokens_per_second": 12150.28
    },
    {
      "epoch": 0.26797737678427147,
      "grad_norm": 0.8818213747826239,
      "learning_rate": 9.563746601521681e-06,
      "loss": 0.37488436698913574,
      "num_input_tokens_seen": 126905040,
      "step": 3980,
      "train_runtime": 10444.587,
      "train_tokens_per_second": 12150.317
    },
    {
      "epoch": 0.2683140317802316,
      "grad_norm": 0.8918529631279178,
      "learning_rate": 9.562665802726961e-06,
      "loss": 0.4101956844329834,
      "num_input_tokens_seen": 127067696,
      "step": 3985,
      "train_runtime": 10457.3572,
      "train_tokens_per_second": 12151.033
    },
    {
      "epoch": 0.26865068677619175,
      "grad_norm": 0.9196599081003479,
      "learning_rate": 9.561583727997297e-06,
      "loss": 0.4102800369262695,
      "num_input_tokens_seen": 127224496,
      "step": 3990,
      "train_runtime": 10470.3267,
      "train_tokens_per_second": 12150.958
    },
    {
      "epoch": 0.2689873417721519,
      "grad_norm": 1.1778946563404906,
      "learning_rate": 9.560500377635288e-06,
      "loss": 0.4151009082794189,
      "num_input_tokens_seen": 127383952,
      "step": 3995,
      "train_runtime": 10482.9527,
      "train_tokens_per_second": 12151.534
    },
    {
      "epoch": 0.26932399676811203,
      "grad_norm": 0.9621523823599923,
      "learning_rate": 9.559415751943893e-06,
      "loss": 0.4220369815826416,
      "num_input_tokens_seen": 127542080,
      "step": 4000,
      "train_runtime": 10496.592,
      "train_tokens_per_second": 12150.809
    },
    {
      "epoch": 0.26966065176407217,
      "grad_norm": 0.8641340293653632,
      "learning_rate": 9.558329851226422e-06,
      "loss": 0.4216341018676758,
      "num_input_tokens_seen": 127702264,
      "step": 4005,
      "train_runtime": 10509.7822,
      "train_tokens_per_second": 12150.8
    },
    {
      "epoch": 0.2699973067600323,
      "grad_norm": 0.8592458141913455,
      "learning_rate": 9.557242675786542e-06,
      "loss": 0.4289440631866455,
      "num_input_tokens_seen": 127864824,
      "step": 4010,
      "train_runtime": 10522.7404,
      "train_tokens_per_second": 12151.286
    },
    {
      "epoch": 0.27033396175599245,
      "grad_norm": 0.8034543033623522,
      "learning_rate": 9.55615422592828e-06,
      "loss": 0.44036264419555665,
      "num_input_tokens_seen": 128026096,
      "step": 4015,
      "train_runtime": 10535.8616,
      "train_tokens_per_second": 12151.46
    },
    {
      "epoch": 0.2706706167519526,
      "grad_norm": 0.9076276734306452,
      "learning_rate": 9.555064501956018e-06,
      "loss": 0.42022361755371096,
      "num_input_tokens_seen": 128183400,
      "step": 4020,
      "train_runtime": 10549.2538,
      "train_tokens_per_second": 12150.945
    },
    {
      "epoch": 0.2710072717479127,
      "grad_norm": 0.8070570838578593,
      "learning_rate": 9.553973504174493e-06,
      "loss": 0.3850049257278442,
      "num_input_tokens_seen": 128345912,
      "step": 4025,
      "train_runtime": 10561.9287,
      "train_tokens_per_second": 12151.75
    },
    {
      "epoch": 0.27134392674387287,
      "grad_norm": 0.8797477977719831,
      "learning_rate": 9.5528812328888e-06,
      "loss": 0.42907538414001467,
      "num_input_tokens_seen": 128504512,
      "step": 4030,
      "train_runtime": 10575.6451,
      "train_tokens_per_second": 12150.986
    },
    {
      "epoch": 0.271680581739833,
      "grad_norm": 1.4841886312416466,
      "learning_rate": 9.551787688404386e-06,
      "loss": 0.3940990447998047,
      "num_input_tokens_seen": 128661440,
      "step": 4035,
      "train_runtime": 10589.0185,
      "train_tokens_per_second": 12150.46
    },
    {
      "epoch": 0.27201723673579314,
      "grad_norm": 0.8376724870244265,
      "learning_rate": 9.550692871027061e-06,
      "loss": 0.39757022857666013,
      "num_input_tokens_seen": 128823512,
      "step": 4040,
      "train_runtime": 10601.4746,
      "train_tokens_per_second": 12151.471
    },
    {
      "epoch": 0.2723538917317533,
      "grad_norm": 0.940089191419943,
      "learning_rate": 9.549596781062984e-06,
      "loss": 0.37547945976257324,
      "num_input_tokens_seen": 128987352,
      "step": 4045,
      "train_runtime": 10613.9244,
      "train_tokens_per_second": 12152.654
    },
    {
      "epoch": 0.2726905467277134,
      "grad_norm": 1.0051950843366402,
      "learning_rate": 9.548499418818677e-06,
      "loss": 0.38057746887207033,
      "num_input_tokens_seen": 129146048,
      "step": 4050,
      "train_runtime": 10627.3533,
      "train_tokens_per_second": 12152.231
    },
    {
      "epoch": 0.27302720172367356,
      "grad_norm": 0.8633820138217068,
      "learning_rate": 9.547400784601011e-06,
      "loss": 0.4335151672363281,
      "num_input_tokens_seen": 129306696,
      "step": 4055,
      "train_runtime": 10640.8708,
      "train_tokens_per_second": 12151.89
    },
    {
      "epoch": 0.2733638567196337,
      "grad_norm": 0.8537731535280159,
      "learning_rate": 9.546300878717216e-06,
      "loss": 0.4264829158782959,
      "num_input_tokens_seen": 129467632,
      "step": 4060,
      "train_runtime": 10654.5637,
      "train_tokens_per_second": 12151.378
    },
    {
      "epoch": 0.27370051171559384,
      "grad_norm": 0.7561876786663848,
      "learning_rate": 9.545199701474877e-06,
      "loss": 0.4020991325378418,
      "num_input_tokens_seen": 129631472,
      "step": 4065,
      "train_runtime": 10666.9987,
      "train_tokens_per_second": 12152.572
    },
    {
      "epoch": 0.274037166711554,
      "grad_norm": 0.8677494015248345,
      "learning_rate": 9.544097253181935e-06,
      "loss": 0.38918609619140626,
      "num_input_tokens_seen": 129788824,
      "step": 4070,
      "train_runtime": 10680.0504,
      "train_tokens_per_second": 12152.454
    },
    {
      "epoch": 0.2743738217075141,
      "grad_norm": 0.8372079147921017,
      "learning_rate": 9.542993534146687e-06,
      "loss": 0.4239842414855957,
      "num_input_tokens_seen": 129947792,
      "step": 4075,
      "train_runtime": 10693.3772,
      "train_tokens_per_second": 12152.175
    },
    {
      "epoch": 0.27471047670347426,
      "grad_norm": 0.8691398700152315,
      "learning_rate": 9.541888544677784e-06,
      "loss": 0.40438117980957033,
      "num_input_tokens_seen": 130105792,
      "step": 4080,
      "train_runtime": 10705.8545,
      "train_tokens_per_second": 12152.77
    },
    {
      "epoch": 0.2750471316994344,
      "grad_norm": 0.8466844758295657,
      "learning_rate": 9.540782285084232e-06,
      "loss": 0.36170125007629395,
      "num_input_tokens_seen": 130260848,
      "step": 4085,
      "train_runtime": 10718.7616,
      "train_tokens_per_second": 12152.602
    },
    {
      "epoch": 0.27538378669539454,
      "grad_norm": 0.8411277603364619,
      "learning_rate": 9.539674755675392e-06,
      "loss": 0.39405972957611085,
      "num_input_tokens_seen": 130421640,
      "step": 4090,
      "train_runtime": 10732.8636,
      "train_tokens_per_second": 12151.616
    },
    {
      "epoch": 0.2757204416913547,
      "grad_norm": 3.2958252601384914,
      "learning_rate": 9.538565956760983e-06,
      "loss": 0.36522512435913085,
      "num_input_tokens_seen": 130583688,
      "step": 4095,
      "train_runtime": 10746.1308,
      "train_tokens_per_second": 12151.694
    },
    {
      "epoch": 0.2760570966873148,
      "grad_norm": 0.8613146328251964,
      "learning_rate": 9.537455888651078e-06,
      "loss": 0.4022481918334961,
      "num_input_tokens_seen": 130747352,
      "step": 4100,
      "train_runtime": 10759.1879,
      "train_tokens_per_second": 12152.158
    },
    {
      "epoch": 0.27639375168327496,
      "grad_norm": 1.3186680984598695,
      "learning_rate": 9.536344551656103e-06,
      "loss": 0.4055627822875977,
      "num_input_tokens_seen": 130908840,
      "step": 4105,
      "train_runtime": 10772.8159,
      "train_tokens_per_second": 12151.775
    },
    {
      "epoch": 0.2767304066792351,
      "grad_norm": 0.8543652458654861,
      "learning_rate": 9.535231946086838e-06,
      "loss": 0.40320382118225095,
      "num_input_tokens_seen": 131065584,
      "step": 4110,
      "train_runtime": 10786.1636,
      "train_tokens_per_second": 12151.27
    },
    {
      "epoch": 0.27706706167519524,
      "grad_norm": 0.8340108259232891,
      "learning_rate": 9.534118072254421e-06,
      "loss": 0.41141462326049805,
      "num_input_tokens_seen": 131227928,
      "step": 4115,
      "train_runtime": 10798.7898,
      "train_tokens_per_second": 12152.096
    },
    {
      "epoch": 0.2774037166711554,
      "grad_norm": 0.8529357998550278,
      "learning_rate": 9.533002930470345e-06,
      "loss": 0.3830364227294922,
      "num_input_tokens_seen": 131388264,
      "step": 4120,
      "train_runtime": 10811.601,
      "train_tokens_per_second": 12152.526
    },
    {
      "epoch": 0.2777403716671155,
      "grad_norm": 0.7250703589785162,
      "learning_rate": 9.531886521046452e-06,
      "loss": 0.38754801750183104,
      "num_input_tokens_seen": 131544184,
      "step": 4125,
      "train_runtime": 10824.9104,
      "train_tokens_per_second": 12151.988
    },
    {
      "epoch": 0.27807702666307565,
      "grad_norm": 0.8427427927057928,
      "learning_rate": 9.530768844294947e-06,
      "loss": 0.4119289398193359,
      "num_input_tokens_seen": 131703456,
      "step": 4130,
      "train_runtime": 10837.9826,
      "train_tokens_per_second": 12152.027
    },
    {
      "epoch": 0.2784136816590358,
      "grad_norm": 0.8791185440797648,
      "learning_rate": 9.52964990052838e-06,
      "loss": 0.42822465896606443,
      "num_input_tokens_seen": 131861344,
      "step": 4135,
      "train_runtime": 10851.2268,
      "train_tokens_per_second": 12151.745
    },
    {
      "epoch": 0.27875033665499593,
      "grad_norm": 0.820978091717965,
      "learning_rate": 9.528529690059663e-06,
      "loss": 0.3799431800842285,
      "num_input_tokens_seen": 132024592,
      "step": 4140,
      "train_runtime": 10864.1584,
      "train_tokens_per_second": 12152.307
    },
    {
      "epoch": 0.2790869916509561,
      "grad_norm": 0.9925027913446373,
      "learning_rate": 9.52740821320206e-06,
      "loss": 0.4137421607971191,
      "num_input_tokens_seen": 132182088,
      "step": 4145,
      "train_runtime": 10877.5079,
      "train_tokens_per_second": 12151.872
    },
    {
      "epoch": 0.2794236466469162,
      "grad_norm": 0.8612955543167724,
      "learning_rate": 9.526285470269185e-06,
      "loss": 0.44510555267333984,
      "num_input_tokens_seen": 132344056,
      "step": 4150,
      "train_runtime": 10891.8434,
      "train_tokens_per_second": 12150.749
    },
    {
      "epoch": 0.2797603016428764,
      "grad_norm": 0.8456208582653145,
      "learning_rate": 9.525161461575012e-06,
      "loss": 0.40246405601501467,
      "num_input_tokens_seen": 132499456,
      "step": 4155,
      "train_runtime": 10905.0227,
      "train_tokens_per_second": 12150.315
    },
    {
      "epoch": 0.28009695663883655,
      "grad_norm": 0.791031346916803,
      "learning_rate": 9.524036187433867e-06,
      "loss": 0.3801803827285767,
      "num_input_tokens_seen": 132654568,
      "step": 4160,
      "train_runtime": 10918.5434,
      "train_tokens_per_second": 12149.475
    },
    {
      "epoch": 0.2804336116347967,
      "grad_norm": 0.9901434372888882,
      "learning_rate": 9.522909648160427e-06,
      "loss": 0.39869370460510256,
      "num_input_tokens_seen": 132811560,
      "step": 4165,
      "train_runtime": 10931.6236,
      "train_tokens_per_second": 12149.299
    },
    {
      "epoch": 0.2807702666307568,
      "grad_norm": 0.9555149736622928,
      "learning_rate": 9.521781844069727e-06,
      "loss": 0.4037034034729004,
      "num_input_tokens_seen": 132971800,
      "step": 4170,
      "train_runtime": 10944.5622,
      "train_tokens_per_second": 12149.577
    },
    {
      "epoch": 0.28110692162671697,
      "grad_norm": 0.8749219237389935,
      "learning_rate": 9.520652775477154e-06,
      "loss": 0.40973973274230957,
      "num_input_tokens_seen": 133130120,
      "step": 4175,
      "train_runtime": 10957.7183,
      "train_tokens_per_second": 12149.438
    },
    {
      "epoch": 0.2814435766226771,
      "grad_norm": 0.7967954387626621,
      "learning_rate": 9.519522442698447e-06,
      "loss": 0.3809692621231079,
      "num_input_tokens_seen": 133282672,
      "step": 4180,
      "train_runtime": 10971.0443,
      "train_tokens_per_second": 12148.586
    },
    {
      "epoch": 0.28178023161863724,
      "grad_norm": 0.8389319962323508,
      "learning_rate": 9.518390846049699e-06,
      "loss": 0.40291223526000974,
      "num_input_tokens_seen": 133441208,
      "step": 4185,
      "train_runtime": 10983.5519,
      "train_tokens_per_second": 12149.185
    },
    {
      "epoch": 0.2821168866145974,
      "grad_norm": 0.8022457546594207,
      "learning_rate": 9.517257985847361e-06,
      "loss": 0.43160762786865237,
      "num_input_tokens_seen": 133600032,
      "step": 4190,
      "train_runtime": 10996.1743,
      "train_tokens_per_second": 12149.683
    },
    {
      "epoch": 0.2824535416105575,
      "grad_norm": 0.8403775711561599,
      "learning_rate": 9.516123862408232e-06,
      "loss": 0.45180859565734866,
      "num_input_tokens_seen": 133763832,
      "step": 4195,
      "train_runtime": 11009.9286,
      "train_tokens_per_second": 12149.382
    },
    {
      "epoch": 0.28279019660651766,
      "grad_norm": 0.9073939948857712,
      "learning_rate": 9.514988476049466e-06,
      "loss": 0.3903691053390503,
      "num_input_tokens_seen": 133921352,
      "step": 4200,
      "train_runtime": 11023.2191,
      "train_tokens_per_second": 12149.024
    },
    {
      "epoch": 0.2831268516024778,
      "grad_norm": 0.7845901590607888,
      "learning_rate": 9.513851827088567e-06,
      "loss": 0.40480241775512693,
      "num_input_tokens_seen": 134077496,
      "step": 4205,
      "train_runtime": 11035.7154,
      "train_tokens_per_second": 12149.416
    },
    {
      "epoch": 0.28346350659843794,
      "grad_norm": 0.9707194632348691,
      "learning_rate": 9.512713915843402e-06,
      "loss": 0.3853065013885498,
      "num_input_tokens_seen": 134238456,
      "step": 4210,
      "train_runtime": 11048.8357,
      "train_tokens_per_second": 12149.557
    },
    {
      "epoch": 0.2838001615943981,
      "grad_norm": 0.7882020899530009,
      "learning_rate": 9.511574742632177e-06,
      "loss": 0.38508694171905516,
      "num_input_tokens_seen": 134401968,
      "step": 4215,
      "train_runtime": 11061.3181,
      "train_tokens_per_second": 12150.629
    },
    {
      "epoch": 0.2841368165903582,
      "grad_norm": 0.8245499350242305,
      "learning_rate": 9.510434307773464e-06,
      "loss": 0.46449713706970214,
      "num_input_tokens_seen": 134563912,
      "step": 4220,
      "train_runtime": 11074.4166,
      "train_tokens_per_second": 12150.88
    },
    {
      "epoch": 0.28447347158631836,
      "grad_norm": 0.8270810808899612,
      "learning_rate": 9.509292611586179e-06,
      "loss": 0.3642393112182617,
      "num_input_tokens_seen": 134713656,
      "step": 4225,
      "train_runtime": 11087.3679,
      "train_tokens_per_second": 12150.193
    },
    {
      "epoch": 0.2848101265822785,
      "grad_norm": 0.8514366882422044,
      "learning_rate": 9.508149654389592e-06,
      "loss": 0.40613679885864257,
      "num_input_tokens_seen": 134877496,
      "step": 4230,
      "train_runtime": 11099.8146,
      "train_tokens_per_second": 12151.329
    },
    {
      "epoch": 0.28514678157823864,
      "grad_norm": 0.7351343130063843,
      "learning_rate": 9.507005436503331e-06,
      "loss": 0.38147788047790526,
      "num_input_tokens_seen": 135039464,
      "step": 4235,
      "train_runtime": 11112.3947,
      "train_tokens_per_second": 12152.148
    },
    {
      "epoch": 0.2854834365741988,
      "grad_norm": 0.8921418832255089,
      "learning_rate": 9.505859958247373e-06,
      "loss": 0.41287670135498045,
      "num_input_tokens_seen": 135200480,
      "step": 4240,
      "train_runtime": 11126.0739,
      "train_tokens_per_second": 12151.679
    },
    {
      "epoch": 0.2858200915701589,
      "grad_norm": 1.0955574578627245,
      "learning_rate": 9.504713219942045e-06,
      "loss": 0.38956336975097655,
      "num_input_tokens_seen": 135362656,
      "step": 4245,
      "train_runtime": 11138.8697,
      "train_tokens_per_second": 12152.279
    },
    {
      "epoch": 0.28615674656611906,
      "grad_norm": 1.0346635278780059,
      "learning_rate": 9.503565221908029e-06,
      "loss": 0.40523953437805177,
      "num_input_tokens_seen": 135521928,
      "step": 4250,
      "train_runtime": 11152.0297,
      "train_tokens_per_second": 12152.221
    },
    {
      "epoch": 0.2864934015620792,
      "grad_norm": 0.8283715162137227,
      "learning_rate": 9.50241596446636e-06,
      "loss": 0.43901691436767576,
      "num_input_tokens_seen": 135681936,
      "step": 4255,
      "train_runtime": 11165.178,
      "train_tokens_per_second": 12152.241
    },
    {
      "epoch": 0.28683005655803934,
      "grad_norm": 0.8646236196491948,
      "learning_rate": 9.501265447938423e-06,
      "loss": 0.39167323112487795,
      "num_input_tokens_seen": 135838168,
      "step": 4260,
      "train_runtime": 11178.9682,
      "train_tokens_per_second": 12151.226
    },
    {
      "epoch": 0.2871667115539995,
      "grad_norm": 0.909317286454628,
      "learning_rate": 9.500113672645958e-06,
      "loss": 0.4129340648651123,
      "num_input_tokens_seen": 135995232,
      "step": 4265,
      "train_runtime": 11192.2296,
      "train_tokens_per_second": 12150.862
    },
    {
      "epoch": 0.2875033665499596,
      "grad_norm": 0.8463895506085269,
      "learning_rate": 9.498960638911054e-06,
      "loss": 0.38132309913635254,
      "num_input_tokens_seen": 136151416,
      "step": 4270,
      "train_runtime": 11205.0992,
      "train_tokens_per_second": 12150.844
    },
    {
      "epoch": 0.28784002154591976,
      "grad_norm": 1.3972308100657642,
      "learning_rate": 9.497806347056153e-06,
      "loss": 0.4125657081604004,
      "num_input_tokens_seen": 136314760,
      "step": 4275,
      "train_runtime": 11218.7557,
      "train_tokens_per_second": 12150.613
    },
    {
      "epoch": 0.2881766765418799,
      "grad_norm": 0.9522163184661055,
      "learning_rate": 9.49665079740405e-06,
      "loss": 0.42870287895202636,
      "num_input_tokens_seen": 136473352,
      "step": 4280,
      "train_runtime": 11231.4159,
      "train_tokens_per_second": 12151.037
    },
    {
      "epoch": 0.28851333153784003,
      "grad_norm": 1.1180231154347653,
      "learning_rate": 9.495493990277889e-06,
      "loss": 0.41233291625976565,
      "num_input_tokens_seen": 136630312,
      "step": 4285,
      "train_runtime": 11244.4219,
      "train_tokens_per_second": 12150.941
    },
    {
      "epoch": 0.2888499865338002,
      "grad_norm": 0.8815055094974447,
      "learning_rate": 9.49433592600117e-06,
      "loss": 0.4076373100280762,
      "num_input_tokens_seen": 136794152,
      "step": 4290,
      "train_runtime": 11256.8586,
      "train_tokens_per_second": 12152.072
    },
    {
      "epoch": 0.2891866415297603,
      "grad_norm": 1.0161708837721504,
      "learning_rate": 9.49317660489774e-06,
      "loss": 0.41091256141662597,
      "num_input_tokens_seen": 136956864,
      "step": 4295,
      "train_runtime": 11269.5927,
      "train_tokens_per_second": 12152.778
    },
    {
      "epoch": 0.28952329652572045,
      "grad_norm": 1.349032180296885,
      "learning_rate": 9.4920160272918e-06,
      "loss": 0.3965152263641357,
      "num_input_tokens_seen": 137117528,
      "step": 4300,
      "train_runtime": 11282.0668,
      "train_tokens_per_second": 12153.582
    },
    {
      "epoch": 0.2898599515216806,
      "grad_norm": 1.094701939369197,
      "learning_rate": 9.490854193507904e-06,
      "loss": 0.4082478523254395,
      "num_input_tokens_seen": 137275328,
      "step": 4305,
      "train_runtime": 11295.4408,
      "train_tokens_per_second": 12153.163
    },
    {
      "epoch": 0.29019660651764073,
      "grad_norm": 0.9570476290737923,
      "learning_rate": 9.489691103870951e-06,
      "loss": 0.4289418697357178,
      "num_input_tokens_seen": 137433528,
      "step": 4310,
      "train_runtime": 11307.9997,
      "train_tokens_per_second": 12153.655
    },
    {
      "epoch": 0.29053326151360087,
      "grad_norm": 0.9234343510694029,
      "learning_rate": 9.488526758706198e-06,
      "loss": 0.39855406284332273,
      "num_input_tokens_seen": 137592952,
      "step": 4315,
      "train_runtime": 11320.7537,
      "train_tokens_per_second": 12154.045
    },
    {
      "epoch": 0.290869916509561,
      "grad_norm": 0.9703115717445585,
      "learning_rate": 9.48736115833925e-06,
      "loss": 0.4281930446624756,
      "num_input_tokens_seen": 137745600,
      "step": 4320,
      "train_runtime": 11334.1515,
      "train_tokens_per_second": 12153.146
    },
    {
      "epoch": 0.29120657150552115,
      "grad_norm": 0.8103780205610209,
      "learning_rate": 9.486194303096062e-06,
      "loss": 0.3839207410812378,
      "num_input_tokens_seen": 137908784,
      "step": 4325,
      "train_runtime": 11347.7586,
      "train_tokens_per_second": 12152.954
    },
    {
      "epoch": 0.2915432265014813,
      "grad_norm": 0.8243675909986489,
      "learning_rate": 9.485026193302945e-06,
      "loss": 0.3825213432312012,
      "num_input_tokens_seen": 138070808,
      "step": 4330,
      "train_runtime": 11360.9226,
      "train_tokens_per_second": 12153.133
    },
    {
      "epoch": 0.29187988149744143,
      "grad_norm": 0.7249358032108992,
      "learning_rate": 9.483856829286552e-06,
      "loss": 0.44112935066223147,
      "num_input_tokens_seen": 138232360,
      "step": 4335,
      "train_runtime": 11373.382,
      "train_tokens_per_second": 12154.024
    },
    {
      "epoch": 0.29221653649340157,
      "grad_norm": 1.0451893486261747,
      "learning_rate": 9.482686211373896e-06,
      "loss": 0.4225608825683594,
      "num_input_tokens_seen": 138395136,
      "step": 4340,
      "train_runtime": 11386.1979,
      "train_tokens_per_second": 12154.64
    },
    {
      "epoch": 0.2925531914893617,
      "grad_norm": 0.9436916338064438,
      "learning_rate": 9.481514339892335e-06,
      "loss": 0.4296980381011963,
      "num_input_tokens_seen": 138557648,
      "step": 4345,
      "train_runtime": 11398.8661,
      "train_tokens_per_second": 12155.389
    },
    {
      "epoch": 0.29288984648532185,
      "grad_norm": 0.7728240407150901,
      "learning_rate": 9.48034121516958e-06,
      "loss": 0.4123666286468506,
      "num_input_tokens_seen": 138715080,
      "step": 4350,
      "train_runtime": 11412.1857,
      "train_tokens_per_second": 12154.997
    },
    {
      "epoch": 0.293226501481282,
      "grad_norm": 0.7324110291364622,
      "learning_rate": 9.47916683753369e-06,
      "loss": 0.4054971694946289,
      "num_input_tokens_seen": 138875448,
      "step": 4355,
      "train_runtime": 11425.3298,
      "train_tokens_per_second": 12155.049
    },
    {
      "epoch": 0.2935631564772421,
      "grad_norm": 0.8851222055351642,
      "learning_rate": 9.477991207313077e-06,
      "loss": 0.41019449234008787,
      "num_input_tokens_seen": 139035344,
      "step": 4360,
      "train_runtime": 11437.9101,
      "train_tokens_per_second": 12155.66
    },
    {
      "epoch": 0.29389981147320227,
      "grad_norm": 0.8419158898950759,
      "learning_rate": 9.476814324836504e-06,
      "loss": 0.39893355369567873,
      "num_input_tokens_seen": 139197800,
      "step": 4365,
      "train_runtime": 11450.5808,
      "train_tokens_per_second": 12156.397
    },
    {
      "epoch": 0.2942364664691624,
      "grad_norm": 0.9172784511100571,
      "learning_rate": 9.475636190433078e-06,
      "loss": 0.46141653060913085,
      "num_input_tokens_seen": 139359936,
      "step": 4370,
      "train_runtime": 11463.0487,
      "train_tokens_per_second": 12157.319
    },
    {
      "epoch": 0.29457312146512254,
      "grad_norm": 0.8131494523021554,
      "learning_rate": 9.474456804432264e-06,
      "loss": 0.4125767707824707,
      "num_input_tokens_seen": 139522080,
      "step": 4375,
      "train_runtime": 11476.2288,
      "train_tokens_per_second": 12157.485
    },
    {
      "epoch": 0.2949097764610827,
      "grad_norm": 0.8801724628849412,
      "learning_rate": 9.473276167163872e-06,
      "loss": 0.40921797752380373,
      "num_input_tokens_seen": 139684512,
      "step": 4380,
      "train_runtime": 11488.6749,
      "train_tokens_per_second": 12158.453
    },
    {
      "epoch": 0.2952464314570428,
      "grad_norm": 0.817729180466838,
      "learning_rate": 9.472094278958066e-06,
      "loss": 0.436159086227417,
      "num_input_tokens_seen": 139847496,
      "step": 4385,
      "train_runtime": 11501.4919,
      "train_tokens_per_second": 12159.074
    },
    {
      "epoch": 0.29558308645300296,
      "grad_norm": 0.8156745528230597,
      "learning_rate": 9.470911140145353e-06,
      "loss": 0.41635875701904296,
      "num_input_tokens_seen": 140004552,
      "step": 4390,
      "train_runtime": 11514.1639,
      "train_tokens_per_second": 12159.333
    },
    {
      "epoch": 0.2959197414489631,
      "grad_norm": 0.9033147544759869,
      "learning_rate": 9.469726751056599e-06,
      "loss": 0.40703392028808594,
      "num_input_tokens_seen": 140166880,
      "step": 4395,
      "train_runtime": 11527.281,
      "train_tokens_per_second": 12159.579
    },
    {
      "epoch": 0.29625639644492324,
      "grad_norm": 1.0566716332279595,
      "learning_rate": 9.468541112023009e-06,
      "loss": 0.43402862548828125,
      "num_input_tokens_seen": 140323904,
      "step": 4400,
      "train_runtime": 11540.9513,
      "train_tokens_per_second": 12158.781
    },
    {
      "epoch": 0.2965930514408834,
      "grad_norm": 0.8266801927187695,
      "learning_rate": 9.467354223376149e-06,
      "loss": 0.41649560928344725,
      "num_input_tokens_seen": 140476704,
      "step": 4405,
      "train_runtime": 11554.866,
      "train_tokens_per_second": 12157.363
    },
    {
      "epoch": 0.2969297064368435,
      "grad_norm": 0.8412248467398555,
      "learning_rate": 9.466166085447923e-06,
      "loss": 0.39827594757080076,
      "num_input_tokens_seen": 140634016,
      "step": 4410,
      "train_runtime": 11568.2122,
      "train_tokens_per_second": 12156.936
    },
    {
      "epoch": 0.29726636143280366,
      "grad_norm": 0.856232253356463,
      "learning_rate": 9.464976698570595e-06,
      "loss": 0.3855783939361572,
      "num_input_tokens_seen": 140784784,
      "step": 4415,
      "train_runtime": 11582.342,
      "train_tokens_per_second": 12155.122
    },
    {
      "epoch": 0.2976030164287638,
      "grad_norm": 0.8481816554616582,
      "learning_rate": 9.463786063076769e-06,
      "loss": 0.43189396858215334,
      "num_input_tokens_seen": 140943328,
      "step": 4420,
      "train_runtime": 11595.3464,
      "train_tokens_per_second": 12155.163
    },
    {
      "epoch": 0.29793967142472394,
      "grad_norm": 0.8715209931650221,
      "learning_rate": 9.462594179299408e-06,
      "loss": 0.34606108665466306,
      "num_input_tokens_seen": 141102432,
      "step": 4425,
      "train_runtime": 11608.3572,
      "train_tokens_per_second": 12155.246
    },
    {
      "epoch": 0.2982763264206841,
      "grad_norm": 0.8899438755819059,
      "learning_rate": 9.461401047571811e-06,
      "loss": 0.3818617343902588,
      "num_input_tokens_seen": 141265112,
      "step": 4430,
      "train_runtime": 11621.3112,
      "train_tokens_per_second": 12155.695
    },
    {
      "epoch": 0.2986129814166442,
      "grad_norm": 0.8002666354318052,
      "learning_rate": 9.460206668227639e-06,
      "loss": 0.41319904327392576,
      "num_input_tokens_seen": 141428840,
      "step": 4435,
      "train_runtime": 11634.3807,
      "train_tokens_per_second": 12156.112
    },
    {
      "epoch": 0.29894963641260436,
      "grad_norm": 0.7284442701043069,
      "learning_rate": 9.459011041600895e-06,
      "loss": 0.390230131149292,
      "num_input_tokens_seen": 141592680,
      "step": 4440,
      "train_runtime": 11646.8126,
      "train_tokens_per_second": 12157.204
    },
    {
      "epoch": 0.2992862914085645,
      "grad_norm": 0.7891894798493347,
      "learning_rate": 9.457814168025932e-06,
      "loss": 0.37724034786224364,
      "num_input_tokens_seen": 141752448,
      "step": 4445,
      "train_runtime": 11659.6033,
      "train_tokens_per_second": 12157.57
    },
    {
      "epoch": 0.29962294640452464,
      "grad_norm": 0.9299250490290664,
      "learning_rate": 9.456616047837452e-06,
      "loss": 0.41147918701171876,
      "num_input_tokens_seen": 141905464,
      "step": 4450,
      "train_runtime": 11672.6854,
      "train_tokens_per_second": 12157.054
    },
    {
      "epoch": 0.2999596014004848,
      "grad_norm": 0.8540909625650474,
      "learning_rate": 9.455416681370506e-06,
      "loss": 0.44285898208618163,
      "num_input_tokens_seen": 142066760,
      "step": 4455,
      "train_runtime": 11685.7211,
      "train_tokens_per_second": 12157.295
    },
    {
      "epoch": 0.3002962563964449,
      "grad_norm": 1.4763148051228883,
      "learning_rate": 9.45421606896049e-06,
      "loss": 0.42873735427856446,
      "num_input_tokens_seen": 142229688,
      "step": 4460,
      "train_runtime": 11698.4995,
      "train_tokens_per_second": 12157.943
    },
    {
      "epoch": 0.30063291139240506,
      "grad_norm": 0.8497188670993348,
      "learning_rate": 9.453014210943155e-06,
      "loss": 0.4182133674621582,
      "num_input_tokens_seen": 142391240,
      "step": 4465,
      "train_runtime": 11711.0648,
      "train_tokens_per_second": 12158.693
    },
    {
      "epoch": 0.3009695663883652,
      "grad_norm": 0.8082557462294843,
      "learning_rate": 9.451811107654596e-06,
      "loss": 0.39871602058410643,
      "num_input_tokens_seen": 142552992,
      "step": 4470,
      "train_runtime": 11723.6574,
      "train_tokens_per_second": 12159.43
    },
    {
      "epoch": 0.30130622138432533,
      "grad_norm": 0.9053717388781125,
      "learning_rate": 9.450606759431255e-06,
      "loss": 0.4201824188232422,
      "num_input_tokens_seen": 142714592,
      "step": 4475,
      "train_runtime": 11736.1645,
      "train_tokens_per_second": 12160.241
    },
    {
      "epoch": 0.3016428763802855,
      "grad_norm": 1.1379930610932143,
      "learning_rate": 9.449401166609928e-06,
      "loss": 0.43520140647888184,
      "num_input_tokens_seen": 142872288,
      "step": 4480,
      "train_runtime": 11750.0031,
      "train_tokens_per_second": 12159.34
    },
    {
      "epoch": 0.3019795313762456,
      "grad_norm": 0.8193985705001284,
      "learning_rate": 9.448194329527752e-06,
      "loss": 0.41699957847595215,
      "num_input_tokens_seen": 143030336,
      "step": 4485,
      "train_runtime": 11763.089,
      "train_tokens_per_second": 12159.25
    },
    {
      "epoch": 0.30231618637220575,
      "grad_norm": 0.7565937337060026,
      "learning_rate": 9.446986248522216e-06,
      "loss": 0.3980448246002197,
      "num_input_tokens_seen": 143188152,
      "step": 4490,
      "train_runtime": 11776.1576,
      "train_tokens_per_second": 12159.157
    },
    {
      "epoch": 0.3026528413681659,
      "grad_norm": 0.7855961183604351,
      "learning_rate": 9.445776923931157e-06,
      "loss": 0.38038811683654783,
      "num_input_tokens_seen": 143341280,
      "step": 4495,
      "train_runtime": 11788.8192,
      "train_tokens_per_second": 12159.087
    },
    {
      "epoch": 0.30298949636412603,
      "grad_norm": 0.9074275710891186,
      "learning_rate": 9.444566356092754e-06,
      "loss": 0.4193134307861328,
      "num_input_tokens_seen": 143503248,
      "step": 4500,
      "train_runtime": 11802.6779,
      "train_tokens_per_second": 12158.533
    },
    {
      "epoch": 0.30332615136008617,
      "grad_norm": 0.8561921751722225,
      "learning_rate": 9.443354545345545e-06,
      "loss": 0.3744628429412842,
      "num_input_tokens_seen": 143663264,
      "step": 4505,
      "train_runtime": 11815.8114,
      "train_tokens_per_second": 12158.561
    },
    {
      "epoch": 0.3036628063560463,
      "grad_norm": 0.6971244335977618,
      "learning_rate": 9.442141492028406e-06,
      "loss": 0.3868942022323608,
      "num_input_tokens_seen": 143826192,
      "step": 4510,
      "train_runtime": 11829.5405,
      "train_tokens_per_second": 12158.223
    },
    {
      "epoch": 0.30399946135200645,
      "grad_norm": 0.9156373436448563,
      "learning_rate": 9.440927196480563e-06,
      "loss": 0.397089409828186,
      "num_input_tokens_seen": 143987960,
      "step": 4515,
      "train_runtime": 11842.3872,
      "train_tokens_per_second": 12158.694
    },
    {
      "epoch": 0.3043361163479666,
      "grad_norm": 0.7312761786649533,
      "learning_rate": 9.439711659041593e-06,
      "loss": 0.39508748054504395,
      "num_input_tokens_seen": 144148208,
      "step": 4520,
      "train_runtime": 11855.9172,
      "train_tokens_per_second": 12158.335
    },
    {
      "epoch": 0.30467277134392673,
      "grad_norm": 0.8197796372255761,
      "learning_rate": 9.438494880051413e-06,
      "loss": 0.4180232048034668,
      "num_input_tokens_seen": 144308840,
      "step": 4525,
      "train_runtime": 11868.7404,
      "train_tokens_per_second": 12158.733
    },
    {
      "epoch": 0.30500942633988687,
      "grad_norm": 0.8001528057679329,
      "learning_rate": 9.437276859850293e-06,
      "loss": 0.3978069543838501,
      "num_input_tokens_seen": 144469856,
      "step": 4530,
      "train_runtime": 11882.3043,
      "train_tokens_per_second": 12158.404
    },
    {
      "epoch": 0.305346081335847,
      "grad_norm": 0.9047970325185543,
      "learning_rate": 9.436057598778848e-06,
      "loss": 0.4406630516052246,
      "num_input_tokens_seen": 144629648,
      "step": 4535,
      "train_runtime": 11895.832,
      "train_tokens_per_second": 12158.01
    },
    {
      "epoch": 0.30568273633180715,
      "grad_norm": 0.8524924733353111,
      "learning_rate": 9.434837097178043e-06,
      "loss": 0.46436290740966796,
      "num_input_tokens_seen": 144790904,
      "step": 4540,
      "train_runtime": 11909.1159,
      "train_tokens_per_second": 12157.989
    },
    {
      "epoch": 0.3060193913277673,
      "grad_norm": 0.9196616779628041,
      "learning_rate": 9.433615355389183e-06,
      "loss": 0.42666077613830566,
      "num_input_tokens_seen": 144945576,
      "step": 4545,
      "train_runtime": 11922.4772,
      "train_tokens_per_second": 12157.337
    },
    {
      "epoch": 0.3063560463237274,
      "grad_norm": 0.8038992071450518,
      "learning_rate": 9.432392373753926e-06,
      "loss": 0.4093827247619629,
      "num_input_tokens_seen": 145106704,
      "step": 4550,
      "train_runtime": 11935.5079,
      "train_tokens_per_second": 12157.564
    },
    {
      "epoch": 0.30669270131968757,
      "grad_norm": 0.7660744108716969,
      "learning_rate": 9.431168152614278e-06,
      "loss": 0.4056282043457031,
      "num_input_tokens_seen": 145267504,
      "step": 4555,
      "train_runtime": 11948.5275,
      "train_tokens_per_second": 12157.775
    },
    {
      "epoch": 0.3070293563156477,
      "grad_norm": 0.8265055133962735,
      "learning_rate": 9.429942692312585e-06,
      "loss": 0.37420063018798827,
      "num_input_tokens_seen": 145425192,
      "step": 4560,
      "train_runtime": 11961.5804,
      "train_tokens_per_second": 12157.69
    },
    {
      "epoch": 0.30736601131160785,
      "grad_norm": 0.6646715297703406,
      "learning_rate": 9.428715993191546e-06,
      "loss": 0.3812062740325928,
      "num_input_tokens_seen": 145587880,
      "step": 4565,
      "train_runtime": 11975.0302,
      "train_tokens_per_second": 12157.621
    },
    {
      "epoch": 0.307702666307568,
      "grad_norm": 0.8232878462191243,
      "learning_rate": 9.427488055594199e-06,
      "loss": 0.4013862609863281,
      "num_input_tokens_seen": 145746552,
      "step": 4570,
      "train_runtime": 11989.8051,
      "train_tokens_per_second": 12155.873
    },
    {
      "epoch": 0.3080393213035281,
      "grad_norm": 0.9078231146166715,
      "learning_rate": 9.426258879863937e-06,
      "loss": 0.4333981513977051,
      "num_input_tokens_seen": 145910392,
      "step": 4575,
      "train_runtime": 12002.2394,
      "train_tokens_per_second": 12156.931
    },
    {
      "epoch": 0.30837597629948826,
      "grad_norm": 1.0674900960925655,
      "learning_rate": 9.425028466344494e-06,
      "loss": 0.36445639133453367,
      "num_input_tokens_seen": 146067464,
      "step": 4580,
      "train_runtime": 12014.8232,
      "train_tokens_per_second": 12157.271
    },
    {
      "epoch": 0.3087126312954484,
      "grad_norm": 1.174643176821385,
      "learning_rate": 9.42379681537995e-06,
      "loss": 0.3764669895172119,
      "num_input_tokens_seen": 146226792,
      "step": 4585,
      "train_runtime": 12027.5759,
      "train_tokens_per_second": 12157.628
    },
    {
      "epoch": 0.30904928629140854,
      "grad_norm": 0.9052483555370321,
      "learning_rate": 9.422563927314732e-06,
      "loss": 0.4020869731903076,
      "num_input_tokens_seen": 146385584,
      "step": 4590,
      "train_runtime": 12040.2597,
      "train_tokens_per_second": 12158.009
    },
    {
      "epoch": 0.3093859412873687,
      "grad_norm": 0.8941761904858248,
      "learning_rate": 9.421329802493615e-06,
      "loss": 0.3765645742416382,
      "num_input_tokens_seen": 146546272,
      "step": 4595,
      "train_runtime": 12053.0465,
      "train_tokens_per_second": 12158.442
    },
    {
      "epoch": 0.3097225962833288,
      "grad_norm": 0.8172363692615738,
      "learning_rate": 9.420094441261717e-06,
      "loss": 0.3999391317367554,
      "num_input_tokens_seen": 146706520,
      "step": 4600,
      "train_runtime": 12065.951,
      "train_tokens_per_second": 12158.72
    },
    {
      "epoch": 0.31005925127928896,
      "grad_norm": 0.8654481893727304,
      "learning_rate": 9.418857843964506e-06,
      "loss": 0.3883073806762695,
      "num_input_tokens_seen": 146863528,
      "step": 4605,
      "train_runtime": 12078.8423,
      "train_tokens_per_second": 12158.742
    },
    {
      "epoch": 0.3103959062752491,
      "grad_norm": 0.7922873475473806,
      "learning_rate": 9.417620010947786e-06,
      "loss": 0.42149972915649414,
      "num_input_tokens_seen": 147017824,
      "step": 4610,
      "train_runtime": 12091.4001,
      "train_tokens_per_second": 12158.875
    },
    {
      "epoch": 0.31073256127120924,
      "grad_norm": 0.7773461059949922,
      "learning_rate": 9.416380942557719e-06,
      "loss": 0.3709684371948242,
      "num_input_tokens_seen": 147180312,
      "step": 4615,
      "train_runtime": 12104.3074,
      "train_tokens_per_second": 12159.334
    },
    {
      "epoch": 0.3110692162671694,
      "grad_norm": 1.3474794135025365,
      "learning_rate": 9.415140639140803e-06,
      "loss": 0.4084783554077148,
      "num_input_tokens_seen": 147336656,
      "step": 4620,
      "train_runtime": 12117.9338,
      "train_tokens_per_second": 12158.563
    },
    {
      "epoch": 0.3114058712631295,
      "grad_norm": 0.7998900902540631,
      "learning_rate": 9.413899101043887e-06,
      "loss": 0.4386431694030762,
      "num_input_tokens_seen": 147491840,
      "step": 4625,
      "train_runtime": 12131.2798,
      "train_tokens_per_second": 12157.979
    },
    {
      "epoch": 0.31174252625908966,
      "grad_norm": 0.9229934872160875,
      "learning_rate": 9.412656328614162e-06,
      "loss": 0.4202288627624512,
      "num_input_tokens_seen": 147649512,
      "step": 4630,
      "train_runtime": 12144.4037,
      "train_tokens_per_second": 12157.823
    },
    {
      "epoch": 0.3120791812550498,
      "grad_norm": 0.8744546564002237,
      "learning_rate": 9.411412322199165e-06,
      "loss": 0.41219310760498046,
      "num_input_tokens_seen": 147812056,
      "step": 4635,
      "train_runtime": 12158.1718,
      "train_tokens_per_second": 12157.425
    },
    {
      "epoch": 0.31241583625100994,
      "grad_norm": 0.7083917759084929,
      "learning_rate": 9.41016708214678e-06,
      "loss": 0.4056905746459961,
      "num_input_tokens_seen": 147970528,
      "step": 4640,
      "train_runtime": 12171.6511,
      "train_tokens_per_second": 12156.981
    },
    {
      "epoch": 0.31275249124697013,
      "grad_norm": 0.8235496817803302,
      "learning_rate": 9.408920608805233e-06,
      "loss": 0.40030794143676757,
      "num_input_tokens_seen": 148130632,
      "step": 4645,
      "train_runtime": 12184.3441,
      "train_tokens_per_second": 12157.456
    },
    {
      "epoch": 0.31308914624293027,
      "grad_norm": 0.7825366245386461,
      "learning_rate": 9.407672902523097e-06,
      "loss": 0.412338924407959,
      "num_input_tokens_seen": 148293280,
      "step": 4650,
      "train_runtime": 12197.0494,
      "train_tokens_per_second": 12158.127
    },
    {
      "epoch": 0.3134258012388904,
      "grad_norm": 0.8335452999443579,
      "learning_rate": 9.406423963649287e-06,
      "loss": 0.402410888671875,
      "num_input_tokens_seen": 148450520,
      "step": 4655,
      "train_runtime": 12209.7691,
      "train_tokens_per_second": 12158.34
    },
    {
      "epoch": 0.31376245623485055,
      "grad_norm": 0.7840629327492388,
      "learning_rate": 9.405173792533069e-06,
      "loss": 0.3869786262512207,
      "num_input_tokens_seen": 148612952,
      "step": 4660,
      "train_runtime": 12222.3325,
      "train_tokens_per_second": 12159.132
    },
    {
      "epoch": 0.3140991112308107,
      "grad_norm": 0.7910144308714715,
      "learning_rate": 9.403922389524045e-06,
      "loss": 0.4036257743835449,
      "num_input_tokens_seen": 148769664,
      "step": 4665,
      "train_runtime": 12234.9384,
      "train_tokens_per_second": 12159.413
    },
    {
      "epoch": 0.31443576622677083,
      "grad_norm": 0.833877535981925,
      "learning_rate": 9.40266975497217e-06,
      "loss": 0.4103482723236084,
      "num_input_tokens_seen": 148926896,
      "step": 4670,
      "train_runtime": 12248.1091,
      "train_tokens_per_second": 12159.174
    },
    {
      "epoch": 0.31477242122273097,
      "grad_norm": 0.8485970901895429,
      "learning_rate": 9.401415889227736e-06,
      "loss": 0.42368435859680176,
      "num_input_tokens_seen": 149090048,
      "step": 4675,
      "train_runtime": 12261.1324,
      "train_tokens_per_second": 12159.566
    },
    {
      "epoch": 0.3151090762186911,
      "grad_norm": 0.7825259543818295,
      "learning_rate": 9.400160792641385e-06,
      "loss": 0.4236462593078613,
      "num_input_tokens_seen": 149253888,
      "step": 4680,
      "train_runtime": 12273.5615,
      "train_tokens_per_second": 12160.601
    },
    {
      "epoch": 0.31544573121465125,
      "grad_norm": 0.769365701722422,
      "learning_rate": 9.398904465564097e-06,
      "loss": 0.40953617095947265,
      "num_input_tokens_seen": 149413600,
      "step": 4685,
      "train_runtime": 12286.7741,
      "train_tokens_per_second": 12160.523
    },
    {
      "epoch": 0.3157823862106114,
      "grad_norm": 0.7674338149775708,
      "learning_rate": 9.397646908347203e-06,
      "loss": 0.3950378894805908,
      "num_input_tokens_seen": 149570592,
      "step": 4690,
      "train_runtime": 12299.472,
      "train_tokens_per_second": 12160.733
    },
    {
      "epoch": 0.3161190412065715,
      "grad_norm": 1.5817771573879629,
      "learning_rate": 9.396388121342375e-06,
      "loss": 0.37178051471710205,
      "num_input_tokens_seen": 149734192,
      "step": 4695,
      "train_runtime": 12312.522,
      "train_tokens_per_second": 12161.131
    },
    {
      "epoch": 0.31645569620253167,
      "grad_norm": 0.9386685224889469,
      "learning_rate": 9.395128104901628e-06,
      "loss": 0.37784833908081056,
      "num_input_tokens_seen": 149890656,
      "step": 4700,
      "train_runtime": 12325.5045,
      "train_tokens_per_second": 12161.016
    },
    {
      "epoch": 0.3167923511984918,
      "grad_norm": 0.8455532511447771,
      "learning_rate": 9.393866859377321e-06,
      "loss": 0.43805780410766604,
      "num_input_tokens_seen": 150051288,
      "step": 4705,
      "train_runtime": 12338.4461,
      "train_tokens_per_second": 12161.279
    },
    {
      "epoch": 0.31712900619445195,
      "grad_norm": 0.8408644125122793,
      "learning_rate": 9.392604385122157e-06,
      "loss": 0.42035617828369143,
      "num_input_tokens_seen": 150213736,
      "step": 4710,
      "train_runtime": 12351.1061,
      "train_tokens_per_second": 12161.966
    },
    {
      "epoch": 0.3174656611904121,
      "grad_norm": 0.7855649433639134,
      "learning_rate": 9.391340682489185e-06,
      "loss": 0.3809244155883789,
      "num_input_tokens_seen": 150371992,
      "step": 4715,
      "train_runtime": 12364.4079,
      "train_tokens_per_second": 12161.682
    },
    {
      "epoch": 0.3178023161863722,
      "grad_norm": 0.9612928324671719,
      "learning_rate": 9.390075751831794e-06,
      "loss": 0.42129902839660643,
      "num_input_tokens_seen": 150532456,
      "step": 4720,
      "train_runtime": 12377.7085,
      "train_tokens_per_second": 12161.577
    },
    {
      "epoch": 0.31813897118233236,
      "grad_norm": 0.916937940209648,
      "learning_rate": 9.388809593503718e-06,
      "loss": 0.4207742214202881,
      "num_input_tokens_seen": 150691232,
      "step": 4725,
      "train_runtime": 12390.3201,
      "train_tokens_per_second": 12162.013
    },
    {
      "epoch": 0.3184756261782925,
      "grad_norm": 0.9869853480917334,
      "learning_rate": 9.387542207859034e-06,
      "loss": 0.46114530563354494,
      "num_input_tokens_seen": 150849560,
      "step": 4730,
      "train_runtime": 12403.2652,
      "train_tokens_per_second": 12162.085
    },
    {
      "epoch": 0.31881228117425264,
      "grad_norm": 0.8468363939082243,
      "learning_rate": 9.386273595252161e-06,
      "loss": 0.4186831474304199,
      "num_input_tokens_seen": 151011472,
      "step": 4735,
      "train_runtime": 12415.8054,
      "train_tokens_per_second": 12162.841
    },
    {
      "epoch": 0.3191489361702128,
      "grad_norm": 0.8779763566849437,
      "learning_rate": 9.385003756037865e-06,
      "loss": 0.4159707069396973,
      "num_input_tokens_seen": 151172248,
      "step": 4740,
      "train_runtime": 12428.681,
      "train_tokens_per_second": 12163.177
    },
    {
      "epoch": 0.3194855911661729,
      "grad_norm": 1.0800308048510612,
      "learning_rate": 9.383732690571253e-06,
      "loss": 0.4065821647644043,
      "num_input_tokens_seen": 151336088,
      "step": 4745,
      "train_runtime": 12441.1123,
      "train_tokens_per_second": 12164.193
    },
    {
      "epoch": 0.31982224616213306,
      "grad_norm": 0.7787481334432229,
      "learning_rate": 9.382460399207769e-06,
      "loss": 0.41455612182617185,
      "num_input_tokens_seen": 151497904,
      "step": 4750,
      "train_runtime": 12453.6968,
      "train_tokens_per_second": 12164.894
    },
    {
      "epoch": 0.3201589011580932,
      "grad_norm": 0.7989212668442146,
      "learning_rate": 9.381186882303212e-06,
      "loss": 0.38765130043029783,
      "num_input_tokens_seen": 151647912,
      "step": 4755,
      "train_runtime": 12466.2562,
      "train_tokens_per_second": 12164.672
    },
    {
      "epoch": 0.32049555615405334,
      "grad_norm": 0.8564718229985108,
      "learning_rate": 9.379912140213713e-06,
      "loss": 0.4071049690246582,
      "num_input_tokens_seen": 151802808,
      "step": 4760,
      "train_runtime": 12479.2361,
      "train_tokens_per_second": 12164.431
    },
    {
      "epoch": 0.3208322111500135,
      "grad_norm": 0.7406210590047412,
      "learning_rate": 9.37863617329575e-06,
      "loss": 0.38042144775390624,
      "num_input_tokens_seen": 151963024,
      "step": 4765,
      "train_runtime": 12493.3746,
      "train_tokens_per_second": 12163.489
    },
    {
      "epoch": 0.3211688661459736,
      "grad_norm": 0.9020902505405521,
      "learning_rate": 9.377358981906145e-06,
      "loss": 0.3874586343765259,
      "num_input_tokens_seen": 152121440,
      "step": 4770,
      "train_runtime": 12506.1342,
      "train_tokens_per_second": 12163.746
    },
    {
      "epoch": 0.32150552114193376,
      "grad_norm": 0.7649066199895285,
      "learning_rate": 9.376080566402059e-06,
      "loss": 0.4204458713531494,
      "num_input_tokens_seen": 152285280,
      "step": 4775,
      "train_runtime": 12518.5723,
      "train_tokens_per_second": 12164.748
    },
    {
      "epoch": 0.3218421761378939,
      "grad_norm": 0.7384489913772869,
      "learning_rate": 9.374800927140994e-06,
      "loss": 0.3313518285751343,
      "num_input_tokens_seen": 152448192,
      "step": 4780,
      "train_runtime": 12531.5765,
      "train_tokens_per_second": 12165.125
    },
    {
      "epoch": 0.32217883113385404,
      "grad_norm": 0.8386048485384894,
      "learning_rate": 9.373520064480804e-06,
      "loss": 0.38408641815185546,
      "num_input_tokens_seen": 152610000,
      "step": 4785,
      "train_runtime": 12544.1684,
      "train_tokens_per_second": 12165.812
    },
    {
      "epoch": 0.3225154861298142,
      "grad_norm": 0.824439348353975,
      "learning_rate": 9.372237978779672e-06,
      "loss": 0.4035375118255615,
      "num_input_tokens_seen": 152772720,
      "step": 4790,
      "train_runtime": 12556.6327,
      "train_tokens_per_second": 12166.695
    },
    {
      "epoch": 0.3228521411257743,
      "grad_norm": 0.8334478415115991,
      "learning_rate": 9.37095467039613e-06,
      "loss": 0.3929436206817627,
      "num_input_tokens_seen": 152926368,
      "step": 4795,
      "train_runtime": 12569.2435,
      "train_tokens_per_second": 12166.712
    },
    {
      "epoch": 0.32318879612173446,
      "grad_norm": 0.8072416528920171,
      "learning_rate": 9.369670139689056e-06,
      "loss": 0.3829780340194702,
      "num_input_tokens_seen": 153084776,
      "step": 4800,
      "train_runtime": 12582.1322,
      "train_tokens_per_second": 12166.839
    },
    {
      "epoch": 0.3235254511176946,
      "grad_norm": 0.7767829514462551,
      "learning_rate": 9.368384387017659e-06,
      "loss": 0.38355996608734133,
      "num_input_tokens_seen": 153238984,
      "step": 4805,
      "train_runtime": 12596.4786,
      "train_tokens_per_second": 12165.224
    },
    {
      "epoch": 0.32386210611365474,
      "grad_norm": 0.8005623513813104,
      "learning_rate": 9.367097412741497e-06,
      "loss": 0.3950089931488037,
      "num_input_tokens_seen": 153393936,
      "step": 4810,
      "train_runtime": 12609.5254,
      "train_tokens_per_second": 12164.925
    },
    {
      "epoch": 0.3241987611096149,
      "grad_norm": 0.976355566941992,
      "learning_rate": 9.36580921722047e-06,
      "loss": 0.41298389434814453,
      "num_input_tokens_seen": 153556688,
      "step": 4815,
      "train_runtime": 12622.232,
      "train_tokens_per_second": 12165.573
    },
    {
      "epoch": 0.324535416105575,
      "grad_norm": 0.7889158643184929,
      "learning_rate": 9.364519800814818e-06,
      "loss": 0.3670580625534058,
      "num_input_tokens_seen": 153719616,
      "step": 4820,
      "train_runtime": 12634.6591,
      "train_tokens_per_second": 12166.503
    },
    {
      "epoch": 0.32487207110153515,
      "grad_norm": 0.6594362745631708,
      "learning_rate": 9.36322916388512e-06,
      "loss": 0.3989641904830933,
      "num_input_tokens_seen": 153880488,
      "step": 4825,
      "train_runtime": 12647.1236,
      "train_tokens_per_second": 12167.232
    },
    {
      "epoch": 0.3252087260974953,
      "grad_norm": 0.7912959674385803,
      "learning_rate": 9.3619373067923e-06,
      "loss": 0.36187832355499266,
      "num_input_tokens_seen": 154042224,
      "step": 4830,
      "train_runtime": 12659.6591,
      "train_tokens_per_second": 12167.96
    },
    {
      "epoch": 0.32554538109345543,
      "grad_norm": 0.7861261637169029,
      "learning_rate": 9.360644229897622e-06,
      "loss": 0.4024681091308594,
      "num_input_tokens_seen": 154201664,
      "step": 4835,
      "train_runtime": 12672.5838,
      "train_tokens_per_second": 12168.131
    },
    {
      "epoch": 0.32588203608941557,
      "grad_norm": 1.7371560857219417,
      "learning_rate": 9.35934993356269e-06,
      "loss": 0.42070827484130857,
      "num_input_tokens_seen": 154360976,
      "step": 4840,
      "train_runtime": 12685.9092,
      "train_tokens_per_second": 12167.908
    },
    {
      "epoch": 0.3262186910853757,
      "grad_norm": 0.9320942600152324,
      "learning_rate": 9.35805441814945e-06,
      "loss": 0.35839457511901857,
      "num_input_tokens_seen": 154521192,
      "step": 4845,
      "train_runtime": 12698.3262,
      "train_tokens_per_second": 12168.627
    },
    {
      "epoch": 0.32655534608133585,
      "grad_norm": 0.879351394336123,
      "learning_rate": 9.356757684020188e-06,
      "loss": 0.42413878440856934,
      "num_input_tokens_seen": 154676968,
      "step": 4850,
      "train_runtime": 12711.4483,
      "train_tokens_per_second": 12168.32
    },
    {
      "epoch": 0.326892001077296,
      "grad_norm": 0.907509715514265,
      "learning_rate": 9.355459731537533e-06,
      "loss": 0.3926476240158081,
      "num_input_tokens_seen": 154836584,
      "step": 4855,
      "train_runtime": 12724.2091,
      "train_tokens_per_second": 12168.661
    },
    {
      "epoch": 0.32722865607325613,
      "grad_norm": 0.9458456785324451,
      "learning_rate": 9.354160561064451e-06,
      "loss": 0.41061720848083494,
      "num_input_tokens_seen": 154997536,
      "step": 4860,
      "train_runtime": 12737.7757,
      "train_tokens_per_second": 12168.336
    },
    {
      "epoch": 0.32756531106921627,
      "grad_norm": 0.8010538144988205,
      "learning_rate": 9.352860172964254e-06,
      "loss": 0.43629021644592286,
      "num_input_tokens_seen": 155158880,
      "step": 4865,
      "train_runtime": 12750.805,
      "train_tokens_per_second": 12168.556
    },
    {
      "epoch": 0.3279019660651764,
      "grad_norm": 0.8690731601875019,
      "learning_rate": 9.35155856760059e-06,
      "loss": 0.35876505374908446,
      "num_input_tokens_seen": 155321408,
      "step": 4870,
      "train_runtime": 12763.4917,
      "train_tokens_per_second": 12169.194
    },
    {
      "epoch": 0.32823862106113655,
      "grad_norm": 0.7890294783649087,
      "learning_rate": 9.35025574533745e-06,
      "loss": 0.41234002113342283,
      "num_input_tokens_seen": 155483600,
      "step": 4875,
      "train_runtime": 12776.1295,
      "train_tokens_per_second": 12169.852
    },
    {
      "epoch": 0.3285752760570967,
      "grad_norm": 0.8823416631189168,
      "learning_rate": 9.34895170653916e-06,
      "loss": 0.39826986789703367,
      "num_input_tokens_seen": 155644456,
      "step": 4880,
      "train_runtime": 12789.2606,
      "train_tokens_per_second": 12169.934
    },
    {
      "epoch": 0.3289119310530568,
      "grad_norm": 0.7832857690809821,
      "learning_rate": 9.347646451570394e-06,
      "loss": 0.3966914176940918,
      "num_input_tokens_seen": 155803504,
      "step": 4885,
      "train_runtime": 12802.5671,
      "train_tokens_per_second": 12169.708
    },
    {
      "epoch": 0.32924858604901697,
      "grad_norm": 0.840055873131497,
      "learning_rate": 9.346339980796162e-06,
      "loss": 0.43817987442016604,
      "num_input_tokens_seen": 155966968,
      "step": 4890,
      "train_runtime": 12815.6307,
      "train_tokens_per_second": 12170.058
    },
    {
      "epoch": 0.3295852410449771,
      "grad_norm": 0.8014174183838414,
      "learning_rate": 9.345032294581813e-06,
      "loss": 0.41841840744018555,
      "num_input_tokens_seen": 156126984,
      "step": 4895,
      "train_runtime": 12828.4869,
      "train_tokens_per_second": 12170.335
    },
    {
      "epoch": 0.32992189604093725,
      "grad_norm": 0.804468546314331,
      "learning_rate": 9.343723393293038e-06,
      "loss": 0.3968712568283081,
      "num_input_tokens_seen": 156287560,
      "step": 4900,
      "train_runtime": 12842.0225,
      "train_tokens_per_second": 12170.011
    },
    {
      "epoch": 0.3302585510368974,
      "grad_norm": 0.9095252272234827,
      "learning_rate": 9.342413277295869e-06,
      "loss": 0.40456581115722656,
      "num_input_tokens_seen": 156447272,
      "step": 4905,
      "train_runtime": 12854.5586,
      "train_tokens_per_second": 12170.567
    },
    {
      "epoch": 0.3305952060328575,
      "grad_norm": 0.7203551553891038,
      "learning_rate": 9.341101946956672e-06,
      "loss": 0.3821167230606079,
      "num_input_tokens_seen": 156607864,
      "step": 4910,
      "train_runtime": 12868.8587,
      "train_tokens_per_second": 12169.522
    },
    {
      "epoch": 0.33093186102881766,
      "grad_norm": 0.80282151693793,
      "learning_rate": 9.33978940264216e-06,
      "loss": 0.42123165130615237,
      "num_input_tokens_seen": 156765504,
      "step": 4915,
      "train_runtime": 12883.3998,
      "train_tokens_per_second": 12168.023
    },
    {
      "epoch": 0.3312685160247778,
      "grad_norm": 0.9697382535598529,
      "learning_rate": 9.338475644719377e-06,
      "loss": 0.4235970497131348,
      "num_input_tokens_seen": 156927656,
      "step": 4920,
      "train_runtime": 12895.9771,
      "train_tokens_per_second": 12168.729
    },
    {
      "epoch": 0.33160517102073794,
      "grad_norm": 0.8772457722559294,
      "learning_rate": 9.337160673555719e-06,
      "loss": 0.3926654815673828,
      "num_input_tokens_seen": 157086904,
      "step": 4925,
      "train_runtime": 12909.3567,
      "train_tokens_per_second": 12168.453
    },
    {
      "epoch": 0.3319418260166981,
      "grad_norm": 0.7722093241333533,
      "learning_rate": 9.335844489518905e-06,
      "loss": 0.40118021965026857,
      "num_input_tokens_seen": 157244616,
      "step": 4930,
      "train_runtime": 12922.3594,
      "train_tokens_per_second": 12168.414
    },
    {
      "epoch": 0.3322784810126582,
      "grad_norm": 0.8203445572972957,
      "learning_rate": 9.334527092977008e-06,
      "loss": 0.3770778179168701,
      "num_input_tokens_seen": 157398016,
      "step": 4935,
      "train_runtime": 12935.9395,
      "train_tokens_per_second": 12167.498
    },
    {
      "epoch": 0.33261513600861836,
      "grad_norm": 0.79769655241568,
      "learning_rate": 9.33320848429843e-06,
      "loss": 0.3995029449462891,
      "num_input_tokens_seen": 157561856,
      "step": 4940,
      "train_runtime": 12948.3755,
      "train_tokens_per_second": 12168.465
    },
    {
      "epoch": 0.3329517910045785,
      "grad_norm": 0.815590516356396,
      "learning_rate": 9.331888663851917e-06,
      "loss": 0.40761737823486327,
      "num_input_tokens_seen": 157722800,
      "step": 4945,
      "train_runtime": 12961.1902,
      "train_tokens_per_second": 12168.852
    },
    {
      "epoch": 0.33328844600053864,
      "grad_norm": 0.8978773502517382,
      "learning_rate": 9.330567632006552e-06,
      "loss": 0.4201383590698242,
      "num_input_tokens_seen": 157884016,
      "step": 4950,
      "train_runtime": 12973.6641,
      "train_tokens_per_second": 12169.578
    },
    {
      "epoch": 0.3336251009964988,
      "grad_norm": 0.8681429501018025,
      "learning_rate": 9.329245389131759e-06,
      "loss": 0.4080531597137451,
      "num_input_tokens_seen": 158046352,
      "step": 4955,
      "train_runtime": 12986.8358,
      "train_tokens_per_second": 12169.735
    },
    {
      "epoch": 0.3339617559924589,
      "grad_norm": 0.8239020738381359,
      "learning_rate": 9.327921935597298e-06,
      "loss": 0.3816958427429199,
      "num_input_tokens_seen": 158203704,
      "step": 4960,
      "train_runtime": 12999.6804,
      "train_tokens_per_second": 12169.815
    },
    {
      "epoch": 0.33429841098841906,
      "grad_norm": 0.9343386065339763,
      "learning_rate": 9.326597271773267e-06,
      "loss": 0.3920842885971069,
      "num_input_tokens_seen": 158359960,
      "step": 4965,
      "train_runtime": 13013.0866,
      "train_tokens_per_second": 12169.285
    },
    {
      "epoch": 0.3346350659843792,
      "grad_norm": 0.8200770490423416,
      "learning_rate": 9.325271398030107e-06,
      "loss": 0.4145952701568604,
      "num_input_tokens_seen": 158519944,
      "step": 4970,
      "train_runtime": 13026.2511,
      "train_tokens_per_second": 12169.268
    },
    {
      "epoch": 0.33497172098033934,
      "grad_norm": 0.9053857793792193,
      "learning_rate": 9.323944314738591e-06,
      "loss": 0.40570497512817383,
      "num_input_tokens_seen": 158681360,
      "step": 4975,
      "train_runtime": 13040.1541,
      "train_tokens_per_second": 12168.672
    },
    {
      "epoch": 0.3353083759762995,
      "grad_norm": 0.7996888086126086,
      "learning_rate": 9.322616022269838e-06,
      "loss": 0.3849011421203613,
      "num_input_tokens_seen": 158837128,
      "step": 4980,
      "train_runtime": 13053.0305,
      "train_tokens_per_second": 12168.602
    },
    {
      "epoch": 0.3356450309722596,
      "grad_norm": 0.8483129981929501,
      "learning_rate": 9.321286520995297e-06,
      "loss": 0.40636348724365234,
      "num_input_tokens_seen": 158994224,
      "step": 4985,
      "train_runtime": 13066.5499,
      "train_tokens_per_second": 12168.034
    },
    {
      "epoch": 0.33598168596821976,
      "grad_norm": 0.9507363918325171,
      "learning_rate": 9.319955811286758e-06,
      "loss": 0.3894704818725586,
      "num_input_tokens_seen": 159155960,
      "step": 4990,
      "train_runtime": 13079.096,
      "train_tokens_per_second": 12168.728
    },
    {
      "epoch": 0.3363183409641799,
      "grad_norm": 0.760280975510304,
      "learning_rate": 9.318623893516354e-06,
      "loss": 0.3991784334182739,
      "num_input_tokens_seen": 159313664,
      "step": 4995,
      "train_runtime": 13092.5355,
      "train_tokens_per_second": 12168.282
    },
    {
      "epoch": 0.33665499596014004,
      "grad_norm": 0.8652066070950722,
      "learning_rate": 9.317290768056548e-06,
      "loss": 0.4180765628814697,
      "num_input_tokens_seen": 159469480,
      "step": 5000,
      "train_runtime": 13106.0662,
      "train_tokens_per_second": 12167.608
    },
    {
      "epoch": 0.3369916509561002,
      "grad_norm": 0.7994492721646552,
      "learning_rate": 9.315956435280147e-06,
      "loss": 0.39785919189453123,
      "num_input_tokens_seen": 159624728,
      "step": 5005,
      "train_runtime": 13118.6223,
      "train_tokens_per_second": 12167.797
    },
    {
      "epoch": 0.3373283059520603,
      "grad_norm": 0.8243698850386498,
      "learning_rate": 9.314620895560292e-06,
      "loss": 0.39923934936523436,
      "num_input_tokens_seen": 159781920,
      "step": 5010,
      "train_runtime": 13131.6053,
      "train_tokens_per_second": 12167.737
    },
    {
      "epoch": 0.33766496094802045,
      "grad_norm": 0.9829489011292506,
      "learning_rate": 9.313284149270459e-06,
      "loss": 0.43280954360961915,
      "num_input_tokens_seen": 159941592,
      "step": 5015,
      "train_runtime": 13144.9858,
      "train_tokens_per_second": 12167.498
    },
    {
      "epoch": 0.3380016159439806,
      "grad_norm": 0.8358057172912196,
      "learning_rate": 9.311946196784469e-06,
      "loss": 0.3909532308578491,
      "num_input_tokens_seen": 160100576,
      "step": 5020,
      "train_runtime": 13158.0765,
      "train_tokens_per_second": 12167.476
    },
    {
      "epoch": 0.33833827093994073,
      "grad_norm": 0.7922816876792169,
      "learning_rate": 9.310607038476476e-06,
      "loss": 0.4048126220703125,
      "num_input_tokens_seen": 160257800,
      "step": 5025,
      "train_runtime": 13170.9561,
      "train_tokens_per_second": 12167.515
    },
    {
      "epoch": 0.3386749259359009,
      "grad_norm": 0.8270426201720833,
      "learning_rate": 9.30926667472097e-06,
      "loss": 0.3984680652618408,
      "num_input_tokens_seen": 160419888,
      "step": 5030,
      "train_runtime": 13183.4271,
      "train_tokens_per_second": 12168.299
    },
    {
      "epoch": 0.339011580931861,
      "grad_norm": 0.8461148763888398,
      "learning_rate": 9.307925105892779e-06,
      "loss": 0.4075295448303223,
      "num_input_tokens_seen": 160582976,
      "step": 5035,
      "train_runtime": 13196.2883,
      "train_tokens_per_second": 12168.799
    },
    {
      "epoch": 0.33934823592782115,
      "grad_norm": 0.8641955395243164,
      "learning_rate": 9.30658233236707e-06,
      "loss": 0.40024585723876954,
      "num_input_tokens_seen": 160746816,
      "step": 5040,
      "train_runtime": 13208.7224,
      "train_tokens_per_second": 12169.747
    },
    {
      "epoch": 0.3396848909237813,
      "grad_norm": 0.7877497901050131,
      "learning_rate": 9.30523835451934e-06,
      "loss": 0.3929188966751099,
      "num_input_tokens_seen": 160906304,
      "step": 5045,
      "train_runtime": 13221.2214,
      "train_tokens_per_second": 12170.306
    },
    {
      "epoch": 0.34002154591974143,
      "grad_norm": 0.8559007486668534,
      "learning_rate": 9.303893172725437e-06,
      "loss": 0.43031940460205076,
      "num_input_tokens_seen": 161066264,
      "step": 5050,
      "train_runtime": 13233.6825,
      "train_tokens_per_second": 12170.933
    },
    {
      "epoch": 0.34035820091570157,
      "grad_norm": 0.8386124456749575,
      "learning_rate": 9.302546787361533e-06,
      "loss": 0.4001013278961182,
      "num_input_tokens_seen": 161226864,
      "step": 5055,
      "train_runtime": 13246.5457,
      "train_tokens_per_second": 12171.238
    },
    {
      "epoch": 0.3406948559116617,
      "grad_norm": 0.9832603552616476,
      "learning_rate": 9.301199198804139e-06,
      "loss": 0.38755548000335693,
      "num_input_tokens_seen": 161384936,
      "step": 5060,
      "train_runtime": 13259.7888,
      "train_tokens_per_second": 12171.003
    },
    {
      "epoch": 0.34103151090762185,
      "grad_norm": 0.8415817747022057,
      "learning_rate": 9.299850407430102e-06,
      "loss": 0.43251619338989256,
      "num_input_tokens_seen": 161548576,
      "step": 5065,
      "train_runtime": 13272.9376,
      "train_tokens_per_second": 12171.275
    },
    {
      "epoch": 0.341368165903582,
      "grad_norm": 0.7918084834966653,
      "learning_rate": 9.298500413616613e-06,
      "loss": 0.3998404502868652,
      "num_input_tokens_seen": 161706704,
      "step": 5070,
      "train_runtime": 13285.5914,
      "train_tokens_per_second": 12171.585
    },
    {
      "epoch": 0.3417048208995421,
      "grad_norm": 0.8379729927001233,
      "learning_rate": 9.297149217741188e-06,
      "loss": 0.38343086242675783,
      "num_input_tokens_seen": 161861952,
      "step": 5075,
      "train_runtime": 13298.3515,
      "train_tokens_per_second": 12171.58
    },
    {
      "epoch": 0.34204147589550227,
      "grad_norm": 0.8195439727473155,
      "learning_rate": 9.295796820181688e-06,
      "loss": 0.42993960380554197,
      "num_input_tokens_seen": 162022432,
      "step": 5080,
      "train_runtime": 13311.4181,
      "train_tokens_per_second": 12171.688
    },
    {
      "epoch": 0.3423781308914624,
      "grad_norm": 0.8061875320272616,
      "learning_rate": 9.294443221316305e-06,
      "loss": 0.3996783018112183,
      "num_input_tokens_seen": 162182064,
      "step": 5085,
      "train_runtime": 13324.518,
      "train_tokens_per_second": 12171.702
    },
    {
      "epoch": 0.34271478588742255,
      "grad_norm": 0.731048556024357,
      "learning_rate": 9.293088421523569e-06,
      "loss": 0.4098616600036621,
      "num_input_tokens_seen": 162339840,
      "step": 5090,
      "train_runtime": 13338.381,
      "train_tokens_per_second": 12170.88
    },
    {
      "epoch": 0.3430514408833827,
      "grad_norm": 0.8657234214130419,
      "learning_rate": 9.291732421182346e-06,
      "loss": 0.40361852645874025,
      "num_input_tokens_seen": 162500768,
      "step": 5095,
      "train_runtime": 13351.6498,
      "train_tokens_per_second": 12170.838
    },
    {
      "epoch": 0.3433880958793428,
      "grad_norm": 0.8613102507340347,
      "learning_rate": 9.290375220671837e-06,
      "loss": 0.3931206464767456,
      "num_input_tokens_seen": 162659344,
      "step": 5100,
      "train_runtime": 13365.1727,
      "train_tokens_per_second": 12170.388
    },
    {
      "epoch": 0.34372475087530296,
      "grad_norm": 0.8702383602516465,
      "learning_rate": 9.28901682037158e-06,
      "loss": 0.3695573091506958,
      "num_input_tokens_seen": 162813336,
      "step": 5105,
      "train_runtime": 13378.1977,
      "train_tokens_per_second": 12170.05
    },
    {
      "epoch": 0.3440614058712631,
      "grad_norm": 0.7968940810354391,
      "learning_rate": 9.287657220661442e-06,
      "loss": 0.41938180923461915,
      "num_input_tokens_seen": 162975448,
      "step": 5110,
      "train_runtime": 13391.4901,
      "train_tokens_per_second": 12170.076
    },
    {
      "epoch": 0.34439806086722324,
      "grad_norm": 0.973285452878614,
      "learning_rate": 9.286296421921636e-06,
      "loss": 0.38599095344543455,
      "num_input_tokens_seen": 163130048,
      "step": 5115,
      "train_runtime": 13404.0948,
      "train_tokens_per_second": 12170.165
    },
    {
      "epoch": 0.3447347158631834,
      "grad_norm": 0.8632588369833437,
      "learning_rate": 9.284934424532706e-06,
      "loss": 0.45295009613037107,
      "num_input_tokens_seen": 163287184,
      "step": 5120,
      "train_runtime": 13416.5447,
      "train_tokens_per_second": 12170.584
    },
    {
      "epoch": 0.3450713708591435,
      "grad_norm": 0.7564900214036272,
      "learning_rate": 9.283571228875525e-06,
      "loss": 0.3599048137664795,
      "num_input_tokens_seen": 163440256,
      "step": 5125,
      "train_runtime": 13430.2464,
      "train_tokens_per_second": 12169.565
    },
    {
      "epoch": 0.3454080258551037,
      "grad_norm": 0.8924664211318143,
      "learning_rate": 9.282206835331311e-06,
      "loss": 0.4371681213378906,
      "num_input_tokens_seen": 163601080,
      "step": 5130,
      "train_runtime": 13443.1298,
      "train_tokens_per_second": 12169.865
    },
    {
      "epoch": 0.34574468085106386,
      "grad_norm": 0.7825280868736085,
      "learning_rate": 9.28084124428161e-06,
      "loss": 0.3430314540863037,
      "num_input_tokens_seen": 163760984,
      "step": 5135,
      "train_runtime": 13456.045,
      "train_tokens_per_second": 12170.068
    },
    {
      "epoch": 0.346081335847024,
      "grad_norm": 1.5002090633337872,
      "learning_rate": 9.279474456108305e-06,
      "loss": 0.3737651824951172,
      "num_input_tokens_seen": 163924824,
      "step": 5140,
      "train_runtime": 13468.4932,
      "train_tokens_per_second": 12170.985
    },
    {
      "epoch": 0.34641799084298414,
      "grad_norm": 0.8436504377406616,
      "learning_rate": 9.278106471193615e-06,
      "loss": 0.36956377029418946,
      "num_input_tokens_seen": 164085368,
      "step": 5145,
      "train_runtime": 13482.2317,
      "train_tokens_per_second": 12170.49
    },
    {
      "epoch": 0.3467546458389443,
      "grad_norm": 0.9267689020286888,
      "learning_rate": 9.276737289920093e-06,
      "loss": 0.4128605842590332,
      "num_input_tokens_seen": 164244920,
      "step": 5150,
      "train_runtime": 13494.9426,
      "train_tokens_per_second": 12170.85
    },
    {
      "epoch": 0.3470913008349044,
      "grad_norm": 0.8201254055115633,
      "learning_rate": 9.275366912670624e-06,
      "loss": 0.40582122802734377,
      "num_input_tokens_seen": 164398576,
      "step": 5155,
      "train_runtime": 13507.5812,
      "train_tokens_per_second": 12170.838
    },
    {
      "epoch": 0.34742795583086455,
      "grad_norm": 0.881610349853351,
      "learning_rate": 9.273995339828432e-06,
      "loss": 0.3749147176742554,
      "num_input_tokens_seen": 164556400,
      "step": 5160,
      "train_runtime": 13521.267,
      "train_tokens_per_second": 12170.191
    },
    {
      "epoch": 0.3477646108268247,
      "grad_norm": 0.8786679347737169,
      "learning_rate": 9.272622571777072e-06,
      "loss": 0.4523941993713379,
      "num_input_tokens_seen": 164715704,
      "step": 5165,
      "train_runtime": 13534.2339,
      "train_tokens_per_second": 12170.301
    },
    {
      "epoch": 0.34810126582278483,
      "grad_norm": 0.7689584997649798,
      "learning_rate": 9.271248608900434e-06,
      "loss": 0.33801937103271484,
      "num_input_tokens_seen": 164878472,
      "step": 5170,
      "train_runtime": 13547.822,
      "train_tokens_per_second": 12170.109
    },
    {
      "epoch": 0.348437920818745,
      "grad_norm": 0.9206282375177812,
      "learning_rate": 9.269873451582741e-06,
      "loss": 0.4213468074798584,
      "num_input_tokens_seen": 165037824,
      "step": 5175,
      "train_runtime": 13562.3222,
      "train_tokens_per_second": 12168.847
    },
    {
      "epoch": 0.3487745758147051,
      "grad_norm": 0.8836613417353555,
      "learning_rate": 9.268497100208556e-06,
      "loss": 0.4191449642181396,
      "num_input_tokens_seen": 165195096,
      "step": 5180,
      "train_runtime": 13575.0516,
      "train_tokens_per_second": 12169.022
    },
    {
      "epoch": 0.34911123081066525,
      "grad_norm": 0.8914245007990452,
      "learning_rate": 9.267119555162765e-06,
      "loss": 0.38235011100769045,
      "num_input_tokens_seen": 165358280,
      "step": 5185,
      "train_runtime": 13587.9476,
      "train_tokens_per_second": 12169.482
    },
    {
      "epoch": 0.3494478858066254,
      "grad_norm": 0.8533894512176639,
      "learning_rate": 9.2657408168306e-06,
      "loss": 0.38951256275177004,
      "num_input_tokens_seen": 165517600,
      "step": 5190,
      "train_runtime": 13600.976,
      "train_tokens_per_second": 12169.538
    },
    {
      "epoch": 0.34978454080258553,
      "grad_norm": 0.9022772579251818,
      "learning_rate": 9.264360885597617e-06,
      "loss": 0.4006654262542725,
      "num_input_tokens_seen": 165677344,
      "step": 5195,
      "train_runtime": 13613.4185,
      "train_tokens_per_second": 12170.15
    },
    {
      "epoch": 0.35012119579854567,
      "grad_norm": 0.874376830598592,
      "learning_rate": 9.262979761849709e-06,
      "loss": 0.4045432090759277,
      "num_input_tokens_seen": 165840408,
      "step": 5200,
      "train_runtime": 13626.3327,
      "train_tokens_per_second": 12170.583
    },
    {
      "epoch": 0.3504578507945058,
      "grad_norm": 0.9160254727219028,
      "learning_rate": 9.261597445973106e-06,
      "loss": 0.4306468963623047,
      "num_input_tokens_seen": 166001240,
      "step": 5205,
      "train_runtime": 13638.8719,
      "train_tokens_per_second": 12171.185
    },
    {
      "epoch": 0.35079450579046595,
      "grad_norm": 0.8837276917778313,
      "learning_rate": 9.260213938354365e-06,
      "loss": 0.3968240976333618,
      "num_input_tokens_seen": 166158608,
      "step": 5210,
      "train_runtime": 13651.827,
      "train_tokens_per_second": 12171.163
    },
    {
      "epoch": 0.3511311607864261,
      "grad_norm": 0.7712525171864484,
      "learning_rate": 9.25882923938038e-06,
      "loss": 0.4188511371612549,
      "num_input_tokens_seen": 166315536,
      "step": 5215,
      "train_runtime": 13665.3447,
      "train_tokens_per_second": 12170.607
    },
    {
      "epoch": 0.35146781578238623,
      "grad_norm": 0.9159058917952215,
      "learning_rate": 9.257443349438382e-06,
      "loss": 0.42844762802124026,
      "num_input_tokens_seen": 166475480,
      "step": 5220,
      "train_runtime": 13677.906,
      "train_tokens_per_second": 12171.123
    },
    {
      "epoch": 0.35180447077834637,
      "grad_norm": 0.9075235164525466,
      "learning_rate": 9.256056268915924e-06,
      "loss": 0.411525297164917,
      "num_input_tokens_seen": 166639064,
      "step": 5225,
      "train_runtime": 13690.9142,
      "train_tokens_per_second": 12171.507
    },
    {
      "epoch": 0.3521411257743065,
      "grad_norm": 0.7413431616163271,
      "learning_rate": 9.254667998200906e-06,
      "loss": 0.4199207305908203,
      "num_input_tokens_seen": 166799520,
      "step": 5230,
      "train_runtime": 13703.6821,
      "train_tokens_per_second": 12171.876
    },
    {
      "epoch": 0.35247778077026665,
      "grad_norm": 0.8502337993119097,
      "learning_rate": 9.253278537681548e-06,
      "loss": 0.42876243591308594,
      "num_input_tokens_seen": 166960712,
      "step": 5235,
      "train_runtime": 13716.1955,
      "train_tokens_per_second": 12172.524
    },
    {
      "epoch": 0.3528144357662268,
      "grad_norm": 0.8667935075836979,
      "learning_rate": 9.251887887746407e-06,
      "loss": 0.410980749130249,
      "num_input_tokens_seen": 167123768,
      "step": 5240,
      "train_runtime": 13729.7701,
      "train_tokens_per_second": 12172.365
    },
    {
      "epoch": 0.3531510907621869,
      "grad_norm": 1.0943017680301057,
      "learning_rate": 9.25049604878438e-06,
      "loss": 0.3909928798675537,
      "num_input_tokens_seen": 167285568,
      "step": 5245,
      "train_runtime": 13742.8361,
      "train_tokens_per_second": 12172.565
    },
    {
      "epoch": 0.35348774575814707,
      "grad_norm": 0.855698377539809,
      "learning_rate": 9.249103021184684e-06,
      "loss": 0.41027007102966306,
      "num_input_tokens_seen": 167448152,
      "step": 5250,
      "train_runtime": 13756.9221,
      "train_tokens_per_second": 12171.92
    },
    {
      "epoch": 0.3538244007541072,
      "grad_norm": 0.8949901826951719,
      "learning_rate": 9.24770880533688e-06,
      "loss": 0.4402909278869629,
      "num_input_tokens_seen": 167611920,
      "step": 5255,
      "train_runtime": 13770.0535,
      "train_tokens_per_second": 12172.205
    },
    {
      "epoch": 0.35416105575006734,
      "grad_norm": 0.9934360109662183,
      "learning_rate": 9.24631340163085e-06,
      "loss": 0.43441362380981446,
      "num_input_tokens_seen": 167759112,
      "step": 5260,
      "train_runtime": 13783.3241,
      "train_tokens_per_second": 12171.165
    },
    {
      "epoch": 0.3544977107460275,
      "grad_norm": 0.8656170451196837,
      "learning_rate": 9.244916810456822e-06,
      "loss": 0.3673239231109619,
      "num_input_tokens_seen": 167921080,
      "step": 5265,
      "train_runtime": 13795.7758,
      "train_tokens_per_second": 12171.92
    },
    {
      "epoch": 0.3548343657419876,
      "grad_norm": 0.7529848291338072,
      "learning_rate": 9.24351903220534e-06,
      "loss": 0.40373005867004397,
      "num_input_tokens_seen": 168082592,
      "step": 5270,
      "train_runtime": 13808.747,
      "train_tokens_per_second": 12172.183
    },
    {
      "epoch": 0.35517102073794776,
      "grad_norm": 0.9375049476682222,
      "learning_rate": 9.242120067267295e-06,
      "loss": 0.38245971202850343,
      "num_input_tokens_seen": 168240328,
      "step": 5275,
      "train_runtime": 13821.8773,
      "train_tokens_per_second": 12172.032
    },
    {
      "epoch": 0.3555076757339079,
      "grad_norm": 0.8709047347805172,
      "learning_rate": 9.2407199160339e-06,
      "loss": 0.40095043182373047,
      "num_input_tokens_seen": 168398304,
      "step": 5280,
      "train_runtime": 13835.0664,
      "train_tokens_per_second": 12171.846
    },
    {
      "epoch": 0.35584433072986804,
      "grad_norm": 0.8986209107879686,
      "learning_rate": 9.239318578896701e-06,
      "loss": 0.39379229545593264,
      "num_input_tokens_seen": 168556416,
      "step": 5285,
      "train_runtime": 13848.5603,
      "train_tokens_per_second": 12171.404
    },
    {
      "epoch": 0.3561809857258282,
      "grad_norm": 0.8952588111347618,
      "learning_rate": 9.237916056247583e-06,
      "loss": 0.40959725379943845,
      "num_input_tokens_seen": 168714328,
      "step": 5290,
      "train_runtime": 13861.6497,
      "train_tokens_per_second": 12171.302
    },
    {
      "epoch": 0.3565176407217883,
      "grad_norm": 1.1077750281959418,
      "learning_rate": 9.23651234847875e-06,
      "loss": 0.4346426010131836,
      "num_input_tokens_seen": 168874192,
      "step": 5295,
      "train_runtime": 13875.131,
      "train_tokens_per_second": 12170.998
    },
    {
      "epoch": 0.35685429571774846,
      "grad_norm": 0.8059544405243951,
      "learning_rate": 9.23510745598275e-06,
      "loss": 0.40315923690795896,
      "num_input_tokens_seen": 169026760,
      "step": 5300,
      "train_runtime": 13888.4047,
      "train_tokens_per_second": 12170.351
    },
    {
      "epoch": 0.3571909507137086,
      "grad_norm": 0.6723592906916273,
      "learning_rate": 9.233701379152456e-06,
      "loss": 0.38354921340942383,
      "num_input_tokens_seen": 169186368,
      "step": 5305,
      "train_runtime": 13901.4029,
      "train_tokens_per_second": 12170.453
    },
    {
      "epoch": 0.35752760570966874,
      "grad_norm": 0.8182667707693383,
      "learning_rate": 9.232294118381069e-06,
      "loss": 0.41822118759155275,
      "num_input_tokens_seen": 169346664,
      "step": 5310,
      "train_runtime": 13914.3013,
      "train_tokens_per_second": 12170.691
    },
    {
      "epoch": 0.3578642607056289,
      "grad_norm": 0.808438568114564,
      "learning_rate": 9.230885674062129e-06,
      "loss": 0.38440418243408203,
      "num_input_tokens_seen": 169505032,
      "step": 5315,
      "train_runtime": 13926.8374,
      "train_tokens_per_second": 12171.107
    },
    {
      "epoch": 0.358200915701589,
      "grad_norm": 0.9084068636232191,
      "learning_rate": 9.2294760465895e-06,
      "loss": 0.4449141502380371,
      "num_input_tokens_seen": 169665216,
      "step": 5320,
      "train_runtime": 13939.6295,
      "train_tokens_per_second": 12171.429
    },
    {
      "epoch": 0.35853757069754916,
      "grad_norm": 1.26081903620349,
      "learning_rate": 9.228065236357383e-06,
      "loss": 0.4199626922607422,
      "num_input_tokens_seen": 169824280,
      "step": 5325,
      "train_runtime": 13952.1976,
      "train_tokens_per_second": 12171.866
    },
    {
      "epoch": 0.3588742256935093,
      "grad_norm": 0.7600938530344189,
      "learning_rate": 9.226653243760304e-06,
      "loss": 0.3862861156463623,
      "num_input_tokens_seen": 169983664,
      "step": 5330,
      "train_runtime": 13965.3009,
      "train_tokens_per_second": 12171.858
    },
    {
      "epoch": 0.35921088068946944,
      "grad_norm": 0.8019187098213592,
      "learning_rate": 9.225240069193124e-06,
      "loss": 0.4181028366088867,
      "num_input_tokens_seen": 170147312,
      "step": 5335,
      "train_runtime": 13978.3125,
      "train_tokens_per_second": 12172.235
    },
    {
      "epoch": 0.3595475356854296,
      "grad_norm": 0.9319117172807395,
      "learning_rate": 9.223825713051034e-06,
      "loss": 0.40991716384887694,
      "num_input_tokens_seen": 170304840,
      "step": 5340,
      "train_runtime": 13991.2879,
      "train_tokens_per_second": 12172.206
    },
    {
      "epoch": 0.3598841906813897,
      "grad_norm": 0.8476252757020172,
      "learning_rate": 9.222410175729549e-06,
      "loss": 0.3886315107345581,
      "num_input_tokens_seen": 170467448,
      "step": 5345,
      "train_runtime": 14003.9776,
      "train_tokens_per_second": 12172.788
    },
    {
      "epoch": 0.36022084567734985,
      "grad_norm": 0.7147961669642899,
      "learning_rate": 9.220993457624523e-06,
      "loss": 0.3842186689376831,
      "num_input_tokens_seen": 170627320,
      "step": 5350,
      "train_runtime": 14016.4194,
      "train_tokens_per_second": 12173.389
    },
    {
      "epoch": 0.36055750067331,
      "grad_norm": 0.8959018010745233,
      "learning_rate": 9.219575559132136e-06,
      "loss": 0.42137994766235354,
      "num_input_tokens_seen": 170786160,
      "step": 5355,
      "train_runtime": 14029.6972,
      "train_tokens_per_second": 12173.189
    },
    {
      "epoch": 0.36089415566927013,
      "grad_norm": 0.8706592271679208,
      "learning_rate": 9.2181564806489e-06,
      "loss": 0.37853660583496096,
      "num_input_tokens_seen": 170947008,
      "step": 5360,
      "train_runtime": 14043.8386,
      "train_tokens_per_second": 12172.385
    },
    {
      "epoch": 0.3612308106652303,
      "grad_norm": 0.7741015293316653,
      "learning_rate": 9.216736222571655e-06,
      "loss": 0.37652509212493895,
      "num_input_tokens_seen": 171107304,
      "step": 5365,
      "train_runtime": 14056.5568,
      "train_tokens_per_second": 12172.775
    },
    {
      "epoch": 0.3615674656611904,
      "grad_norm": 0.8747891666388579,
      "learning_rate": 9.215314785297572e-06,
      "loss": 0.43374929428100584,
      "num_input_tokens_seen": 171259032,
      "step": 5370,
      "train_runtime": 14070.4661,
      "train_tokens_per_second": 12171.525
    },
    {
      "epoch": 0.36190412065715055,
      "grad_norm": 0.8248728924065581,
      "learning_rate": 9.21389216922415e-06,
      "loss": 0.3997281789779663,
      "num_input_tokens_seen": 171421592,
      "step": 5375,
      "train_runtime": 14084.1203,
      "train_tokens_per_second": 12171.267
    },
    {
      "epoch": 0.3622407756531107,
      "grad_norm": 1.127273081754096,
      "learning_rate": 9.212468374749222e-06,
      "loss": 0.4404153823852539,
      "num_input_tokens_seen": 171581768,
      "step": 5380,
      "train_runtime": 14097.602,
      "train_tokens_per_second": 12170.99
    },
    {
      "epoch": 0.36257743064907083,
      "grad_norm": 0.8409125355103068,
      "learning_rate": 9.211043402270942e-06,
      "loss": 0.396565580368042,
      "num_input_tokens_seen": 171739688,
      "step": 5385,
      "train_runtime": 14110.4921,
      "train_tokens_per_second": 12171.063
    },
    {
      "epoch": 0.36291408564503097,
      "grad_norm": 0.782349328647705,
      "learning_rate": 9.209617252187806e-06,
      "loss": 0.37641682624816897,
      "num_input_tokens_seen": 171899328,
      "step": 5390,
      "train_runtime": 14124.3435,
      "train_tokens_per_second": 12170.43
    },
    {
      "epoch": 0.3632507406409911,
      "grad_norm": 1.5789828430298494,
      "learning_rate": 9.208189924898627e-06,
      "loss": 0.3491178035736084,
      "num_input_tokens_seen": 172055864,
      "step": 5395,
      "train_runtime": 14137.5415,
      "train_tokens_per_second": 12170.14
    },
    {
      "epoch": 0.36358739563695125,
      "grad_norm": 0.8591137013189266,
      "learning_rate": 9.206761420802554e-06,
      "loss": 0.4361734390258789,
      "num_input_tokens_seen": 172216224,
      "step": 5400,
      "train_runtime": 14150.8361,
      "train_tokens_per_second": 12170.039
    },
    {
      "epoch": 0.3639240506329114,
      "grad_norm": 0.793100446784095,
      "learning_rate": 9.205331740299065e-06,
      "loss": 0.39650869369506836,
      "num_input_tokens_seen": 172373320,
      "step": 5405,
      "train_runtime": 14163.9108,
      "train_tokens_per_second": 12169.896
    },
    {
      "epoch": 0.36426070562887153,
      "grad_norm": 0.7357072663371851,
      "learning_rate": 9.203900883787967e-06,
      "loss": 0.4177663326263428,
      "num_input_tokens_seen": 172532312,
      "step": 5410,
      "train_runtime": 14176.6289,
      "train_tokens_per_second": 12170.193
    },
    {
      "epoch": 0.36459736062483167,
      "grad_norm": 0.9359744188774798,
      "learning_rate": 9.202468851669388e-06,
      "loss": 0.42070827484130857,
      "num_input_tokens_seen": 172694568,
      "step": 5415,
      "train_runtime": 14189.9651,
      "train_tokens_per_second": 12170.19
    },
    {
      "epoch": 0.3649340156207918,
      "grad_norm": 0.9566702448992348,
      "learning_rate": 9.201035644343797e-06,
      "loss": 0.4181356906890869,
      "num_input_tokens_seen": 172852960,
      "step": 5420,
      "train_runtime": 14203.5279,
      "train_tokens_per_second": 12169.72
    },
    {
      "epoch": 0.36527067061675195,
      "grad_norm": 0.857426528144976,
      "learning_rate": 9.199601262211985e-06,
      "loss": 0.41099071502685547,
      "num_input_tokens_seen": 173009896,
      "step": 5425,
      "train_runtime": 14217.8906,
      "train_tokens_per_second": 12168.464
    },
    {
      "epoch": 0.3656073256127121,
      "grad_norm": 0.8090681659423706,
      "learning_rate": 9.19816570567507e-06,
      "loss": 0.40360093116760254,
      "num_input_tokens_seen": 173172528,
      "step": 5430,
      "train_runtime": 14231.2608,
      "train_tokens_per_second": 12168.46
    },
    {
      "epoch": 0.3659439806086722,
      "grad_norm": 0.9198145252434449,
      "learning_rate": 9.196728975134503e-06,
      "loss": 0.40351595878601076,
      "num_input_tokens_seen": 173330768,
      "step": 5435,
      "train_runtime": 14244.5371,
      "train_tokens_per_second": 12168.227
    },
    {
      "epoch": 0.36628063560463237,
      "grad_norm": 0.8871786057809372,
      "learning_rate": 9.195291070992061e-06,
      "loss": 0.40013465881347654,
      "num_input_tokens_seen": 173489168,
      "step": 5440,
      "train_runtime": 14257.3716,
      "train_tokens_per_second": 12168.384
    },
    {
      "epoch": 0.3666172906005925,
      "grad_norm": 0.7591105123533345,
      "learning_rate": 9.193851993649849e-06,
      "loss": 0.3656224489212036,
      "num_input_tokens_seen": 173651608,
      "step": 5445,
      "train_runtime": 14270.0293,
      "train_tokens_per_second": 12168.973
    },
    {
      "epoch": 0.36695394559655264,
      "grad_norm": 0.7917581616134372,
      "learning_rate": 9.1924117435103e-06,
      "loss": 0.3992220640182495,
      "num_input_tokens_seen": 173813640,
      "step": 5450,
      "train_runtime": 14284.594,
      "train_tokens_per_second": 12167.909
    },
    {
      "epoch": 0.3672906005925128,
      "grad_norm": 0.9644535132477472,
      "learning_rate": 9.190970320976176e-06,
      "loss": 0.38292953968048093,
      "num_input_tokens_seen": 173975976,
      "step": 5455,
      "train_runtime": 14297.4913,
      "train_tokens_per_second": 12168.287
    },
    {
      "epoch": 0.3676272555884729,
      "grad_norm": 0.9272476811861222,
      "learning_rate": 9.189527726450565e-06,
      "loss": 0.4454700469970703,
      "num_input_tokens_seen": 174126728,
      "step": 5460,
      "train_runtime": 14311.2685,
      "train_tokens_per_second": 12167.107
    },
    {
      "epoch": 0.36796391058443306,
      "grad_norm": 0.7631388334899264,
      "learning_rate": 9.188083960336885e-06,
      "loss": 0.3804170608520508,
      "num_input_tokens_seen": 174286720,
      "step": 5465,
      "train_runtime": 14324.3185,
      "train_tokens_per_second": 12167.191
    },
    {
      "epoch": 0.3683005655803932,
      "grad_norm": 0.8911542710537709,
      "learning_rate": 9.186639023038879e-06,
      "loss": 0.4201070785522461,
      "num_input_tokens_seen": 174444640,
      "step": 5470,
      "train_runtime": 14336.8437,
      "train_tokens_per_second": 12167.576
    },
    {
      "epoch": 0.36863722057635334,
      "grad_norm": 0.8904517292247853,
      "learning_rate": 9.185192914960622e-06,
      "loss": 0.4242672920227051,
      "num_input_tokens_seen": 174603088,
      "step": 5475,
      "train_runtime": 14350.2797,
      "train_tokens_per_second": 12167.225
    },
    {
      "epoch": 0.3689738755723135,
      "grad_norm": 0.8099982686077623,
      "learning_rate": 9.183745636506511e-06,
      "loss": 0.3790933609008789,
      "num_input_tokens_seen": 174756968,
      "step": 5480,
      "train_runtime": 14363.2812,
      "train_tokens_per_second": 12166.925
    },
    {
      "epoch": 0.3693105305682736,
      "grad_norm": 0.7866468441310667,
      "learning_rate": 9.182297188081274e-06,
      "loss": 0.42302284240722654,
      "num_input_tokens_seen": 174916744,
      "step": 5485,
      "train_runtime": 14376.4425,
      "train_tokens_per_second": 12166.9
    },
    {
      "epoch": 0.36964718556423376,
      "grad_norm": 0.7676797852311467,
      "learning_rate": 9.180847570089964e-06,
      "loss": 0.38768396377563474,
      "num_input_tokens_seen": 175070976,
      "step": 5490,
      "train_runtime": 14389.8927,
      "train_tokens_per_second": 12166.246
    },
    {
      "epoch": 0.3699838405601939,
      "grad_norm": 0.9132375685904397,
      "learning_rate": 9.179396782937965e-06,
      "loss": 0.45098104476928713,
      "num_input_tokens_seen": 175234816,
      "step": 5495,
      "train_runtime": 14402.3244,
      "train_tokens_per_second": 12167.12
    },
    {
      "epoch": 0.37032049555615404,
      "grad_norm": 0.9392819970580462,
      "learning_rate": 9.17794482703098e-06,
      "loss": 0.448504638671875,
      "num_input_tokens_seen": 175394096,
      "step": 5500,
      "train_runtime": 14415.4204,
      "train_tokens_per_second": 12167.116
    },
    {
      "epoch": 0.3706571505521142,
      "grad_norm": 0.8215392485819866,
      "learning_rate": 9.176491702775049e-06,
      "loss": 0.42099428176879883,
      "num_input_tokens_seen": 175551864,
      "step": 5505,
      "train_runtime": 14429.7417,
      "train_tokens_per_second": 12165.974
    },
    {
      "epoch": 0.3709938055480743,
      "grad_norm": 0.8187564645822528,
      "learning_rate": 9.17503741057653e-06,
      "loss": 0.43409528732299807,
      "num_input_tokens_seen": 175713368,
      "step": 5510,
      "train_runtime": 14442.7162,
      "train_tokens_per_second": 12166.227
    },
    {
      "epoch": 0.37133046054403446,
      "grad_norm": 0.7715462826711784,
      "learning_rate": 9.173581950842111e-06,
      "loss": 0.38469316959381106,
      "num_input_tokens_seen": 175869752,
      "step": 5515,
      "train_runtime": 14455.6556,
      "train_tokens_per_second": 12166.155
    },
    {
      "epoch": 0.3716671155399946,
      "grad_norm": 0.962006007111934,
      "learning_rate": 9.172125323978811e-06,
      "loss": 0.42240176200866697,
      "num_input_tokens_seen": 176029104,
      "step": 5520,
      "train_runtime": 14468.8354,
      "train_tokens_per_second": 12166.087
    },
    {
      "epoch": 0.37200377053595474,
      "grad_norm": 0.837968435330365,
      "learning_rate": 9.170667530393966e-06,
      "loss": 0.39805238246917723,
      "num_input_tokens_seen": 176190264,
      "step": 5525,
      "train_runtime": 14481.3117,
      "train_tokens_per_second": 12166.734
    },
    {
      "epoch": 0.3723404255319149,
      "grad_norm": 0.8076745384422905,
      "learning_rate": 9.169208570495247e-06,
      "loss": 0.3894563436508179,
      "num_input_tokens_seen": 176351328,
      "step": 5530,
      "train_runtime": 14494.4014,
      "train_tokens_per_second": 12166.858
    },
    {
      "epoch": 0.372677080527875,
      "grad_norm": 0.7535944618782825,
      "learning_rate": 9.167748444690646e-06,
      "loss": 0.3739210844039917,
      "num_input_tokens_seen": 176506488,
      "step": 5535,
      "train_runtime": 14507.4727,
      "train_tokens_per_second": 12166.591
    },
    {
      "epoch": 0.37301373552383515,
      "grad_norm": 0.8136183852272242,
      "learning_rate": 9.16628715338848e-06,
      "loss": 0.39924371242523193,
      "num_input_tokens_seen": 176665416,
      "step": 5540,
      "train_runtime": 14521.2476,
      "train_tokens_per_second": 12165.994
    },
    {
      "epoch": 0.3733503905197953,
      "grad_norm": 0.7769219837739544,
      "learning_rate": 9.164824696997398e-06,
      "loss": 0.37765989303588865,
      "num_input_tokens_seen": 176819992,
      "step": 5545,
      "train_runtime": 14533.8681,
      "train_tokens_per_second": 12166.066
    },
    {
      "epoch": 0.37368704551575543,
      "grad_norm": 0.7746481378249906,
      "learning_rate": 9.163361075926371e-06,
      "loss": 0.42786331176757814,
      "num_input_tokens_seen": 176981816,
      "step": 5550,
      "train_runtime": 14546.9976,
      "train_tokens_per_second": 12166.209
    },
    {
      "epoch": 0.3740237005117156,
      "grad_norm": 0.7710835899607028,
      "learning_rate": 9.161896290584695e-06,
      "loss": 0.3802783727645874,
      "num_input_tokens_seen": 177143816,
      "step": 5555,
      "train_runtime": 14560.2464,
      "train_tokens_per_second": 12166.265
    },
    {
      "epoch": 0.3743603555076757,
      "grad_norm": 0.8120536606566661,
      "learning_rate": 9.160430341381991e-06,
      "loss": 0.40530972480773925,
      "num_input_tokens_seen": 177299952,
      "step": 5560,
      "train_runtime": 14573.1894,
      "train_tokens_per_second": 12166.174
    },
    {
      "epoch": 0.37469701050363585,
      "grad_norm": 0.8005440309863602,
      "learning_rate": 9.15896322872821e-06,
      "loss": 0.40654287338256834,
      "num_input_tokens_seen": 177457184,
      "step": 5565,
      "train_runtime": 14586.1781,
      "train_tokens_per_second": 12166.119
    },
    {
      "epoch": 0.375033665499596,
      "grad_norm": 0.8336876755861642,
      "learning_rate": 9.157494953033625e-06,
      "loss": 0.4060182571411133,
      "num_input_tokens_seen": 177616256,
      "step": 5570,
      "train_runtime": 14598.7549,
      "train_tokens_per_second": 12166.535
    },
    {
      "epoch": 0.37537032049555613,
      "grad_norm": 0.7971953206236163,
      "learning_rate": 9.156025514708832e-06,
      "loss": 0.40840883255004884,
      "num_input_tokens_seen": 177780096,
      "step": 5575,
      "train_runtime": 14611.1835,
      "train_tokens_per_second": 12167.399
    },
    {
      "epoch": 0.37570697549151627,
      "grad_norm": 0.9032953055211189,
      "learning_rate": 9.154554914164757e-06,
      "loss": 0.40496139526367186,
      "num_input_tokens_seen": 177943936,
      "step": 5580,
      "train_runtime": 14623.6079,
      "train_tokens_per_second": 12168.265
    },
    {
      "epoch": 0.3760436304874764,
      "grad_norm": 0.9269836124316817,
      "learning_rate": 9.153083151812649e-06,
      "loss": 0.3801657438278198,
      "num_input_tokens_seen": 178106792,
      "step": 5585,
      "train_runtime": 14637.0727,
      "train_tokens_per_second": 12168.198
    },
    {
      "epoch": 0.37638028548343655,
      "grad_norm": 0.7861253764502524,
      "learning_rate": 9.151610228064078e-06,
      "loss": 0.4101823329925537,
      "num_input_tokens_seen": 178266120,
      "step": 5590,
      "train_runtime": 14649.5915,
      "train_tokens_per_second": 12168.675
    },
    {
      "epoch": 0.3767169404793967,
      "grad_norm": 0.9021453681705658,
      "learning_rate": 9.150136143330949e-06,
      "loss": 0.4283175468444824,
      "num_input_tokens_seen": 178425568,
      "step": 5595,
      "train_runtime": 14663.3037,
      "train_tokens_per_second": 12168.17
    },
    {
      "epoch": 0.37705359547535683,
      "grad_norm": 0.7770077485391865,
      "learning_rate": 9.14866089802548e-06,
      "loss": 0.38265461921691896,
      "num_input_tokens_seen": 178587480,
      "step": 5600,
      "train_runtime": 14675.7639,
      "train_tokens_per_second": 12168.871
    },
    {
      "epoch": 0.37739025047131697,
      "grad_norm": 0.9395926061437876,
      "learning_rate": 9.147184492560219e-06,
      "loss": 0.37009391784667967,
      "num_input_tokens_seen": 178746008,
      "step": 5605,
      "train_runtime": 14689.0252,
      "train_tokens_per_second": 12168.677
    },
    {
      "epoch": 0.3777269054672771,
      "grad_norm": 0.8455249908859552,
      "learning_rate": 9.14570692734804e-06,
      "loss": 0.3915083408355713,
      "num_input_tokens_seen": 178905264,
      "step": 5610,
      "train_runtime": 14701.6137,
      "train_tokens_per_second": 12169.09
    },
    {
      "epoch": 0.37806356046323725,
      "grad_norm": 0.9041363350213437,
      "learning_rate": 9.144228202802137e-06,
      "loss": 0.4065399169921875,
      "num_input_tokens_seen": 179066360,
      "step": 5615,
      "train_runtime": 14714.6407,
      "train_tokens_per_second": 12169.265
    },
    {
      "epoch": 0.37840021545919744,
      "grad_norm": 1.2263757573882148,
      "learning_rate": 9.142748319336033e-06,
      "loss": 0.3947492837905884,
      "num_input_tokens_seen": 179228208,
      "step": 5620,
      "train_runtime": 14727.1869,
      "train_tokens_per_second": 12169.887
    },
    {
      "epoch": 0.3787368704551576,
      "grad_norm": 0.7951021213093052,
      "learning_rate": 9.141267277363573e-06,
      "loss": 0.3920163631439209,
      "num_input_tokens_seen": 179380480,
      "step": 5625,
      "train_runtime": 14740.6069,
      "train_tokens_per_second": 12169.138
    },
    {
      "epoch": 0.3790735254511177,
      "grad_norm": 0.8174096183896313,
      "learning_rate": 9.139785077298921e-06,
      "loss": 0.3512788534164429,
      "num_input_tokens_seen": 179541312,
      "step": 5630,
      "train_runtime": 14753.6406,
      "train_tokens_per_second": 12169.289
    },
    {
      "epoch": 0.37941018044707786,
      "grad_norm": 0.8466272717055351,
      "learning_rate": 9.138301719556577e-06,
      "loss": 0.4015397071838379,
      "num_input_tokens_seen": 179701144,
      "step": 5635,
      "train_runtime": 14767.1186,
      "train_tokens_per_second": 12169.005
    },
    {
      "epoch": 0.379746835443038,
      "grad_norm": 0.7415804913081718,
      "learning_rate": 9.136817204551352e-06,
      "loss": 0.3865314722061157,
      "num_input_tokens_seen": 179860816,
      "step": 5640,
      "train_runtime": 14779.9261,
      "train_tokens_per_second": 12169.264
    },
    {
      "epoch": 0.38008349043899814,
      "grad_norm": 1.355504995480586,
      "learning_rate": 9.135331532698385e-06,
      "loss": 0.39690232276916504,
      "num_input_tokens_seen": 180019992,
      "step": 5645,
      "train_runtime": 14793.0097,
      "train_tokens_per_second": 12169.261
    },
    {
      "epoch": 0.3804201454349583,
      "grad_norm": 0.7425264963124952,
      "learning_rate": 9.133844704413144e-06,
      "loss": 0.3869367837905884,
      "num_input_tokens_seen": 180177648,
      "step": 5650,
      "train_runtime": 14806.427,
      "train_tokens_per_second": 12168.881
    },
    {
      "epoch": 0.3807568004309184,
      "grad_norm": 0.7552524034699898,
      "learning_rate": 9.132356720111413e-06,
      "loss": 0.4172785758972168,
      "num_input_tokens_seen": 180336960,
      "step": 5655,
      "train_runtime": 14819.5314,
      "train_tokens_per_second": 12168.871
    },
    {
      "epoch": 0.38109345542687856,
      "grad_norm": 0.8076926411735782,
      "learning_rate": 9.1308675802093e-06,
      "loss": 0.38221077919006347,
      "num_input_tokens_seen": 180495360,
      "step": 5660,
      "train_runtime": 14832.4018,
      "train_tokens_per_second": 12168.991
    },
    {
      "epoch": 0.3814301104228387,
      "grad_norm": 0.8504971136032399,
      "learning_rate": 9.129377285123241e-06,
      "loss": 0.39042105674743655,
      "num_input_tokens_seen": 180653216,
      "step": 5665,
      "train_runtime": 14846.115,
      "train_tokens_per_second": 12168.383
    },
    {
      "epoch": 0.38176676541879884,
      "grad_norm": 0.9151089894670424,
      "learning_rate": 9.127885835269996e-06,
      "loss": 0.4058138370513916,
      "num_input_tokens_seen": 180812832,
      "step": 5670,
      "train_runtime": 14859.4374,
      "train_tokens_per_second": 12168.215
    },
    {
      "epoch": 0.382103420414759,
      "grad_norm": 0.8002759244872248,
      "learning_rate": 9.126393231066636e-06,
      "loss": 0.3943216323852539,
      "num_input_tokens_seen": 180971504,
      "step": 5675,
      "train_runtime": 14872.9419,
      "train_tokens_per_second": 12167.835
    },
    {
      "epoch": 0.3824400754107191,
      "grad_norm": 0.7787199320047549,
      "learning_rate": 9.124899472930568e-06,
      "loss": 0.4211256980895996,
      "num_input_tokens_seen": 181132808,
      "step": 5680,
      "train_runtime": 14886.0525,
      "train_tokens_per_second": 12167.954
    },
    {
      "epoch": 0.38277673040667926,
      "grad_norm": 0.8401227616643309,
      "learning_rate": 9.123404561279517e-06,
      "loss": 0.4256914138793945,
      "num_input_tokens_seen": 181289376,
      "step": 5685,
      "train_runtime": 14899.1143,
      "train_tokens_per_second": 12167.796
    },
    {
      "epoch": 0.3831133854026394,
      "grad_norm": 0.8201153390190173,
      "learning_rate": 9.121908496531527e-06,
      "loss": 0.39665195941925047,
      "num_input_tokens_seen": 181448296,
      "step": 5690,
      "train_runtime": 14912.3678,
      "train_tokens_per_second": 12167.638
    },
    {
      "epoch": 0.38345004039859953,
      "grad_norm": 0.7393431449344484,
      "learning_rate": 9.120411279104971e-06,
      "loss": 0.4066039562225342,
      "num_input_tokens_seen": 181608312,
      "step": 5695,
      "train_runtime": 14924.9902,
      "train_tokens_per_second": 12168.069
    },
    {
      "epoch": 0.3837866953945597,
      "grad_norm": 0.8007982770126021,
      "learning_rate": 9.118912909418538e-06,
      "loss": 0.39641437530517576,
      "num_input_tokens_seen": 181767824,
      "step": 5700,
      "train_runtime": 14937.6696,
      "train_tokens_per_second": 12168.419
    },
    {
      "epoch": 0.3841233503905198,
      "grad_norm": 0.8468123415657721,
      "learning_rate": 9.117413387891246e-06,
      "loss": 0.43598108291625975,
      "num_input_tokens_seen": 181927176,
      "step": 5705,
      "train_runtime": 14950.6981,
      "train_tokens_per_second": 12168.474
    },
    {
      "epoch": 0.38446000538647995,
      "grad_norm": 0.8515726932958023,
      "learning_rate": 9.11591271494243e-06,
      "loss": 0.40558586120605467,
      "num_input_tokens_seen": 182086072,
      "step": 5710,
      "train_runtime": 14963.8442,
      "train_tokens_per_second": 12168.402
    },
    {
      "epoch": 0.3847966603824401,
      "grad_norm": 0.8816896294455118,
      "learning_rate": 9.114410890991748e-06,
      "loss": 0.3944713592529297,
      "num_input_tokens_seen": 182246408,
      "step": 5715,
      "train_runtime": 14976.817,
      "train_tokens_per_second": 12168.567
    },
    {
      "epoch": 0.38513331537840023,
      "grad_norm": 0.830910035563101,
      "learning_rate": 9.112907916459177e-06,
      "loss": 0.37573947906494143,
      "num_input_tokens_seen": 182398464,
      "step": 5720,
      "train_runtime": 14990.4181,
      "train_tokens_per_second": 12167.67
    },
    {
      "epoch": 0.38546997037436037,
      "grad_norm": 0.8454436188931626,
      "learning_rate": 9.111403791765025e-06,
      "loss": 0.4055496692657471,
      "num_input_tokens_seen": 182558360,
      "step": 5725,
      "train_runtime": 15003.9494,
      "train_tokens_per_second": 12167.354
    },
    {
      "epoch": 0.3858066253703205,
      "grad_norm": 0.8200948435098427,
      "learning_rate": 9.109898517329914e-06,
      "loss": 0.3865630626678467,
      "num_input_tokens_seen": 182721176,
      "step": 5730,
      "train_runtime": 15017.4338,
      "train_tokens_per_second": 12167.27
    },
    {
      "epoch": 0.38614328036628065,
      "grad_norm": 0.806582487164594,
      "learning_rate": 9.108392093574785e-06,
      "loss": 0.39389605522155763,
      "num_input_tokens_seen": 182879168,
      "step": 5735,
      "train_runtime": 15029.9144,
      "train_tokens_per_second": 12167.679
    },
    {
      "epoch": 0.3864799353622408,
      "grad_norm": 1.0220999968490372,
      "learning_rate": 9.10688452092091e-06,
      "loss": 0.36595332622528076,
      "num_input_tokens_seen": 183036016,
      "step": 5740,
      "train_runtime": 15043.1609,
      "train_tokens_per_second": 12167.391
    },
    {
      "epoch": 0.38681659035820093,
      "grad_norm": 0.9483636654575818,
      "learning_rate": 9.105375799789874e-06,
      "loss": 0.39159164428710935,
      "num_input_tokens_seen": 183192744,
      "step": 5745,
      "train_runtime": 15056.1087,
      "train_tokens_per_second": 12167.337
    },
    {
      "epoch": 0.38715324535416107,
      "grad_norm": 0.9446174214024551,
      "learning_rate": 9.103865930603586e-06,
      "loss": 0.3879040241241455,
      "num_input_tokens_seen": 183352800,
      "step": 5750,
      "train_runtime": 15068.8803,
      "train_tokens_per_second": 12167.646
    },
    {
      "epoch": 0.3874899003501212,
      "grad_norm": 0.8698776344339706,
      "learning_rate": 9.102354913784279e-06,
      "loss": 0.3768143177032471,
      "num_input_tokens_seen": 183513504,
      "step": 5755,
      "train_runtime": 15082.4847,
      "train_tokens_per_second": 12167.326
    },
    {
      "epoch": 0.38782655534608135,
      "grad_norm": 0.7930820769724194,
      "learning_rate": 9.100842749754499e-06,
      "loss": 0.367011570930481,
      "num_input_tokens_seen": 183675944,
      "step": 5760,
      "train_runtime": 15095.1321,
      "train_tokens_per_second": 12167.892
    },
    {
      "epoch": 0.3881632103420415,
      "grad_norm": 0.75049975851598,
      "learning_rate": 9.099329438937122e-06,
      "loss": 0.41629934310913086,
      "num_input_tokens_seen": 183832944,
      "step": 5765,
      "train_runtime": 15108.5489,
      "train_tokens_per_second": 12167.478
    },
    {
      "epoch": 0.3884998653380016,
      "grad_norm": 0.8929032755400228,
      "learning_rate": 9.09781498175534e-06,
      "loss": 0.36588306427001954,
      "num_input_tokens_seen": 183988360,
      "step": 5770,
      "train_runtime": 15121.7871,
      "train_tokens_per_second": 12167.104
    },
    {
      "epoch": 0.38883652033396177,
      "grad_norm": 0.7853231079312388,
      "learning_rate": 9.096299378632666e-06,
      "loss": 0.3768979549407959,
      "num_input_tokens_seen": 184150184,
      "step": 5775,
      "train_runtime": 15134.3328,
      "train_tokens_per_second": 12167.711
    },
    {
      "epoch": 0.3891731753299219,
      "grad_norm": 0.8219082469266189,
      "learning_rate": 9.09478262999293e-06,
      "loss": 0.4181269645690918,
      "num_input_tokens_seen": 184313928,
      "step": 5780,
      "train_runtime": 15147.4113,
      "train_tokens_per_second": 12168.015
    },
    {
      "epoch": 0.38950983032588204,
      "grad_norm": 0.819243053311245,
      "learning_rate": 9.09326473626029e-06,
      "loss": 0.3907138824462891,
      "num_input_tokens_seen": 184474160,
      "step": 5785,
      "train_runtime": 15160.7168,
      "train_tokens_per_second": 12167.905
    },
    {
      "epoch": 0.3898464853218422,
      "grad_norm": 0.87035948515625,
      "learning_rate": 9.091745697859218e-06,
      "loss": 0.4154670715332031,
      "num_input_tokens_seen": 184635832,
      "step": 5790,
      "train_runtime": 15173.2416,
      "train_tokens_per_second": 12168.516
    },
    {
      "epoch": 0.3901831403178023,
      "grad_norm": 0.9080500459010776,
      "learning_rate": 9.090225515214512e-06,
      "loss": 0.40126667022705076,
      "num_input_tokens_seen": 184794224,
      "step": 5795,
      "train_runtime": 15185.8968,
      "train_tokens_per_second": 12168.805
    },
    {
      "epoch": 0.39051979531376246,
      "grad_norm": 0.897892240078457,
      "learning_rate": 9.088704188751281e-06,
      "loss": 0.41934909820556643,
      "num_input_tokens_seen": 184951904,
      "step": 5800,
      "train_runtime": 15199.3745,
      "train_tokens_per_second": 12168.389
    },
    {
      "epoch": 0.3908564503097226,
      "grad_norm": 0.9187883276401101,
      "learning_rate": 9.087181718894963e-06,
      "loss": 0.4034393310546875,
      "num_input_tokens_seen": 185115744,
      "step": 5805,
      "train_runtime": 15211.7969,
      "train_tokens_per_second": 12169.223
    },
    {
      "epoch": 0.39119310530568274,
      "grad_norm": 0.8018265047341445,
      "learning_rate": 9.085658106071309e-06,
      "loss": 0.3991898536682129,
      "num_input_tokens_seen": 185272992,
      "step": 5810,
      "train_runtime": 15225.0177,
      "train_tokens_per_second": 12168.984
    },
    {
      "epoch": 0.3915297603016429,
      "grad_norm": 0.8042124878943258,
      "learning_rate": 9.084133350706394e-06,
      "loss": 0.39764938354492185,
      "num_input_tokens_seen": 185430776,
      "step": 5815,
      "train_runtime": 15238.0174,
      "train_tokens_per_second": 12168.957
    },
    {
      "epoch": 0.391866415297603,
      "grad_norm": 1.1724651938015462,
      "learning_rate": 9.082607453226613e-06,
      "loss": 0.368347692489624,
      "num_input_tokens_seen": 185590712,
      "step": 5820,
      "train_runtime": 15251.7535,
      "train_tokens_per_second": 12168.484
    },
    {
      "epoch": 0.39220307029356316,
      "grad_norm": 0.8206580766339412,
      "learning_rate": 9.081080414058674e-06,
      "loss": 0.4035919666290283,
      "num_input_tokens_seen": 185746704,
      "step": 5825,
      "train_runtime": 15264.759,
      "train_tokens_per_second": 12168.335
    },
    {
      "epoch": 0.3925397252895233,
      "grad_norm": 0.9899489284328178,
      "learning_rate": 9.079552233629612e-06,
      "loss": 0.3828219175338745,
      "num_input_tokens_seen": 185905824,
      "step": 5830,
      "train_runtime": 15277.3724,
      "train_tokens_per_second": 12168.704
    },
    {
      "epoch": 0.39287638028548344,
      "grad_norm": 0.8675174625974806,
      "learning_rate": 9.078022912366776e-06,
      "loss": 0.3968531131744385,
      "num_input_tokens_seen": 186066216,
      "step": 5835,
      "train_runtime": 15290.1079,
      "train_tokens_per_second": 12169.058
    },
    {
      "epoch": 0.3932130352814436,
      "grad_norm": 0.8760718072859601,
      "learning_rate": 9.076492450697835e-06,
      "loss": 0.41472568511962893,
      "num_input_tokens_seen": 186228752,
      "step": 5840,
      "train_runtime": 15303.4815,
      "train_tokens_per_second": 12169.045
    },
    {
      "epoch": 0.3935496902774037,
      "grad_norm": 1.0266589139098974,
      "learning_rate": 9.074960849050782e-06,
      "loss": 0.41312284469604493,
      "num_input_tokens_seen": 186387896,
      "step": 5845,
      "train_runtime": 15316.5546,
      "train_tokens_per_second": 12169.049
    },
    {
      "epoch": 0.39388634527336386,
      "grad_norm": 0.7634786648002566,
      "learning_rate": 9.07342810785392e-06,
      "loss": 0.3797821044921875,
      "num_input_tokens_seen": 186541912,
      "step": 5850,
      "train_runtime": 15330.1905,
      "train_tokens_per_second": 12168.271
    },
    {
      "epoch": 0.394223000269324,
      "grad_norm": 0.844673236007135,
      "learning_rate": 9.071894227535878e-06,
      "loss": 0.377911901473999,
      "num_input_tokens_seen": 186698896,
      "step": 5855,
      "train_runtime": 15342.837,
      "train_tokens_per_second": 12168.473
    },
    {
      "epoch": 0.39455965526528414,
      "grad_norm": 0.8915012174115184,
      "learning_rate": 9.070359208525598e-06,
      "loss": 0.411223030090332,
      "num_input_tokens_seen": 186858400,
      "step": 5860,
      "train_runtime": 15356.1349,
      "train_tokens_per_second": 12168.322
    },
    {
      "epoch": 0.3948963102612443,
      "grad_norm": 0.8370255662698854,
      "learning_rate": 9.068823051252346e-06,
      "loss": 0.40084524154663087,
      "num_input_tokens_seen": 187021040,
      "step": 5865,
      "train_runtime": 15369.4971,
      "train_tokens_per_second": 12168.325
    },
    {
      "epoch": 0.3952329652572044,
      "grad_norm": 0.8029216927292948,
      "learning_rate": 9.067285756145702e-06,
      "loss": 0.3786440134048462,
      "num_input_tokens_seen": 187183768,
      "step": 5870,
      "train_runtime": 15382.2157,
      "train_tokens_per_second": 12168.843
    },
    {
      "epoch": 0.39556962025316456,
      "grad_norm": 0.8524017651600102,
      "learning_rate": 9.065747323635565e-06,
      "loss": 0.34367804527282714,
      "num_input_tokens_seen": 187330264,
      "step": 5875,
      "train_runtime": 15395.6644,
      "train_tokens_per_second": 12167.728
    },
    {
      "epoch": 0.3959062752491247,
      "grad_norm": 0.9063536628710379,
      "learning_rate": 9.064207754152154e-06,
      "loss": 0.42354135513305663,
      "num_input_tokens_seen": 187487432,
      "step": 5880,
      "train_runtime": 15408.226,
      "train_tokens_per_second": 12168.009
    },
    {
      "epoch": 0.39624293024508483,
      "grad_norm": 0.8052408528246583,
      "learning_rate": 9.062667048126005e-06,
      "loss": 0.39244871139526366,
      "num_input_tokens_seen": 187650960,
      "step": 5885,
      "train_runtime": 15421.2318,
      "train_tokens_per_second": 12168.351
    },
    {
      "epoch": 0.396579585241045,
      "grad_norm": 0.8609627530954982,
      "learning_rate": 9.061125205987971e-06,
      "loss": 0.42192506790161133,
      "num_input_tokens_seen": 187809824,
      "step": 5890,
      "train_runtime": 15434.7336,
      "train_tokens_per_second": 12167.999
    },
    {
      "epoch": 0.3969162402370051,
      "grad_norm": 0.9358273490435912,
      "learning_rate": 9.059582228169222e-06,
      "loss": 0.38229267597198485,
      "num_input_tokens_seen": 187962480,
      "step": 5895,
      "train_runtime": 15448.1777,
      "train_tokens_per_second": 12167.291
    },
    {
      "epoch": 0.39725289523296525,
      "grad_norm": 0.8098644138468156,
      "learning_rate": 9.058038115101248e-06,
      "loss": 0.40517187118530273,
      "num_input_tokens_seen": 188125032,
      "step": 5900,
      "train_runtime": 15460.8407,
      "train_tokens_per_second": 12167.84
    },
    {
      "epoch": 0.3975895502289254,
      "grad_norm": 0.7648835503770965,
      "learning_rate": 9.056492867215857e-06,
      "loss": 0.3762061595916748,
      "num_input_tokens_seen": 188281880,
      "step": 5905,
      "train_runtime": 15474.5271,
      "train_tokens_per_second": 12167.214
    },
    {
      "epoch": 0.39792620522488553,
      "grad_norm": 0.747794607520484,
      "learning_rate": 9.05494648494517e-06,
      "loss": 0.39612545967102053,
      "num_input_tokens_seen": 188438496,
      "step": 5910,
      "train_runtime": 15487.9809,
      "train_tokens_per_second": 12166.757
    },
    {
      "epoch": 0.39826286022084567,
      "grad_norm": 0.7620731357615933,
      "learning_rate": 9.053398968721629e-06,
      "loss": 0.3705871343612671,
      "num_input_tokens_seen": 188602120,
      "step": 5915,
      "train_runtime": 15501.7093,
      "train_tokens_per_second": 12166.537
    },
    {
      "epoch": 0.3985995152168058,
      "grad_norm": 0.8018640078537024,
      "learning_rate": 9.05185031897799e-06,
      "loss": 0.4284857749938965,
      "num_input_tokens_seen": 188763320,
      "step": 5920,
      "train_runtime": 15514.5217,
      "train_tokens_per_second": 12166.88
    },
    {
      "epoch": 0.39893617021276595,
      "grad_norm": 0.866387989255959,
      "learning_rate": 9.050300536147331e-06,
      "loss": 0.4351509094238281,
      "num_input_tokens_seen": 188927160,
      "step": 5925,
      "train_runtime": 15526.9376,
      "train_tokens_per_second": 12167.703
    },
    {
      "epoch": 0.3992728252087261,
      "grad_norm": 0.8936389951020064,
      "learning_rate": 9.048749620663044e-06,
      "loss": 0.3660414218902588,
      "num_input_tokens_seen": 189090312,
      "step": 5930,
      "train_runtime": 15539.7926,
      "train_tokens_per_second": 12168.136
    },
    {
      "epoch": 0.39960948020468623,
      "grad_norm": 0.803482186573226,
      "learning_rate": 9.047197572958834e-06,
      "loss": 0.387343168258667,
      "num_input_tokens_seen": 189252584,
      "step": 5935,
      "train_runtime": 15552.2237,
      "train_tokens_per_second": 12168.844
    },
    {
      "epoch": 0.39994613520064637,
      "grad_norm": 0.838252065087558,
      "learning_rate": 9.04564439346873e-06,
      "loss": 0.3894481897354126,
      "num_input_tokens_seen": 189406888,
      "step": 5940,
      "train_runtime": 15564.8973,
      "train_tokens_per_second": 12168.849
    },
    {
      "epoch": 0.4002827901966065,
      "grad_norm": 0.9667848225791741,
      "learning_rate": 9.044090082627071e-06,
      "loss": 0.4214315414428711,
      "num_input_tokens_seen": 189559896,
      "step": 5945,
      "train_runtime": 15577.8575,
      "train_tokens_per_second": 12168.547
    },
    {
      "epoch": 0.40061944519256665,
      "grad_norm": 0.8400420293920484,
      "learning_rate": 9.042534640868514e-06,
      "loss": 0.408921480178833,
      "num_input_tokens_seen": 189711728,
      "step": 5950,
      "train_runtime": 15590.6384,
      "train_tokens_per_second": 12168.31
    },
    {
      "epoch": 0.4009561001885268,
      "grad_norm": 0.7616169642484772,
      "learning_rate": 9.04097806862804e-06,
      "loss": 0.415556001663208,
      "num_input_tokens_seen": 189870832,
      "step": 5955,
      "train_runtime": 15603.7912,
      "train_tokens_per_second": 12168.25
    },
    {
      "epoch": 0.4012927551844869,
      "grad_norm": 0.8858349537977714,
      "learning_rate": 9.03942036634093e-06,
      "loss": 0.4213587760925293,
      "num_input_tokens_seen": 190028904,
      "step": 5960,
      "train_runtime": 15617.2964,
      "train_tokens_per_second": 12167.849
    },
    {
      "epoch": 0.40162941018044707,
      "grad_norm": 0.9141605664703594,
      "learning_rate": 9.037861534442797e-06,
      "loss": 0.37289316654205323,
      "num_input_tokens_seen": 190191840,
      "step": 5965,
      "train_runtime": 15630.0631,
      "train_tokens_per_second": 12168.335
    },
    {
      "epoch": 0.4019660651764072,
      "grad_norm": 0.8192586079814421,
      "learning_rate": 9.036301573369563e-06,
      "loss": 0.4016674041748047,
      "num_input_tokens_seen": 190355496,
      "step": 5970,
      "train_runtime": 15643.1143,
      "train_tokens_per_second": 12168.644
    },
    {
      "epoch": 0.40230272017236735,
      "grad_norm": 0.8120674008486066,
      "learning_rate": 9.034740483557465e-06,
      "loss": 0.3864711284637451,
      "num_input_tokens_seen": 190515816,
      "step": 5975,
      "train_runtime": 15655.8952,
      "train_tokens_per_second": 12168.951
    },
    {
      "epoch": 0.4026393751683275,
      "grad_norm": 0.8375662565507036,
      "learning_rate": 9.033178265443055e-06,
      "loss": 0.3972316265106201,
      "num_input_tokens_seen": 190676784,
      "step": 5980,
      "train_runtime": 15669.5932,
      "train_tokens_per_second": 12168.585
    },
    {
      "epoch": 0.4029760301642876,
      "grad_norm": 0.7948683684826887,
      "learning_rate": 9.031614919463206e-06,
      "loss": 0.3926998615264893,
      "num_input_tokens_seen": 190836056,
      "step": 5985,
      "train_runtime": 15682.0858,
      "train_tokens_per_second": 12169.048
    },
    {
      "epoch": 0.40331268516024776,
      "grad_norm": 0.7750158583199912,
      "learning_rate": 9.030050446055099e-06,
      "loss": 0.3480082988739014,
      "num_input_tokens_seen": 190992424,
      "step": 5990,
      "train_runtime": 15694.8333,
      "train_tokens_per_second": 12169.127
    },
    {
      "epoch": 0.4036493401562079,
      "grad_norm": 0.8056815087789841,
      "learning_rate": 9.028484845656235e-06,
      "loss": 0.39787003993988035,
      "num_input_tokens_seen": 191155872,
      "step": 5995,
      "train_runtime": 15707.2746,
      "train_tokens_per_second": 12169.894
    },
    {
      "epoch": 0.40398599515216804,
      "grad_norm": 0.7618788072626257,
      "learning_rate": 9.026918118704431e-06,
      "loss": 0.45204849243164064,
      "num_input_tokens_seen": 191316712,
      "step": 6000,
      "train_runtime": 15720.3448,
      "train_tokens_per_second": 12170.007
    },
    {
      "epoch": 0.4043226501481282,
      "grad_norm": 0.7065275851884097,
      "learning_rate": 9.025350265637816e-06,
      "loss": 0.36794228553771974,
      "num_input_tokens_seen": 191479024,
      "step": 6005,
      "train_runtime": 15733.3655,
      "train_tokens_per_second": 12170.252
    },
    {
      "epoch": 0.4046593051440883,
      "grad_norm": 1.037368970805501,
      "learning_rate": 9.023781286894834e-06,
      "loss": 0.381566596031189,
      "num_input_tokens_seen": 191634136,
      "step": 6010,
      "train_runtime": 15746.1296,
      "train_tokens_per_second": 12170.237
    },
    {
      "epoch": 0.40499596014004846,
      "grad_norm": 0.8031970303960078,
      "learning_rate": 9.022211182914247e-06,
      "loss": 0.4001512050628662,
      "num_input_tokens_seen": 191791816,
      "step": 6015,
      "train_runtime": 15758.7662,
      "train_tokens_per_second": 12170.484
    },
    {
      "epoch": 0.4053326151360086,
      "grad_norm": 0.811674031567323,
      "learning_rate": 9.020639954135128e-06,
      "loss": 0.3773888826370239,
      "num_input_tokens_seen": 191948704,
      "step": 6020,
      "train_runtime": 15771.8651,
      "train_tokens_per_second": 12170.324
    },
    {
      "epoch": 0.40566927013196874,
      "grad_norm": 0.7368352838298349,
      "learning_rate": 9.019067600996867e-06,
      "loss": 0.3350774049758911,
      "num_input_tokens_seen": 192103376,
      "step": 6025,
      "train_runtime": 15785.3402,
      "train_tokens_per_second": 12169.733
    },
    {
      "epoch": 0.4060059251279289,
      "grad_norm": 0.8228241168164091,
      "learning_rate": 9.017494123939169e-06,
      "loss": 0.39459714889526365,
      "num_input_tokens_seen": 192263096,
      "step": 6030,
      "train_runtime": 15798.0287,
      "train_tokens_per_second": 12170.069
    },
    {
      "epoch": 0.406342580123889,
      "grad_norm": 0.7493182645039433,
      "learning_rate": 9.015919523402048e-06,
      "loss": 0.39789183139801027,
      "num_input_tokens_seen": 192423632,
      "step": 6035,
      "train_runtime": 15811.4076,
      "train_tokens_per_second": 12169.924
    },
    {
      "epoch": 0.40667923511984916,
      "grad_norm": 0.944138570357275,
      "learning_rate": 9.014343799825838e-06,
      "loss": 0.405488109588623,
      "num_input_tokens_seen": 192576160,
      "step": 6040,
      "train_runtime": 15824.8331,
      "train_tokens_per_second": 12169.238
    },
    {
      "epoch": 0.4070158901158093,
      "grad_norm": 0.7229076218989445,
      "learning_rate": 9.012766953651186e-06,
      "loss": 0.4446002960205078,
      "num_input_tokens_seen": 192738024,
      "step": 6045,
      "train_runtime": 15837.3665,
      "train_tokens_per_second": 12169.828
    },
    {
      "epoch": 0.40735254511176944,
      "grad_norm": 0.7346346622124152,
      "learning_rate": 9.01118898531905e-06,
      "loss": 0.42488913536071776,
      "num_input_tokens_seen": 192896808,
      "step": 6050,
      "train_runtime": 15851.3029,
      "train_tokens_per_second": 12169.145
    },
    {
      "epoch": 0.4076892001077296,
      "grad_norm": 0.7545877021696753,
      "learning_rate": 9.009609895270708e-06,
      "loss": 0.36740038394927976,
      "num_input_tokens_seen": 193060096,
      "step": 6055,
      "train_runtime": 15864.2103,
      "train_tokens_per_second": 12169.537
    },
    {
      "epoch": 0.4080258551036897,
      "grad_norm": 0.8929205988945362,
      "learning_rate": 9.008029683947743e-06,
      "loss": 0.432479190826416,
      "num_input_tokens_seen": 193219616,
      "step": 6060,
      "train_runtime": 15876.8956,
      "train_tokens_per_second": 12169.861
    },
    {
      "epoch": 0.40836251009964986,
      "grad_norm": 0.837476773262991,
      "learning_rate": 9.006448351792057e-06,
      "loss": 0.41374797821044923,
      "num_input_tokens_seen": 193380576,
      "step": 6065,
      "train_runtime": 15890.3347,
      "train_tokens_per_second": 12169.698
    },
    {
      "epoch": 0.40869916509561,
      "grad_norm": 0.7238584969936781,
      "learning_rate": 9.004865899245864e-06,
      "loss": 0.39906086921691897,
      "num_input_tokens_seen": 193540592,
      "step": 6070,
      "train_runtime": 15902.9456,
      "train_tokens_per_second": 12170.11
    },
    {
      "epoch": 0.40903582009157013,
      "grad_norm": 0.8608805638553695,
      "learning_rate": 9.003282326751694e-06,
      "loss": 0.4137234687805176,
      "num_input_tokens_seen": 193702528,
      "step": 6075,
      "train_runtime": 15916.4029,
      "train_tokens_per_second": 12169.994
    },
    {
      "epoch": 0.4093724750875303,
      "grad_norm": 0.9214624756540415,
      "learning_rate": 9.001697634752387e-06,
      "loss": 0.40547928810119627,
      "num_input_tokens_seen": 193857872,
      "step": 6080,
      "train_runtime": 15929.847,
      "train_tokens_per_second": 12169.475
    },
    {
      "epoch": 0.4097091300834904,
      "grad_norm": 0.8070033670926007,
      "learning_rate": 9.000111823691097e-06,
      "loss": 0.39362249374389646,
      "num_input_tokens_seen": 194020144,
      "step": 6085,
      "train_runtime": 15943.1037,
      "train_tokens_per_second": 12169.534
    },
    {
      "epoch": 0.41004578507945055,
      "grad_norm": 0.8981515344845221,
      "learning_rate": 8.99852489401129e-06,
      "loss": 0.4239961624145508,
      "num_input_tokens_seen": 194180504,
      "step": 6090,
      "train_runtime": 15955.5323,
      "train_tokens_per_second": 12170.105
    },
    {
      "epoch": 0.4103824400754107,
      "grad_norm": 0.8182106166403889,
      "learning_rate": 8.996936846156748e-06,
      "loss": 0.3660509824752808,
      "num_input_tokens_seen": 194338048,
      "step": 6095,
      "train_runtime": 15968.6049,
      "train_tokens_per_second": 12170.008
    },
    {
      "epoch": 0.41071909507137083,
      "grad_norm": 0.9047333854399864,
      "learning_rate": 8.995347680571563e-06,
      "loss": 0.4100959777832031,
      "num_input_tokens_seen": 194499064,
      "step": 6100,
      "train_runtime": 15982.0207,
      "train_tokens_per_second": 12169.867
    },
    {
      "epoch": 0.41105575006733097,
      "grad_norm": 0.7778770496312896,
      "learning_rate": 8.993757397700137e-06,
      "loss": 0.4050309181213379,
      "num_input_tokens_seen": 194660184,
      "step": 6105,
      "train_runtime": 15995.5461,
      "train_tokens_per_second": 12169.649
    },
    {
      "epoch": 0.41139240506329117,
      "grad_norm": 1.4907452336465898,
      "learning_rate": 8.99216599798719e-06,
      "loss": 0.4434063911437988,
      "num_input_tokens_seen": 194824024,
      "step": 6110,
      "train_runtime": 16007.9622,
      "train_tokens_per_second": 12170.445
    },
    {
      "epoch": 0.4117290600592513,
      "grad_norm": 0.7858470845855655,
      "learning_rate": 8.990573481877753e-06,
      "loss": 0.37927401065826416,
      "num_input_tokens_seen": 194980144,
      "step": 6115,
      "train_runtime": 16021.803,
      "train_tokens_per_second": 12169.676
    },
    {
      "epoch": 0.41206571505521145,
      "grad_norm": 0.8632857159384224,
      "learning_rate": 8.988979849817167e-06,
      "loss": 0.36629629135131836,
      "num_input_tokens_seen": 195129536,
      "step": 6120,
      "train_runtime": 16035.0786,
      "train_tokens_per_second": 12168.917
    },
    {
      "epoch": 0.4124023700511716,
      "grad_norm": 0.8252474045339431,
      "learning_rate": 8.987385102251085e-06,
      "loss": 0.3823172330856323,
      "num_input_tokens_seen": 195284256,
      "step": 6125,
      "train_runtime": 16048.6856,
      "train_tokens_per_second": 12168.24
    },
    {
      "epoch": 0.4127390250471317,
      "grad_norm": 0.8412642559178806,
      "learning_rate": 8.985789239625475e-06,
      "loss": 0.4026922702789307,
      "num_input_tokens_seen": 195444920,
      "step": 6130,
      "train_runtime": 16061.1364,
      "train_tokens_per_second": 12168.81
    },
    {
      "epoch": 0.41307568004309186,
      "grad_norm": 0.8540486966451508,
      "learning_rate": 8.984192262386613e-06,
      "loss": 0.37354564666748047,
      "num_input_tokens_seen": 195607008,
      "step": 6135,
      "train_runtime": 16073.9344,
      "train_tokens_per_second": 12169.205
    },
    {
      "epoch": 0.413412335039052,
      "grad_norm": 0.8107912129309226,
      "learning_rate": 8.98259417098109e-06,
      "loss": 0.40629425048828127,
      "num_input_tokens_seen": 195770784,
      "step": 6140,
      "train_runtime": 16086.9606,
      "train_tokens_per_second": 12169.532
    },
    {
      "epoch": 0.41374899003501214,
      "grad_norm": 0.8114748297424631,
      "learning_rate": 8.980994965855808e-06,
      "loss": 0.40453157424926756,
      "num_input_tokens_seen": 195929880,
      "step": 6145,
      "train_runtime": 16099.3823,
      "train_tokens_per_second": 12170.025
    },
    {
      "epoch": 0.4140856450309723,
      "grad_norm": 0.8195637603118618,
      "learning_rate": 8.979394647457976e-06,
      "loss": 0.4065361976623535,
      "num_input_tokens_seen": 196090936,
      "step": 6150,
      "train_runtime": 16111.8583,
      "train_tokens_per_second": 12170.597
    },
    {
      "epoch": 0.4144223000269324,
      "grad_norm": 0.7491849435521459,
      "learning_rate": 8.97779321623512e-06,
      "loss": 0.35030360221862794,
      "num_input_tokens_seen": 196245128,
      "step": 6155,
      "train_runtime": 16125.0804,
      "train_tokens_per_second": 12170.18
    },
    {
      "epoch": 0.41475895502289256,
      "grad_norm": 0.7708602924157264,
      "learning_rate": 8.976190672635077e-06,
      "loss": 0.40508880615234377,
      "num_input_tokens_seen": 196406272,
      "step": 6160,
      "train_runtime": 16138.0809,
      "train_tokens_per_second": 12170.361
    },
    {
      "epoch": 0.4150956100188527,
      "grad_norm": 0.8593333629155138,
      "learning_rate": 8.974587017105991e-06,
      "loss": 0.4211843490600586,
      "num_input_tokens_seen": 196560072,
      "step": 6165,
      "train_runtime": 16150.8737,
      "train_tokens_per_second": 12170.244
    },
    {
      "epoch": 0.41543226501481284,
      "grad_norm": 0.8208749917281156,
      "learning_rate": 8.97298225009632e-06,
      "loss": 0.407952880859375,
      "num_input_tokens_seen": 196721824,
      "step": 6170,
      "train_runtime": 16163.8993,
      "train_tokens_per_second": 12170.444
    },
    {
      "epoch": 0.415768920010773,
      "grad_norm": 0.7562947302772545,
      "learning_rate": 8.971376372054829e-06,
      "loss": 0.42374467849731445,
      "num_input_tokens_seen": 196883024,
      "step": 6175,
      "train_runtime": 16177.0713,
      "train_tokens_per_second": 12170.499
    },
    {
      "epoch": 0.4161055750067331,
      "grad_norm": 0.7955561472836911,
      "learning_rate": 8.969769383430602e-06,
      "loss": 0.39401679039001464,
      "num_input_tokens_seen": 197043040,
      "step": 6180,
      "train_runtime": 16189.7638,
      "train_tokens_per_second": 12170.841
    },
    {
      "epoch": 0.41644223000269326,
      "grad_norm": 0.759803900299695,
      "learning_rate": 8.968161284673027e-06,
      "loss": 0.3870995998382568,
      "num_input_tokens_seen": 197204472,
      "step": 6185,
      "train_runtime": 16202.3082,
      "train_tokens_per_second": 12171.381
    },
    {
      "epoch": 0.4167788849986534,
      "grad_norm": 0.7942920161369126,
      "learning_rate": 8.9665520762318e-06,
      "loss": 0.3831200122833252,
      "num_input_tokens_seen": 197366792,
      "step": 6190,
      "train_runtime": 16215.3475,
      "train_tokens_per_second": 12171.604
    },
    {
      "epoch": 0.41711553999461354,
      "grad_norm": 0.8769219430868023,
      "learning_rate": 8.964941758556934e-06,
      "loss": 0.41658754348754884,
      "num_input_tokens_seen": 197524608,
      "step": 6195,
      "train_runtime": 16229.0733,
      "train_tokens_per_second": 12171.034
    },
    {
      "epoch": 0.4174521949905737,
      "grad_norm": 0.7566960557757276,
      "learning_rate": 8.963330332098747e-06,
      "loss": 0.3765744924545288,
      "num_input_tokens_seen": 197676176,
      "step": 6200,
      "train_runtime": 16241.8149,
      "train_tokens_per_second": 12170.818
    },
    {
      "epoch": 0.4177888499865338,
      "grad_norm": 0.7896100385839332,
      "learning_rate": 8.961717797307872e-06,
      "loss": 0.36784772872924804,
      "num_input_tokens_seen": 197837608,
      "step": 6205,
      "train_runtime": 16254.8732,
      "train_tokens_per_second": 12170.972
    },
    {
      "epoch": 0.41812550498249396,
      "grad_norm": 0.9123405461314523,
      "learning_rate": 8.960104154635248e-06,
      "loss": 0.40808467864990233,
      "num_input_tokens_seen": 197992952,
      "step": 6210,
      "train_runtime": 16267.3764,
      "train_tokens_per_second": 12171.167
    },
    {
      "epoch": 0.4184621599784541,
      "grad_norm": 0.8011919607851222,
      "learning_rate": 8.958489404532125e-06,
      "loss": 0.40915279388427733,
      "num_input_tokens_seen": 198149216,
      "step": 6215,
      "train_runtime": 16280.3392,
      "train_tokens_per_second": 12171.074
    },
    {
      "epoch": 0.41879881497441424,
      "grad_norm": 0.7494244230690413,
      "learning_rate": 8.956873547450062e-06,
      "loss": 0.39697990417480467,
      "num_input_tokens_seen": 198308392,
      "step": 6220,
      "train_runtime": 16293.2682,
      "train_tokens_per_second": 12171.186
    },
    {
      "epoch": 0.4191354699703744,
      "grad_norm": 0.9582653313602688,
      "learning_rate": 8.95525658384093e-06,
      "loss": 0.4475069046020508,
      "num_input_tokens_seen": 198465496,
      "step": 6225,
      "train_runtime": 16306.766,
      "train_tokens_per_second": 12170.745
    },
    {
      "epoch": 0.4194721249663345,
      "grad_norm": 0.7776093919803606,
      "learning_rate": 8.953638514156908e-06,
      "loss": 0.4056656837463379,
      "num_input_tokens_seen": 198625160,
      "step": 6230,
      "train_runtime": 16320.0486,
      "train_tokens_per_second": 12170.623
    },
    {
      "epoch": 0.41980877996229465,
      "grad_norm": 0.7767806422411496,
      "learning_rate": 8.952019338850481e-06,
      "loss": 0.38939704895019533,
      "num_input_tokens_seen": 198784016,
      "step": 6235,
      "train_runtime": 16332.8976,
      "train_tokens_per_second": 12170.775
    },
    {
      "epoch": 0.4201454349582548,
      "grad_norm": 0.9012689676117106,
      "learning_rate": 8.950399058374447e-06,
      "loss": 0.3841665029525757,
      "num_input_tokens_seen": 198945344,
      "step": 6240,
      "train_runtime": 16345.8275,
      "train_tokens_per_second": 12171.017
    },
    {
      "epoch": 0.42048208995421493,
      "grad_norm": 0.8633903218074196,
      "learning_rate": 8.948777673181913e-06,
      "loss": 0.3867220640182495,
      "num_input_tokens_seen": 199107400,
      "step": 6245,
      "train_runtime": 16359.0192,
      "train_tokens_per_second": 12171.109
    },
    {
      "epoch": 0.42081874495017507,
      "grad_norm": 0.8017319486424882,
      "learning_rate": 8.947155183726296e-06,
      "loss": 0.3729018926620483,
      "num_input_tokens_seen": 199269448,
      "step": 6250,
      "train_runtime": 16372.062,
      "train_tokens_per_second": 12171.31
    },
    {
      "epoch": 0.4211553999461352,
      "grad_norm": 0.8163275104027464,
      "learning_rate": 8.94553159046132e-06,
      "loss": 0.37907135486602783,
      "num_input_tokens_seen": 199426096,
      "step": 6255,
      "train_runtime": 16385.2401,
      "train_tokens_per_second": 12171.082
    },
    {
      "epoch": 0.42149205494209535,
      "grad_norm": 0.7607843183411673,
      "learning_rate": 8.943906893841015e-06,
      "loss": 0.38002800941467285,
      "num_input_tokens_seen": 199587736,
      "step": 6260,
      "train_runtime": 16398.4572,
      "train_tokens_per_second": 12171.129
    },
    {
      "epoch": 0.4218287099380555,
      "grad_norm": 0.8887736244229288,
      "learning_rate": 8.942281094319721e-06,
      "loss": 0.43198213577270506,
      "num_input_tokens_seen": 199742616,
      "step": 6265,
      "train_runtime": 16411.2663,
      "train_tokens_per_second": 12171.067
    },
    {
      "epoch": 0.42216536493401563,
      "grad_norm": 0.8187426365560948,
      "learning_rate": 8.940654192352092e-06,
      "loss": 0.40901384353637693,
      "num_input_tokens_seen": 199898128,
      "step": 6270,
      "train_runtime": 16424.6515,
      "train_tokens_per_second": 12170.616
    },
    {
      "epoch": 0.42250201992997577,
      "grad_norm": 0.8789552757882678,
      "learning_rate": 8.939026188393083e-06,
      "loss": 0.3866248607635498,
      "num_input_tokens_seen": 200060936,
      "step": 6275,
      "train_runtime": 16437.4024,
      "train_tokens_per_second": 12171.08
    },
    {
      "epoch": 0.4228386749259359,
      "grad_norm": 0.7203565152395781,
      "learning_rate": 8.937397082897963e-06,
      "loss": 0.3998734951019287,
      "num_input_tokens_seen": 200223864,
      "step": 6280,
      "train_runtime": 16450.9201,
      "train_tokens_per_second": 12170.983
    },
    {
      "epoch": 0.42317532992189605,
      "grad_norm": 0.7756494113576382,
      "learning_rate": 8.935766876322303e-06,
      "loss": 0.4059271335601807,
      "num_input_tokens_seen": 200382296,
      "step": 6285,
      "train_runtime": 16464.0097,
      "train_tokens_per_second": 12170.929
    },
    {
      "epoch": 0.4235119849178562,
      "grad_norm": 0.7467971465269441,
      "learning_rate": 8.934135569121985e-06,
      "loss": 0.41104855537414553,
      "num_input_tokens_seen": 200545160,
      "step": 6290,
      "train_runtime": 16477.1281,
      "train_tokens_per_second": 12171.123
    },
    {
      "epoch": 0.4238486399138163,
      "grad_norm": 0.7831102362746416,
      "learning_rate": 8.932503161753202e-06,
      "loss": 0.37763237953186035,
      "num_input_tokens_seen": 200706664,
      "step": 6295,
      "train_runtime": 16491.3629,
      "train_tokens_per_second": 12170.411
    },
    {
      "epoch": 0.42418529490977647,
      "grad_norm": 0.8711319413573633,
      "learning_rate": 8.930869654672449e-06,
      "loss": 0.398427152633667,
      "num_input_tokens_seen": 200863712,
      "step": 6300,
      "train_runtime": 16503.8862,
      "train_tokens_per_second": 12170.692
    },
    {
      "epoch": 0.4245219499057366,
      "grad_norm": 0.8852711513006012,
      "learning_rate": 8.929235048336531e-06,
      "loss": 0.3873772144317627,
      "num_input_tokens_seen": 201025136,
      "step": 6305,
      "train_runtime": 16517.3164,
      "train_tokens_per_second": 12170.569
    },
    {
      "epoch": 0.42485860490169675,
      "grad_norm": 0.9583855949667159,
      "learning_rate": 8.927599343202563e-06,
      "loss": 0.4271397590637207,
      "num_input_tokens_seen": 201186984,
      "step": 6310,
      "train_runtime": 16529.8803,
      "train_tokens_per_second": 12171.11
    },
    {
      "epoch": 0.4251952598976569,
      "grad_norm": 0.9957624333329684,
      "learning_rate": 8.92596253972796e-06,
      "loss": 0.4059957504272461,
      "num_input_tokens_seen": 201350824,
      "step": 6315,
      "train_runtime": 16542.3331,
      "train_tokens_per_second": 12171.852
    },
    {
      "epoch": 0.425531914893617,
      "grad_norm": 0.8220383823221571,
      "learning_rate": 8.924324638370454e-06,
      "loss": 0.4262758731842041,
      "num_input_tokens_seen": 201514664,
      "step": 6320,
      "train_runtime": 16554.7804,
      "train_tokens_per_second": 12172.597
    },
    {
      "epoch": 0.42586856988957716,
      "grad_norm": 0.6974983819037354,
      "learning_rate": 8.922685639588076e-06,
      "loss": 0.3435081481933594,
      "num_input_tokens_seen": 201667208,
      "step": 6325,
      "train_runtime": 16568.0216,
      "train_tokens_per_second": 12172.075
    },
    {
      "epoch": 0.4262052248855373,
      "grad_norm": 0.8025442664097825,
      "learning_rate": 8.921045543839167e-06,
      "loss": 0.4132560729980469,
      "num_input_tokens_seen": 201828440,
      "step": 6330,
      "train_runtime": 16580.5082,
      "train_tokens_per_second": 12172.633
    },
    {
      "epoch": 0.42654187988149744,
      "grad_norm": 0.7580964703879259,
      "learning_rate": 8.919404351582376e-06,
      "loss": 0.3796800374984741,
      "num_input_tokens_seen": 201986256,
      "step": 6335,
      "train_runtime": 16593.1761,
      "train_tokens_per_second": 12172.851
    },
    {
      "epoch": 0.4268785348774576,
      "grad_norm": 0.7601594560802152,
      "learning_rate": 8.917762063276657e-06,
      "loss": 0.3857969522476196,
      "num_input_tokens_seen": 202141168,
      "step": 6340,
      "train_runtime": 16605.8796,
      "train_tokens_per_second": 12172.867
    },
    {
      "epoch": 0.4272151898734177,
      "grad_norm": 0.7696437366726852,
      "learning_rate": 8.916118679381268e-06,
      "loss": 0.3719454526901245,
      "num_input_tokens_seen": 202302784,
      "step": 6345,
      "train_runtime": 16619.0137,
      "train_tokens_per_second": 12172.972
    },
    {
      "epoch": 0.42755184486937786,
      "grad_norm": 0.7645595854372058,
      "learning_rate": 8.914474200355777e-06,
      "loss": 0.4145723819732666,
      "num_input_tokens_seen": 202463664,
      "step": 6350,
      "train_runtime": 16632.3917,
      "train_tokens_per_second": 12172.853
    },
    {
      "epoch": 0.427888499865338,
      "grad_norm": 0.9344552043434483,
      "learning_rate": 8.912828626660059e-06,
      "loss": 0.4205318450927734,
      "num_input_tokens_seen": 202623624,
      "step": 6355,
      "train_runtime": 16645.6046,
      "train_tokens_per_second": 12172.801
    },
    {
      "epoch": 0.42822515486129814,
      "grad_norm": 0.8541854290730081,
      "learning_rate": 8.911181958754294e-06,
      "loss": 0.4175868988037109,
      "num_input_tokens_seen": 202783832,
      "step": 6360,
      "train_runtime": 16658.0436,
      "train_tokens_per_second": 12173.328
    },
    {
      "epoch": 0.4285618098572583,
      "grad_norm": 0.8073663004497286,
      "learning_rate": 8.909534197098966e-06,
      "loss": 0.418334436416626,
      "num_input_tokens_seen": 202946472,
      "step": 6365,
      "train_runtime": 16670.8113,
      "train_tokens_per_second": 12173.761
    },
    {
      "epoch": 0.4288984648532184,
      "grad_norm": 0.8478465127920592,
      "learning_rate": 8.907885342154865e-06,
      "loss": 0.3911870241165161,
      "num_input_tokens_seen": 203106432,
      "step": 6370,
      "train_runtime": 16684.3891,
      "train_tokens_per_second": 12173.441
    },
    {
      "epoch": 0.42923511984917856,
      "grad_norm": 0.8514826203924523,
      "learning_rate": 8.906235394383089e-06,
      "loss": 0.3916271448135376,
      "num_input_tokens_seen": 203261320,
      "step": 6375,
      "train_runtime": 16698.3456,
      "train_tokens_per_second": 12172.542
    },
    {
      "epoch": 0.4295717748451387,
      "grad_norm": 0.7694754730675708,
      "learning_rate": 8.904584354245042e-06,
      "loss": 0.38933758735656737,
      "num_input_tokens_seen": 203423376,
      "step": 6380,
      "train_runtime": 16712.2435,
      "train_tokens_per_second": 12172.117
    },
    {
      "epoch": 0.42990842984109884,
      "grad_norm": 0.9025773225725443,
      "learning_rate": 8.902932222202433e-06,
      "loss": 0.41636996269226073,
      "num_input_tokens_seen": 203587216,
      "step": 6385,
      "train_runtime": 16724.6822,
      "train_tokens_per_second": 12172.86
    },
    {
      "epoch": 0.430245084837059,
      "grad_norm": 0.7411666207577537,
      "learning_rate": 8.901278998717272e-06,
      "loss": 0.37010984420776366,
      "num_input_tokens_seen": 203745920,
      "step": 6390,
      "train_runtime": 16737.7751,
      "train_tokens_per_second": 12172.82
    },
    {
      "epoch": 0.4305817398330191,
      "grad_norm": 0.8851181563210012,
      "learning_rate": 8.899624684251878e-06,
      "loss": 0.38461480140686033,
      "num_input_tokens_seen": 203906136,
      "step": 6395,
      "train_runtime": 16750.2006,
      "train_tokens_per_second": 12173.355
    },
    {
      "epoch": 0.43091839482897926,
      "grad_norm": 0.7376785945451684,
      "learning_rate": 8.897969279268877e-06,
      "loss": 0.3654215335845947,
      "num_input_tokens_seen": 204065088,
      "step": 6400,
      "train_runtime": 16763.7124,
      "train_tokens_per_second": 12173.025
    },
    {
      "epoch": 0.4312550498249394,
      "grad_norm": 0.767766339824947,
      "learning_rate": 8.896312784231196e-06,
      "loss": 0.37800936698913573,
      "num_input_tokens_seen": 204218200,
      "step": 6405,
      "train_runtime": 16776.3752,
      "train_tokens_per_second": 12172.963
    },
    {
      "epoch": 0.43159170482089954,
      "grad_norm": 0.8647014605767486,
      "learning_rate": 8.89465519960207e-06,
      "loss": 0.4516145706176758,
      "num_input_tokens_seen": 204379112,
      "step": 6410,
      "train_runtime": 16789.4607,
      "train_tokens_per_second": 12173.06
    },
    {
      "epoch": 0.4319283598168597,
      "grad_norm": 0.7586251558449697,
      "learning_rate": 8.892996525845037e-06,
      "loss": 0.3941913604736328,
      "num_input_tokens_seen": 204542648,
      "step": 6415,
      "train_runtime": 16803.1228,
      "train_tokens_per_second": 12172.895
    },
    {
      "epoch": 0.4322650148128198,
      "grad_norm": 0.8113777619222794,
      "learning_rate": 8.89133676342394e-06,
      "loss": 0.3958760738372803,
      "num_input_tokens_seen": 204695928,
      "step": 6420,
      "train_runtime": 16816.5603,
      "train_tokens_per_second": 12172.283
    },
    {
      "epoch": 0.43260166980877995,
      "grad_norm": 0.779944474262146,
      "learning_rate": 8.889675912802923e-06,
      "loss": 0.3866041898727417,
      "num_input_tokens_seen": 204855760,
      "step": 6425,
      "train_runtime": 16829.8761,
      "train_tokens_per_second": 12172.149
    },
    {
      "epoch": 0.4329383248047401,
      "grad_norm": 0.8856747506771431,
      "learning_rate": 8.888013974446443e-06,
      "loss": 0.41153998374938966,
      "num_input_tokens_seen": 205016560,
      "step": 6430,
      "train_runtime": 16842.6785,
      "train_tokens_per_second": 12172.444
    },
    {
      "epoch": 0.43327497980070023,
      "grad_norm": 0.8903721739980905,
      "learning_rate": 8.886350948819253e-06,
      "loss": 0.3754237174987793,
      "num_input_tokens_seen": 205170616,
      "step": 6435,
      "train_runtime": 16856.0772,
      "train_tokens_per_second": 12171.908
    },
    {
      "epoch": 0.4336116347966604,
      "grad_norm": 0.7785309730752001,
      "learning_rate": 8.884686836386412e-06,
      "loss": 0.3853541374206543,
      "num_input_tokens_seen": 205331288,
      "step": 6440,
      "train_runtime": 16869.4471,
      "train_tokens_per_second": 12171.785
    },
    {
      "epoch": 0.4339482897926205,
      "grad_norm": 1.080649968407136,
      "learning_rate": 8.883021637613286e-06,
      "loss": 0.40599737167358396,
      "num_input_tokens_seen": 205492152,
      "step": 6445,
      "train_runtime": 16882.2669,
      "train_tokens_per_second": 12172.071
    },
    {
      "epoch": 0.43428494478858065,
      "grad_norm": 0.7418465419288072,
      "learning_rate": 8.881355352965537e-06,
      "loss": 0.3794271230697632,
      "num_input_tokens_seen": 205651240,
      "step": 6450,
      "train_runtime": 16895.8877,
      "train_tokens_per_second": 12171.674
    },
    {
      "epoch": 0.4346215997845408,
      "grad_norm": 0.7335827511668068,
      "learning_rate": 8.879687982909145e-06,
      "loss": 0.3569803714752197,
      "num_input_tokens_seen": 205805808,
      "step": 6455,
      "train_runtime": 16908.5562,
      "train_tokens_per_second": 12171.696
    },
    {
      "epoch": 0.43495825478050093,
      "grad_norm": 1.001987415047391,
      "learning_rate": 8.878019527910378e-06,
      "loss": 0.4369661331176758,
      "num_input_tokens_seen": 205956232,
      "step": 6460,
      "train_runtime": 16921.8604,
      "train_tokens_per_second": 12171.016
    },
    {
      "epoch": 0.43529490977646107,
      "grad_norm": 0.8005061611445161,
      "learning_rate": 8.876349988435815e-06,
      "loss": 0.46979198455810545,
      "num_input_tokens_seen": 206116112,
      "step": 6465,
      "train_runtime": 16935.1726,
      "train_tokens_per_second": 12170.889
    },
    {
      "epoch": 0.4356315647724212,
      "grad_norm": 0.826756852432038,
      "learning_rate": 8.874679364952339e-06,
      "loss": 0.40633292198181153,
      "num_input_tokens_seen": 206279952,
      "step": 6470,
      "train_runtime": 16947.6033,
      "train_tokens_per_second": 12171.63
    },
    {
      "epoch": 0.43596821976838135,
      "grad_norm": 0.7496165082560022,
      "learning_rate": 8.873007657927135e-06,
      "loss": 0.38650712966918943,
      "num_input_tokens_seen": 206441440,
      "step": 6475,
      "train_runtime": 16960.1135,
      "train_tokens_per_second": 12172.173
    },
    {
      "epoch": 0.4363048747643415,
      "grad_norm": 0.7835729027641085,
      "learning_rate": 8.87133486782769e-06,
      "loss": 0.36956214904785156,
      "num_input_tokens_seen": 206595640,
      "step": 6480,
      "train_runtime": 16974.8493,
      "train_tokens_per_second": 12170.691
    },
    {
      "epoch": 0.4366415297603016,
      "grad_norm": 0.7061650916424851,
      "learning_rate": 8.869660995121792e-06,
      "loss": 0.40279216766357423,
      "num_input_tokens_seen": 206756288,
      "step": 6485,
      "train_runtime": 16987.7467,
      "train_tokens_per_second": 12170.907
    },
    {
      "epoch": 0.43697818475626177,
      "grad_norm": 0.8158854326348044,
      "learning_rate": 8.867986040277537e-06,
      "loss": 0.4118006706237793,
      "num_input_tokens_seen": 206912712,
      "step": 6490,
      "train_runtime": 17001.8686,
      "train_tokens_per_second": 12169.998
    },
    {
      "epoch": 0.4373148397522219,
      "grad_norm": 0.8243070992286381,
      "learning_rate": 8.86631000376332e-06,
      "loss": 0.4177233695983887,
      "num_input_tokens_seen": 207069176,
      "step": 6495,
      "train_runtime": 17015.2578,
      "train_tokens_per_second": 12169.617
    },
    {
      "epoch": 0.43765149474818205,
      "grad_norm": 0.7912651297537403,
      "learning_rate": 8.864632886047841e-06,
      "loss": 0.3682330846786499,
      "num_input_tokens_seen": 207227144,
      "step": 6500,
      "train_runtime": 17028.0961,
      "train_tokens_per_second": 12169.719
    },
    {
      "epoch": 0.4379881497441422,
      "grad_norm": 0.9261598462539096,
      "learning_rate": 8.862954687600097e-06,
      "loss": 0.41315951347351076,
      "num_input_tokens_seen": 207383400,
      "step": 6505,
      "train_runtime": 17040.9889,
      "train_tokens_per_second": 12169.681
    },
    {
      "epoch": 0.4383248047401023,
      "grad_norm": 1.1330731074748384,
      "learning_rate": 8.861275408889393e-06,
      "loss": 0.4136187076568604,
      "num_input_tokens_seen": 207542168,
      "step": 6510,
      "train_runtime": 17054.5497,
      "train_tokens_per_second": 12169.314
    },
    {
      "epoch": 0.43866145973606246,
      "grad_norm": 0.9327818733323059,
      "learning_rate": 8.859595050385332e-06,
      "loss": 0.40981483459472656,
      "num_input_tokens_seen": 207703736,
      "step": 6515,
      "train_runtime": 17068.2618,
      "train_tokens_per_second": 12169.003
    },
    {
      "epoch": 0.4389981147320226,
      "grad_norm": 0.7885621842572841,
      "learning_rate": 8.857913612557825e-06,
      "loss": 0.3984759569168091,
      "num_input_tokens_seen": 207864200,
      "step": 6520,
      "train_runtime": 17081.795,
      "train_tokens_per_second": 12168.756
    },
    {
      "epoch": 0.43933476972798274,
      "grad_norm": 0.8749563441795234,
      "learning_rate": 8.856231095877077e-06,
      "loss": 0.3816246509552002,
      "num_input_tokens_seen": 208019152,
      "step": 6525,
      "train_runtime": 17094.4348,
      "train_tokens_per_second": 12168.823
    },
    {
      "epoch": 0.4396714247239429,
      "grad_norm": 0.8141550743345219,
      "learning_rate": 8.8545475008136e-06,
      "loss": 0.4005760669708252,
      "num_input_tokens_seen": 208179640,
      "step": 6530,
      "train_runtime": 17107.2657,
      "train_tokens_per_second": 12169.077
    },
    {
      "epoch": 0.440008079719903,
      "grad_norm": 0.8987331662227747,
      "learning_rate": 8.852862827838207e-06,
      "loss": 0.37380046844482423,
      "num_input_tokens_seen": 208337936,
      "step": 6535,
      "train_runtime": 17120.9085,
      "train_tokens_per_second": 12168.626
    },
    {
      "epoch": 0.44034473471586316,
      "grad_norm": 0.765956198966461,
      "learning_rate": 8.851177077422008e-06,
      "loss": 0.4113166809082031,
      "num_input_tokens_seen": 208501472,
      "step": 6540,
      "train_runtime": 17133.944,
      "train_tokens_per_second": 12168.913
    },
    {
      "epoch": 0.4406813897118233,
      "grad_norm": 1.0106113443016926,
      "learning_rate": 8.849490250036421e-06,
      "loss": 0.42745304107666016,
      "num_input_tokens_seen": 208657360,
      "step": 6545,
      "train_runtime": 17147.108,
      "train_tokens_per_second": 12168.662
    },
    {
      "epoch": 0.44101804470778344,
      "grad_norm": 0.8229589045758727,
      "learning_rate": 8.847802346153159e-06,
      "loss": 0.3718966722488403,
      "num_input_tokens_seen": 208815528,
      "step": 6550,
      "train_runtime": 17160.7233,
      "train_tokens_per_second": 12168.224
    },
    {
      "epoch": 0.4413546997037436,
      "grad_norm": 0.796960460826291,
      "learning_rate": 8.846113366244241e-06,
      "loss": 0.4133618354797363,
      "num_input_tokens_seen": 208974696,
      "step": 6555,
      "train_runtime": 17173.6823,
      "train_tokens_per_second": 12168.31
    },
    {
      "epoch": 0.4416913546997037,
      "grad_norm": 0.751544981464247,
      "learning_rate": 8.844423310781986e-06,
      "loss": 0.35882039070129396,
      "num_input_tokens_seen": 209137368,
      "step": 6560,
      "train_runtime": 17186.3785,
      "train_tokens_per_second": 12168.786
    },
    {
      "epoch": 0.44202800969566386,
      "grad_norm": 0.8147794467919871,
      "learning_rate": 8.842732180239011e-06,
      "loss": 0.4266206741333008,
      "num_input_tokens_seen": 209298600,
      "step": 6565,
      "train_runtime": 17199.6802,
      "train_tokens_per_second": 12168.75
    },
    {
      "epoch": 0.442364664691624,
      "grad_norm": 0.9851723203083609,
      "learning_rate": 8.841039975088234e-06,
      "loss": 0.3815274953842163,
      "num_input_tokens_seen": 209456896,
      "step": 6570,
      "train_runtime": 17212.5306,
      "train_tokens_per_second": 12168.861
    },
    {
      "epoch": 0.44270131968758414,
      "grad_norm": 0.7881267267746809,
      "learning_rate": 8.839346695802878e-06,
      "loss": 0.46625118255615233,
      "num_input_tokens_seen": 209618920,
      "step": 6575,
      "train_runtime": 17225.6984,
      "train_tokens_per_second": 12168.965
    },
    {
      "epoch": 0.4430379746835443,
      "grad_norm": 0.8247864465981104,
      "learning_rate": 8.837652342856459e-06,
      "loss": 0.40712246894836424,
      "num_input_tokens_seen": 209778592,
      "step": 6580,
      "train_runtime": 17238.7642,
      "train_tokens_per_second": 12169.004
    },
    {
      "epoch": 0.4433746296795044,
      "grad_norm": 0.7479122721100551,
      "learning_rate": 8.835956916722803e-06,
      "loss": 0.37038216590881345,
      "num_input_tokens_seen": 209938736,
      "step": 6585,
      "train_runtime": 17251.7156,
      "train_tokens_per_second": 12169.151
    },
    {
      "epoch": 0.44371128467546456,
      "grad_norm": 0.8037524924384541,
      "learning_rate": 8.834260417876024e-06,
      "loss": 0.4097486972808838,
      "num_input_tokens_seen": 210095024,
      "step": 6590,
      "train_runtime": 17264.6922,
      "train_tokens_per_second": 12169.057
    },
    {
      "epoch": 0.4440479396714247,
      "grad_norm": 0.8117819027739414,
      "learning_rate": 8.832562846790549e-06,
      "loss": 0.39945507049560547,
      "num_input_tokens_seen": 210257736,
      "step": 6595,
      "train_runtime": 17278.1134,
      "train_tokens_per_second": 12169.022
    },
    {
      "epoch": 0.4443845946673849,
      "grad_norm": 0.760431393143732,
      "learning_rate": 8.830864203941092e-06,
      "loss": 0.40523157119750974,
      "num_input_tokens_seen": 210419872,
      "step": 6600,
      "train_runtime": 17292.0997,
      "train_tokens_per_second": 12168.555
    },
    {
      "epoch": 0.44472124966334503,
      "grad_norm": 0.8406259681036701,
      "learning_rate": 8.829164489802677e-06,
      "loss": 0.3919215202331543,
      "num_input_tokens_seen": 210577272,
      "step": 6605,
      "train_runtime": 17305.2143,
      "train_tokens_per_second": 12168.429
    },
    {
      "epoch": 0.44505790465930517,
      "grad_norm": 0.8874134198545599,
      "learning_rate": 8.827463704850622e-06,
      "loss": 0.3872117042541504,
      "num_input_tokens_seen": 210737528,
      "step": 6610,
      "train_runtime": 17318.719,
      "train_tokens_per_second": 12168.194
    },
    {
      "epoch": 0.4453945596552653,
      "grad_norm": 0.8771821703382103,
      "learning_rate": 8.825761849560547e-06,
      "loss": 0.38438212871551514,
      "num_input_tokens_seen": 210896192,
      "step": 6615,
      "train_runtime": 17331.8987,
      "train_tokens_per_second": 12168.095
    },
    {
      "epoch": 0.44573121465122545,
      "grad_norm": 0.892410750033935,
      "learning_rate": 8.824058924408371e-06,
      "loss": 0.40035238265991213,
      "num_input_tokens_seen": 211056312,
      "step": 6620,
      "train_runtime": 17344.3902,
      "train_tokens_per_second": 12168.563
    },
    {
      "epoch": 0.4460678696471856,
      "grad_norm": 0.7760070552432398,
      "learning_rate": 8.822354929870311e-06,
      "loss": 0.39186439514160154,
      "num_input_tokens_seen": 211215168,
      "step": 6625,
      "train_runtime": 17357.5707,
      "train_tokens_per_second": 12168.475
    },
    {
      "epoch": 0.44640452464314573,
      "grad_norm": 0.7836388794703599,
      "learning_rate": 8.820649866422884e-06,
      "loss": 0.40058097839355467,
      "num_input_tokens_seen": 211378952,
      "step": 6630,
      "train_runtime": 17370.6912,
      "train_tokens_per_second": 12168.713
    },
    {
      "epoch": 0.44674117963910587,
      "grad_norm": 0.7406913292362001,
      "learning_rate": 8.818943734542905e-06,
      "loss": 0.3794663429260254,
      "num_input_tokens_seen": 211538880,
      "step": 6635,
      "train_runtime": 17383.4248,
      "train_tokens_per_second": 12168.999
    },
    {
      "epoch": 0.447077834635066,
      "grad_norm": 0.8669209063529325,
      "learning_rate": 8.817236534707486e-06,
      "loss": 0.3923959732055664,
      "num_input_tokens_seen": 211699136,
      "step": 6640,
      "train_runtime": 17396.3703,
      "train_tokens_per_second": 12169.156
    },
    {
      "epoch": 0.44741448963102615,
      "grad_norm": 0.9249704546623426,
      "learning_rate": 8.815528267394045e-06,
      "loss": 0.39086484909057617,
      "num_input_tokens_seen": 211859248,
      "step": 6645,
      "train_runtime": 17408.8413,
      "train_tokens_per_second": 12169.635
    },
    {
      "epoch": 0.4477511446269863,
      "grad_norm": 0.7906495271904804,
      "learning_rate": 8.81381893308029e-06,
      "loss": 0.4134485721588135,
      "num_input_tokens_seen": 212021880,
      "step": 6650,
      "train_runtime": 17421.8541,
      "train_tokens_per_second": 12169.88
    },
    {
      "epoch": 0.4480877996229464,
      "grad_norm": 0.7898332042713607,
      "learning_rate": 8.81210853224423e-06,
      "loss": 0.4078489303588867,
      "num_input_tokens_seen": 212178808,
      "step": 6655,
      "train_runtime": 17435.0406,
      "train_tokens_per_second": 12169.677
    },
    {
      "epoch": 0.44842445461890656,
      "grad_norm": 0.8404495359433463,
      "learning_rate": 8.810397065364177e-06,
      "loss": 0.37212514877319336,
      "num_input_tokens_seen": 212331904,
      "step": 6660,
      "train_runtime": 17448.5209,
      "train_tokens_per_second": 12169.049
    },
    {
      "epoch": 0.4487611096148667,
      "grad_norm": 0.7825132274677199,
      "learning_rate": 8.808684532918735e-06,
      "loss": 0.40340847969055177,
      "num_input_tokens_seen": 212492088,
      "step": 6665,
      "train_runtime": 17462.0371,
      "train_tokens_per_second": 12168.803
    },
    {
      "epoch": 0.44909776461082684,
      "grad_norm": 0.8570183065553999,
      "learning_rate": 8.806970935386807e-06,
      "loss": 0.3904376745223999,
      "num_input_tokens_seen": 212648056,
      "step": 6670,
      "train_runtime": 17475.2084,
      "train_tokens_per_second": 12168.556
    },
    {
      "epoch": 0.449434419606787,
      "grad_norm": 0.8284935573964377,
      "learning_rate": 8.805256273247597e-06,
      "loss": 0.39842920303344725,
      "num_input_tokens_seen": 212808760,
      "step": 6675,
      "train_runtime": 17487.8267,
      "train_tokens_per_second": 12168.965
    },
    {
      "epoch": 0.4497710746027471,
      "grad_norm": 0.796623884357856,
      "learning_rate": 8.803540546980605e-06,
      "loss": 0.40415778160095217,
      "num_input_tokens_seen": 212965304,
      "step": 6680,
      "train_runtime": 17500.828,
      "train_tokens_per_second": 12168.87
    },
    {
      "epoch": 0.45010772959870726,
      "grad_norm": 0.7744746352317847,
      "learning_rate": 8.801823757065628e-06,
      "loss": 0.4022409915924072,
      "num_input_tokens_seen": 213125232,
      "step": 6685,
      "train_runtime": 17514.0128,
      "train_tokens_per_second": 12168.841
    },
    {
      "epoch": 0.4504443845946674,
      "grad_norm": 0.7580401213092398,
      "learning_rate": 8.800105903982758e-06,
      "loss": 0.38105192184448244,
      "num_input_tokens_seen": 213280840,
      "step": 6690,
      "train_runtime": 17527.0911,
      "train_tokens_per_second": 12168.639
    },
    {
      "epoch": 0.45078103959062754,
      "grad_norm": 0.8406139757419543,
      "learning_rate": 8.798386988212393e-06,
      "loss": 0.36024372577667235,
      "num_input_tokens_seen": 213433632,
      "step": 6695,
      "train_runtime": 17540.7379,
      "train_tokens_per_second": 12167.882
    },
    {
      "epoch": 0.4511176945865877,
      "grad_norm": 0.75351032576296,
      "learning_rate": 8.796667010235216e-06,
      "loss": 0.4107832431793213,
      "num_input_tokens_seen": 213596120,
      "step": 6700,
      "train_runtime": 17554.2753,
      "train_tokens_per_second": 12167.755
    },
    {
      "epoch": 0.4514543495825478,
      "grad_norm": 0.8697942812642919,
      "learning_rate": 8.794945970532219e-06,
      "loss": 0.39205708503723147,
      "num_input_tokens_seen": 213756792,
      "step": 6705,
      "train_runtime": 17566.7415,
      "train_tokens_per_second": 12168.267
    },
    {
      "epoch": 0.45179100457850796,
      "grad_norm": 0.8508593484656624,
      "learning_rate": 8.79322386958468e-06,
      "loss": 0.42752652168273925,
      "num_input_tokens_seen": 213916936,
      "step": 6710,
      "train_runtime": 17579.2101,
      "train_tokens_per_second": 12168.746
    },
    {
      "epoch": 0.4521276595744681,
      "grad_norm": 0.8201516596388327,
      "learning_rate": 8.791500707874183e-06,
      "loss": 0.35318779945373535,
      "num_input_tokens_seen": 214078000,
      "step": 6715,
      "train_runtime": 17593.4481,
      "train_tokens_per_second": 12168.052
    },
    {
      "epoch": 0.45246431457042824,
      "grad_norm": 0.8088966745033099,
      "learning_rate": 8.789776485882601e-06,
      "loss": 0.40700445175170896,
      "num_input_tokens_seen": 214240720,
      "step": 6720,
      "train_runtime": 17606.1625,
      "train_tokens_per_second": 12168.507
    },
    {
      "epoch": 0.4528009695663884,
      "grad_norm": 0.789583358396357,
      "learning_rate": 8.788051204092112e-06,
      "loss": 0.38327717781066895,
      "num_input_tokens_seen": 214402544,
      "step": 6725,
      "train_runtime": 17619.6186,
      "train_tokens_per_second": 12168.399
    },
    {
      "epoch": 0.4531376245623485,
      "grad_norm": 0.85960794851224,
      "learning_rate": 8.786324862985183e-06,
      "loss": 0.3902151107788086,
      "num_input_tokens_seen": 214557824,
      "step": 6730,
      "train_runtime": 17632.1516,
      "train_tokens_per_second": 12168.556
    },
    {
      "epoch": 0.45347427955830866,
      "grad_norm": 0.7771633639450803,
      "learning_rate": 8.78459746304458e-06,
      "loss": 0.36953599452972413,
      "num_input_tokens_seen": 214707008,
      "step": 6735,
      "train_runtime": 17646.0626,
      "train_tokens_per_second": 12167.417
    },
    {
      "epoch": 0.4538109345542688,
      "grad_norm": 0.9306154065922544,
      "learning_rate": 8.782869004753363e-06,
      "loss": 0.3993424892425537,
      "num_input_tokens_seen": 214863688,
      "step": 6740,
      "train_runtime": 17658.9199,
      "train_tokens_per_second": 12167.431
    },
    {
      "epoch": 0.45414758955022894,
      "grad_norm": 0.8713376597304905,
      "learning_rate": 8.781139488594892e-06,
      "loss": 0.39494800567626953,
      "num_input_tokens_seen": 215021672,
      "step": 6745,
      "train_runtime": 17671.4196,
      "train_tokens_per_second": 12167.765
    },
    {
      "epoch": 0.4544842445461891,
      "grad_norm": 0.6753005342703616,
      "learning_rate": 8.779408915052821e-06,
      "loss": 0.36996941566467284,
      "num_input_tokens_seen": 215182088,
      "step": 6750,
      "train_runtime": 17685.1772,
      "train_tokens_per_second": 12167.37
    },
    {
      "epoch": 0.4548208995421492,
      "grad_norm": 0.7950795116058244,
      "learning_rate": 8.777677284611096e-06,
      "loss": 0.354027533531189,
      "num_input_tokens_seen": 215332952,
      "step": 6755,
      "train_runtime": 17699.1602,
      "train_tokens_per_second": 12166.281
    },
    {
      "epoch": 0.45515755453810935,
      "grad_norm": 0.8834425196894982,
      "learning_rate": 8.775944597753969e-06,
      "loss": 0.39939677715301514,
      "num_input_tokens_seen": 215488848,
      "step": 6760,
      "train_runtime": 17712.1667,
      "train_tokens_per_second": 12166.148
    },
    {
      "epoch": 0.4554942095340695,
      "grad_norm": 0.7426046058775196,
      "learning_rate": 8.774210854965972e-06,
      "loss": 0.4040395736694336,
      "num_input_tokens_seen": 215648288,
      "step": 6765,
      "train_runtime": 17725.1406,
      "train_tokens_per_second": 12166.238
    },
    {
      "epoch": 0.45583086453002963,
      "grad_norm": 0.8505219793928216,
      "learning_rate": 8.772476056731946e-06,
      "loss": 0.36863436698913576,
      "num_input_tokens_seen": 215810448,
      "step": 6770,
      "train_runtime": 17737.7264,
      "train_tokens_per_second": 12166.748
    },
    {
      "epoch": 0.4561675195259898,
      "grad_norm": 0.7396477168608548,
      "learning_rate": 8.77074020353702e-06,
      "loss": 0.3714876174926758,
      "num_input_tokens_seen": 215969208,
      "step": 6775,
      "train_runtime": 17751.017,
      "train_tokens_per_second": 12166.582
    },
    {
      "epoch": 0.4565041745219499,
      "grad_norm": 0.9123448652752225,
      "learning_rate": 8.76900329586662e-06,
      "loss": 0.4195611000061035,
      "num_input_tokens_seen": 216131120,
      "step": 6780,
      "train_runtime": 17764.2864,
      "train_tokens_per_second": 12166.609
    },
    {
      "epoch": 0.45684082951791005,
      "grad_norm": 0.8231517437976272,
      "learning_rate": 8.767265334206467e-06,
      "loss": 0.38347907066345216,
      "num_input_tokens_seen": 216291168,
      "step": 6785,
      "train_runtime": 17777.2516,
      "train_tokens_per_second": 12166.738
    },
    {
      "epoch": 0.4571774845138702,
      "grad_norm": 0.7977872024966792,
      "learning_rate": 8.765526319042577e-06,
      "loss": 0.3858048439025879,
      "num_input_tokens_seen": 216449456,
      "step": 6790,
      "train_runtime": 17789.792,
      "train_tokens_per_second": 12167.059
    },
    {
      "epoch": 0.45751413950983033,
      "grad_norm": 0.8541744268445306,
      "learning_rate": 8.763786250861258e-06,
      "loss": 0.3648559093475342,
      "num_input_tokens_seen": 216612928,
      "step": 6795,
      "train_runtime": 17802.9115,
      "train_tokens_per_second": 12167.275
    },
    {
      "epoch": 0.45785079450579047,
      "grad_norm": 0.9194657749875546,
      "learning_rate": 8.762045130149114e-06,
      "loss": 0.3854583978652954,
      "num_input_tokens_seen": 216775024,
      "step": 6800,
      "train_runtime": 17815.528,
      "train_tokens_per_second": 12167.757
    },
    {
      "epoch": 0.4581874495017506,
      "grad_norm": 0.7200621597310751,
      "learning_rate": 8.76030295739305e-06,
      "loss": 0.3729256153106689,
      "num_input_tokens_seen": 216937600,
      "step": 6805,
      "train_runtime": 17828.8545,
      "train_tokens_per_second": 12167.781
    },
    {
      "epoch": 0.45852410449771075,
      "grad_norm": 0.7894993409186029,
      "learning_rate": 8.758559733080252e-06,
      "loss": 0.40532307624816893,
      "num_input_tokens_seen": 217096152,
      "step": 6810,
      "train_runtime": 17841.4522,
      "train_tokens_per_second": 12168.076
    },
    {
      "epoch": 0.4588607594936709,
      "grad_norm": 0.8093445467642725,
      "learning_rate": 8.75681545769821e-06,
      "loss": 0.3983329772949219,
      "num_input_tokens_seen": 217259520,
      "step": 6815,
      "train_runtime": 17854.4145,
      "train_tokens_per_second": 12168.392
    },
    {
      "epoch": 0.45919741448963103,
      "grad_norm": 0.8232103268961845,
      "learning_rate": 8.755070131734706e-06,
      "loss": 0.3976221799850464,
      "num_input_tokens_seen": 217421624,
      "step": 6820,
      "train_runtime": 17867.6365,
      "train_tokens_per_second": 12168.46
    },
    {
      "epoch": 0.45953406948559117,
      "grad_norm": 0.8958033190100124,
      "learning_rate": 8.753323755677812e-06,
      "loss": 0.41740145683288576,
      "num_input_tokens_seen": 217579464,
      "step": 6825,
      "train_runtime": 17880.9597,
      "train_tokens_per_second": 12168.221
    },
    {
      "epoch": 0.4598707244815513,
      "grad_norm": 0.8002993819969272,
      "learning_rate": 8.7515763300159e-06,
      "loss": 0.34453320503234863,
      "num_input_tokens_seen": 217742072,
      "step": 6830,
      "train_runtime": 17894.3239,
      "train_tokens_per_second": 12168.22
    },
    {
      "epoch": 0.46020737947751145,
      "grad_norm": 0.7532701777270369,
      "learning_rate": 8.74982785523763e-06,
      "loss": 0.4114655017852783,
      "num_input_tokens_seen": 217902256,
      "step": 6835,
      "train_runtime": 17907.2925,
      "train_tokens_per_second": 12168.353
    },
    {
      "epoch": 0.4605440344734716,
      "grad_norm": 0.7239052161210936,
      "learning_rate": 8.748078331831957e-06,
      "loss": 0.3882114887237549,
      "num_input_tokens_seen": 218065288,
      "step": 6840,
      "train_runtime": 17920.1302,
      "train_tokens_per_second": 12168.733
    },
    {
      "epoch": 0.4608806894694317,
      "grad_norm": 0.8408536904821221,
      "learning_rate": 8.746327760288129e-06,
      "loss": 0.3962335348129272,
      "num_input_tokens_seen": 218228736,
      "step": 6845,
      "train_runtime": 17933.1293,
      "train_tokens_per_second": 12169.027
    },
    {
      "epoch": 0.46121734446539187,
      "grad_norm": 0.8899269826641489,
      "learning_rate": 8.744576141095691e-06,
      "loss": 0.39309852123260497,
      "num_input_tokens_seen": 218385216,
      "step": 6850,
      "train_runtime": 17947.5304,
      "train_tokens_per_second": 12167.981
    },
    {
      "epoch": 0.461553999461352,
      "grad_norm": 0.7850798579701105,
      "learning_rate": 8.742823474744476e-06,
      "loss": 0.3981865406036377,
      "num_input_tokens_seen": 218542896,
      "step": 6855,
      "train_runtime": 17960.6226,
      "train_tokens_per_second": 12167.891
    },
    {
      "epoch": 0.46189065445731214,
      "grad_norm": 0.8890118580060308,
      "learning_rate": 8.74106976172461e-06,
      "loss": 0.4196115016937256,
      "num_input_tokens_seen": 218704176,
      "step": 6860,
      "train_runtime": 17973.0681,
      "train_tokens_per_second": 12168.439
    },
    {
      "epoch": 0.4622273094532723,
      "grad_norm": 0.8224090861959624,
      "learning_rate": 8.739315002526513e-06,
      "loss": 0.39695000648498535,
      "num_input_tokens_seen": 218865072,
      "step": 6865,
      "train_runtime": 17986.7731,
      "train_tokens_per_second": 12168.112
    },
    {
      "epoch": 0.4625639644492324,
      "grad_norm": 0.7164602207810157,
      "learning_rate": 8.737559197640902e-06,
      "loss": 0.36735970973968507,
      "num_input_tokens_seen": 219025624,
      "step": 6870,
      "train_runtime": 18000.1906,
      "train_tokens_per_second": 12167.961
    },
    {
      "epoch": 0.46290061944519256,
      "grad_norm": 0.8357823103032376,
      "learning_rate": 8.735802347558778e-06,
      "loss": 0.4234785079956055,
      "num_input_tokens_seen": 219181568,
      "step": 6875,
      "train_runtime": 18013.3505,
      "train_tokens_per_second": 12167.729
    },
    {
      "epoch": 0.4632372744411527,
      "grad_norm": 0.8309752476170185,
      "learning_rate": 8.734044452771442e-06,
      "loss": 0.38432772159576417,
      "num_input_tokens_seen": 219334728,
      "step": 6880,
      "train_runtime": 18025.8307,
      "train_tokens_per_second": 12167.801
    },
    {
      "epoch": 0.46357392943711284,
      "grad_norm": 0.8811580036900444,
      "learning_rate": 8.73228551377048e-06,
      "loss": 0.4391273021697998,
      "num_input_tokens_seen": 219493880,
      "step": 6885,
      "train_runtime": 18038.4519,
      "train_tokens_per_second": 12168.111
    },
    {
      "epoch": 0.463910584433073,
      "grad_norm": 0.7792481733734428,
      "learning_rate": 8.730525531047776e-06,
      "loss": 0.43122043609619143,
      "num_input_tokens_seen": 219654464,
      "step": 6890,
      "train_runtime": 18052.0699,
      "train_tokens_per_second": 12167.827
    },
    {
      "epoch": 0.4642472394290331,
      "grad_norm": 0.7476440147927917,
      "learning_rate": 8.728764505095503e-06,
      "loss": 0.40107998847961424,
      "num_input_tokens_seen": 219815544,
      "step": 6895,
      "train_runtime": 18064.9371,
      "train_tokens_per_second": 12168.077
    },
    {
      "epoch": 0.46458389442499326,
      "grad_norm": 0.7353933681022436,
      "learning_rate": 8.727002436406126e-06,
      "loss": 0.3747609853744507,
      "num_input_tokens_seen": 219979200,
      "step": 6900,
      "train_runtime": 18077.362,
      "train_tokens_per_second": 12168.767
    },
    {
      "epoch": 0.4649205494209534,
      "grad_norm": 0.7895603215904783,
      "learning_rate": 8.725239325472404e-06,
      "loss": 0.4177891731262207,
      "num_input_tokens_seen": 220139608,
      "step": 6905,
      "train_runtime": 18089.7798,
      "train_tokens_per_second": 12169.281
    },
    {
      "epoch": 0.46525720441691354,
      "grad_norm": 0.8021107552844692,
      "learning_rate": 8.723475172787381e-06,
      "loss": 0.3995223045349121,
      "num_input_tokens_seen": 220300208,
      "step": 6910,
      "train_runtime": 18102.6102,
      "train_tokens_per_second": 12169.527
    },
    {
      "epoch": 0.4655938594128737,
      "grad_norm": 0.9122128866486677,
      "learning_rate": 8.7217099788444e-06,
      "loss": 0.39371824264526367,
      "num_input_tokens_seen": 220461264,
      "step": 6915,
      "train_runtime": 18115.9139,
      "train_tokens_per_second": 12169.481
    },
    {
      "epoch": 0.4659305144088338,
      "grad_norm": 0.7737315036622889,
      "learning_rate": 8.719943744137091e-06,
      "loss": 0.39672868251800536,
      "num_input_tokens_seen": 220622360,
      "step": 6920,
      "train_runtime": 18128.7224,
      "train_tokens_per_second": 12169.769
    },
    {
      "epoch": 0.46626716940479396,
      "grad_norm": 0.7837743297102105,
      "learning_rate": 8.718176469159378e-06,
      "loss": 0.42811408042907717,
      "num_input_tokens_seen": 220785336,
      "step": 6925,
      "train_runtime": 18141.5243,
      "train_tokens_per_second": 12170.165
    },
    {
      "epoch": 0.4666038244007541,
      "grad_norm": 0.8511441920864125,
      "learning_rate": 8.716408154405469e-06,
      "loss": 0.3826107978820801,
      "num_input_tokens_seen": 220944312,
      "step": 6930,
      "train_runtime": 18154.6976,
      "train_tokens_per_second": 12170.09
    },
    {
      "epoch": 0.46694047939671424,
      "grad_norm": 1.1182358195891273,
      "learning_rate": 8.714638800369872e-06,
      "loss": 0.4499194145202637,
      "num_input_tokens_seen": 221095344,
      "step": 6935,
      "train_runtime": 18167.3442,
      "train_tokens_per_second": 12169.932
    },
    {
      "epoch": 0.4672771343926744,
      "grad_norm": 0.7239425685342206,
      "learning_rate": 8.712868407547378e-06,
      "loss": 0.41254420280456544,
      "num_input_tokens_seen": 221256952,
      "step": 6940,
      "train_runtime": 18180.5934,
      "train_tokens_per_second": 12169.952
    },
    {
      "epoch": 0.4676137893886345,
      "grad_norm": 0.8953238184350519,
      "learning_rate": 8.711096976433076e-06,
      "loss": 0.4014854431152344,
      "num_input_tokens_seen": 221419912,
      "step": 6945,
      "train_runtime": 18194.1087,
      "train_tokens_per_second": 12169.869
    },
    {
      "epoch": 0.46795044438459465,
      "grad_norm": 0.688036549653125,
      "learning_rate": 8.709324507522337e-06,
      "loss": 0.3690744400024414,
      "num_input_tokens_seen": 221581240,
      "step": 6950,
      "train_runtime": 18208.4558,
      "train_tokens_per_second": 12169.14
    },
    {
      "epoch": 0.4682870993805548,
      "grad_norm": 0.8439062836244687,
      "learning_rate": 8.707551001310828e-06,
      "loss": 0.3911032438278198,
      "num_input_tokens_seen": 221742400,
      "step": 6955,
      "train_runtime": 18220.923,
      "train_tokens_per_second": 12169.658
    },
    {
      "epoch": 0.46862375437651493,
      "grad_norm": 0.7958934368299119,
      "learning_rate": 8.705776458294503e-06,
      "loss": 0.38139638900756834,
      "num_input_tokens_seen": 221896080,
      "step": 6960,
      "train_runtime": 18233.9691,
      "train_tokens_per_second": 12169.379
    },
    {
      "epoch": 0.4689604093724751,
      "grad_norm": 0.7318534337550506,
      "learning_rate": 8.704000878969608e-06,
      "loss": 0.372223424911499,
      "num_input_tokens_seen": 222055360,
      "step": 6965,
      "train_runtime": 18248.2545,
      "train_tokens_per_second": 12168.581
    },
    {
      "epoch": 0.4692970643684352,
      "grad_norm": 0.7832820185939897,
      "learning_rate": 8.702224263832679e-06,
      "loss": 0.3828082799911499,
      "num_input_tokens_seen": 222218632,
      "step": 6970,
      "train_runtime": 18261.159,
      "train_tokens_per_second": 12168.923
    },
    {
      "epoch": 0.46963371936439535,
      "grad_norm": 0.8041866756514505,
      "learning_rate": 8.700446613380542e-06,
      "loss": 0.4333803653717041,
      "num_input_tokens_seen": 222382472,
      "step": 6975,
      "train_runtime": 18273.5743,
      "train_tokens_per_second": 12169.621
    },
    {
      "epoch": 0.4699703743603555,
      "grad_norm": 0.8351140006367409,
      "learning_rate": 8.698667928110307e-06,
      "loss": 0.37558832168579104,
      "num_input_tokens_seen": 222543544,
      "step": 6980,
      "train_runtime": 18286.0791,
      "train_tokens_per_second": 12170.107
    },
    {
      "epoch": 0.47030702935631563,
      "grad_norm": 1.0237851805051659,
      "learning_rate": 8.696888208519381e-06,
      "loss": 0.3982857227325439,
      "num_input_tokens_seen": 222700984,
      "step": 6985,
      "train_runtime": 18299.021,
      "train_tokens_per_second": 12170.104
    },
    {
      "epoch": 0.47064368435227577,
      "grad_norm": 0.7787491433474013,
      "learning_rate": 8.695107455105454e-06,
      "loss": 0.36977086067199705,
      "num_input_tokens_seen": 222856224,
      "step": 6990,
      "train_runtime": 18312.1219,
      "train_tokens_per_second": 12169.874
    },
    {
      "epoch": 0.4709803393482359,
      "grad_norm": 0.7720287118080041,
      "learning_rate": 8.693325668366513e-06,
      "loss": 0.40237860679626464,
      "num_input_tokens_seen": 223014120,
      "step": 6995,
      "train_runtime": 18325.4079,
      "train_tokens_per_second": 12169.667
    },
    {
      "epoch": 0.47131699434419605,
      "grad_norm": 0.879008593693964,
      "learning_rate": 8.691542848800825e-06,
      "loss": 0.4074204444885254,
      "num_input_tokens_seen": 223173984,
      "step": 7000,
      "train_runtime": 18338.2755,
      "train_tokens_per_second": 12169.846
    },
    {
      "epoch": 0.4716536493401562,
      "grad_norm": 0.9119760727659342,
      "learning_rate": 8.68975899690695e-06,
      "loss": 0.39657690525054934,
      "num_input_tokens_seen": 223335504,
      "step": 7005,
      "train_runtime": 18350.7771,
      "train_tokens_per_second": 12170.357
    },
    {
      "epoch": 0.47199030433611633,
      "grad_norm": 0.756123000198057,
      "learning_rate": 8.687974113183738e-06,
      "loss": 0.3958401679992676,
      "num_input_tokens_seen": 223498296,
      "step": 7010,
      "train_runtime": 18364.2103,
      "train_tokens_per_second": 12170.319
    },
    {
      "epoch": 0.47232695933207647,
      "grad_norm": 0.88566005211872,
      "learning_rate": 8.686188198130326e-06,
      "loss": 0.39900622367858884,
      "num_input_tokens_seen": 223658048,
      "step": 7015,
      "train_runtime": 18377.422,
      "train_tokens_per_second": 12170.262
    },
    {
      "epoch": 0.4726636143280366,
      "grad_norm": 0.7409509036889722,
      "learning_rate": 8.684401252246138e-06,
      "loss": 0.4031393527984619,
      "num_input_tokens_seen": 223821440,
      "step": 7020,
      "train_runtime": 18390.3747,
      "train_tokens_per_second": 12170.575
    },
    {
      "epoch": 0.47300026932399675,
      "grad_norm": 0.8413951860648937,
      "learning_rate": 8.68261327603089e-06,
      "loss": 0.4226849555969238,
      "num_input_tokens_seen": 223979960,
      "step": 7025,
      "train_runtime": 18404.0224,
      "train_tokens_per_second": 12170.163
    },
    {
      "epoch": 0.4733369243199569,
      "grad_norm": 0.9485066890931744,
      "learning_rate": 8.68082426998458e-06,
      "loss": 0.4254887580871582,
      "num_input_tokens_seen": 224143776,
      "step": 7030,
      "train_runtime": 18417.1362,
      "train_tokens_per_second": 12170.393
    },
    {
      "epoch": 0.473673579315917,
      "grad_norm": 0.7758108273203596,
      "learning_rate": 8.6790342346075e-06,
      "loss": 0.3716390609741211,
      "num_input_tokens_seen": 224298384,
      "step": 7035,
      "train_runtime": 18429.7502,
      "train_tokens_per_second": 12170.452
    },
    {
      "epoch": 0.47401023431187717,
      "grad_norm": 0.6717080586827616,
      "learning_rate": 8.677243170400228e-06,
      "loss": 0.37898454666137693,
      "num_input_tokens_seen": 224452768,
      "step": 7040,
      "train_runtime": 18442.3614,
      "train_tokens_per_second": 12170.5
    },
    {
      "epoch": 0.4743468893078373,
      "grad_norm": 0.7943940524806025,
      "learning_rate": 8.675451077863632e-06,
      "loss": 0.3969534397125244,
      "num_input_tokens_seen": 224608128,
      "step": 7045,
      "train_runtime": 18455.5512,
      "train_tokens_per_second": 12170.22
    },
    {
      "epoch": 0.47468354430379744,
      "grad_norm": 0.7793871640043072,
      "learning_rate": 8.67365795749886e-06,
      "loss": 0.3900038719177246,
      "num_input_tokens_seen": 224763912,
      "step": 7050,
      "train_runtime": 18468.2248,
      "train_tokens_per_second": 12170.304
    },
    {
      "epoch": 0.4750201992997576,
      "grad_norm": 0.792065481304858,
      "learning_rate": 8.671863809807355e-06,
      "loss": 0.35427536964416506,
      "num_input_tokens_seen": 224918960,
      "step": 7055,
      "train_runtime": 18481.5992,
      "train_tokens_per_second": 12169.886
    },
    {
      "epoch": 0.4753568542957177,
      "grad_norm": 0.7833904992101803,
      "learning_rate": 8.670068635290844e-06,
      "loss": 0.3800031661987305,
      "num_input_tokens_seen": 225078712,
      "step": 7060,
      "train_runtime": 18494.7178,
      "train_tokens_per_second": 12169.892
    },
    {
      "epoch": 0.47569350929167786,
      "grad_norm": 0.7486941563825608,
      "learning_rate": 8.668272434451343e-06,
      "loss": 0.37044103145599366,
      "num_input_tokens_seen": 225237856,
      "step": 7065,
      "train_runtime": 18507.6596,
      "train_tokens_per_second": 12169.98
    },
    {
      "epoch": 0.476030164287638,
      "grad_norm": 0.7973969961446149,
      "learning_rate": 8.666475207791154e-06,
      "loss": 0.4061105251312256,
      "num_input_tokens_seen": 225391400,
      "step": 7070,
      "train_runtime": 18521.0431,
      "train_tokens_per_second": 12169.477
    },
    {
      "epoch": 0.47636681928359814,
      "grad_norm": 0.7056974726052286,
      "learning_rate": 8.664676955812863e-06,
      "loss": 0.3848630428314209,
      "num_input_tokens_seen": 225554232,
      "step": 7075,
      "train_runtime": 18534.772,
      "train_tokens_per_second": 12169.248
    },
    {
      "epoch": 0.4767034742795583,
      "grad_norm": 0.8314253905216896,
      "learning_rate": 8.662877679019348e-06,
      "loss": 0.3976008892059326,
      "num_input_tokens_seen": 225714616,
      "step": 7080,
      "train_runtime": 18547.9962,
      "train_tokens_per_second": 12169.218
    },
    {
      "epoch": 0.4770401292755185,
      "grad_norm": 0.8089730595552718,
      "learning_rate": 8.66107737791377e-06,
      "loss": 0.4049996376037598,
      "num_input_tokens_seen": 225877832,
      "step": 7085,
      "train_runtime": 18561.6135,
      "train_tokens_per_second": 12169.084
    },
    {
      "epoch": 0.4773767842714786,
      "grad_norm": 0.8280415165088731,
      "learning_rate": 8.659276052999574e-06,
      "loss": 0.36771228313446047,
      "num_input_tokens_seen": 226029392,
      "step": 7090,
      "train_runtime": 18574.4947,
      "train_tokens_per_second": 12168.804
    },
    {
      "epoch": 0.47771343926743876,
      "grad_norm": 0.7937245006086171,
      "learning_rate": 8.657473704780504e-06,
      "loss": 0.4083720684051514,
      "num_input_tokens_seen": 226182576,
      "step": 7095,
      "train_runtime": 18587.1866,
      "train_tokens_per_second": 12168.737
    },
    {
      "epoch": 0.4780500942633989,
      "grad_norm": 0.8082274828008125,
      "learning_rate": 8.65567033376057e-06,
      "loss": 0.38915488719940183,
      "num_input_tokens_seen": 226342936,
      "step": 7100,
      "train_runtime": 18600.0812,
      "train_tokens_per_second": 12168.922
    },
    {
      "epoch": 0.47838674925935903,
      "grad_norm": 0.721123867877659,
      "learning_rate": 8.65386594044409e-06,
      "loss": 0.4009056091308594,
      "num_input_tokens_seen": 226502536,
      "step": 7105,
      "train_runtime": 18612.9093,
      "train_tokens_per_second": 12169.11
    },
    {
      "epoch": 0.4787234042553192,
      "grad_norm": 0.7851747833802771,
      "learning_rate": 8.652060525335647e-06,
      "loss": 0.35300569534301757,
      "num_input_tokens_seen": 226658496,
      "step": 7110,
      "train_runtime": 18626.3958,
      "train_tokens_per_second": 12168.672
    },
    {
      "epoch": 0.4790600592512793,
      "grad_norm": 0.7367297018516998,
      "learning_rate": 8.650254088940125e-06,
      "loss": 0.39135017395019533,
      "num_input_tokens_seen": 226812312,
      "step": 7115,
      "train_runtime": 18639.6761,
      "train_tokens_per_second": 12168.254
    },
    {
      "epoch": 0.47939671424723945,
      "grad_norm": 0.7889127972539557,
      "learning_rate": 8.648446631762686e-06,
      "loss": 0.4005890846252441,
      "num_input_tokens_seen": 226967840,
      "step": 7120,
      "train_runtime": 18653.5656,
      "train_tokens_per_second": 12167.531
    },
    {
      "epoch": 0.4797333692431996,
      "grad_norm": 0.7652681758015323,
      "learning_rate": 8.646638154308781e-06,
      "loss": 0.3942344427108765,
      "num_input_tokens_seen": 227125008,
      "step": 7125,
      "train_runtime": 18667.1049,
      "train_tokens_per_second": 12167.125
    },
    {
      "epoch": 0.48007002423915973,
      "grad_norm": 0.7235084190604785,
      "learning_rate": 8.644828657084144e-06,
      "loss": 0.359389591217041,
      "num_input_tokens_seen": 227286280,
      "step": 7130,
      "train_runtime": 18680.2392,
      "train_tokens_per_second": 12167.204
    },
    {
      "epoch": 0.48040667923511987,
      "grad_norm": 0.7522643507653209,
      "learning_rate": 8.643018140594797e-06,
      "loss": 0.3650181293487549,
      "num_input_tokens_seen": 227440440,
      "step": 7135,
      "train_runtime": 18692.8213,
      "train_tokens_per_second": 12167.261
    },
    {
      "epoch": 0.48074333423108,
      "grad_norm": 0.6960712285977015,
      "learning_rate": 8.641206605347042e-06,
      "loss": 0.3616954326629639,
      "num_input_tokens_seen": 227602432,
      "step": 7140,
      "train_runtime": 18705.9285,
      "train_tokens_per_second": 12167.396
    },
    {
      "epoch": 0.48107998922704015,
      "grad_norm": 1.0043484290836913,
      "learning_rate": 8.639394051847472e-06,
      "loss": 0.37091000080108644,
      "num_input_tokens_seen": 227759264,
      "step": 7145,
      "train_runtime": 18719.1006,
      "train_tokens_per_second": 12167.212
    },
    {
      "epoch": 0.4814166442230003,
      "grad_norm": 1.219247047591243,
      "learning_rate": 8.63758048060296e-06,
      "loss": 0.3871778964996338,
      "num_input_tokens_seen": 227917840,
      "step": 7150,
      "train_runtime": 18732.4163,
      "train_tokens_per_second": 12167.028
    },
    {
      "epoch": 0.48175329921896043,
      "grad_norm": 0.8263739399318326,
      "learning_rate": 8.635765892120666e-06,
      "loss": 0.3666724681854248,
      "num_input_tokens_seen": 228081680,
      "step": 7155,
      "train_runtime": 18744.8474,
      "train_tokens_per_second": 12167.7
    },
    {
      "epoch": 0.48208995421492057,
      "grad_norm": 0.9342396606059361,
      "learning_rate": 8.633950286908032e-06,
      "loss": 0.3934246778488159,
      "num_input_tokens_seen": 228237752,
      "step": 7160,
      "train_runtime": 18757.3623,
      "train_tokens_per_second": 12167.902
    },
    {
      "epoch": 0.4824266092108807,
      "grad_norm": 0.9580504896372071,
      "learning_rate": 8.63213366547279e-06,
      "loss": 0.4105491638183594,
      "num_input_tokens_seen": 228393320,
      "step": 7165,
      "train_runtime": 18770.8746,
      "train_tokens_per_second": 12167.431
    },
    {
      "epoch": 0.48276326420684085,
      "grad_norm": 0.8545310829644184,
      "learning_rate": 8.630316028322952e-06,
      "loss": 0.37929890155792234,
      "num_input_tokens_seen": 228554080,
      "step": 7170,
      "train_runtime": 18784.9313,
      "train_tokens_per_second": 12166.884
    },
    {
      "epoch": 0.483099919202801,
      "grad_norm": 0.8392666194219107,
      "learning_rate": 8.628497375966811e-06,
      "loss": 0.36797289848327636,
      "num_input_tokens_seen": 228717160,
      "step": 7175,
      "train_runtime": 18797.4019,
      "train_tokens_per_second": 12167.488
    },
    {
      "epoch": 0.4834365741987611,
      "grad_norm": 0.8238488351186392,
      "learning_rate": 8.62667770891295e-06,
      "loss": 0.4252311706542969,
      "num_input_tokens_seen": 228881000,
      "step": 7180,
      "train_runtime": 18809.8453,
      "train_tokens_per_second": 12168.149
    },
    {
      "epoch": 0.48377322919472127,
      "grad_norm": 0.6629076604492486,
      "learning_rate": 8.624857027670232e-06,
      "loss": 0.3757671356201172,
      "num_input_tokens_seen": 229042536,
      "step": 7185,
      "train_runtime": 18822.3237,
      "train_tokens_per_second": 12168.664
    },
    {
      "epoch": 0.4841098841906814,
      "grad_norm": 0.6925496567965691,
      "learning_rate": 8.623035332747804e-06,
      "loss": 0.4328261375427246,
      "num_input_tokens_seen": 229196680,
      "step": 7190,
      "train_runtime": 18835.575,
      "train_tokens_per_second": 12168.287
    },
    {
      "epoch": 0.48444653918664154,
      "grad_norm": 0.828165661281709,
      "learning_rate": 8.6212126246551e-06,
      "loss": 0.36807425022125245,
      "num_input_tokens_seen": 229350936,
      "step": 7195,
      "train_runtime": 18849.1423,
      "train_tokens_per_second": 12167.712
    },
    {
      "epoch": 0.4847831941826017,
      "grad_norm": 0.7725903396363935,
      "learning_rate": 8.619388903901833e-06,
      "loss": 0.393758487701416,
      "num_input_tokens_seen": 229510128,
      "step": 7200,
      "train_runtime": 18862.1802,
      "train_tokens_per_second": 12167.741
    },
    {
      "epoch": 0.4851198491785618,
      "grad_norm": 0.8423500182531545,
      "learning_rate": 8.617564170997998e-06,
      "loss": 0.38716416358947753,
      "num_input_tokens_seen": 229672152,
      "step": 7205,
      "train_runtime": 18875.0092,
      "train_tokens_per_second": 12168.055
    },
    {
      "epoch": 0.48545650417452196,
      "grad_norm": 0.7308449425619794,
      "learning_rate": 8.61573842645388e-06,
      "loss": 0.4164746761322021,
      "num_input_tokens_seen": 229833528,
      "step": 7210,
      "train_runtime": 18887.9124,
      "train_tokens_per_second": 12168.286
    },
    {
      "epoch": 0.4857931591704821,
      "grad_norm": 0.7793960754097168,
      "learning_rate": 8.61391167078004e-06,
      "loss": 0.39917917251586915,
      "num_input_tokens_seen": 229997072,
      "step": 7215,
      "train_runtime": 18901.6131,
      "train_tokens_per_second": 12168.119
    },
    {
      "epoch": 0.48612981416644224,
      "grad_norm": 0.8451245235723249,
      "learning_rate": 8.612083904487324e-06,
      "loss": 0.4313366889953613,
      "num_input_tokens_seen": 230156376,
      "step": 7220,
      "train_runtime": 18915.6258,
      "train_tokens_per_second": 12167.526
    },
    {
      "epoch": 0.4864664691624024,
      "grad_norm": 0.8521695692081358,
      "learning_rate": 8.610255128086867e-06,
      "loss": 0.40326638221740724,
      "num_input_tokens_seen": 230313304,
      "step": 7225,
      "train_runtime": 18929.7858,
      "train_tokens_per_second": 12166.715
    },
    {
      "epoch": 0.4868031241583625,
      "grad_norm": 0.8050386510678731,
      "learning_rate": 8.60842534209007e-06,
      "loss": 0.39557466506958006,
      "num_input_tokens_seen": 230475720,
      "step": 7230,
      "train_runtime": 18942.2481,
      "train_tokens_per_second": 12167.284
    },
    {
      "epoch": 0.48713977915432266,
      "grad_norm": 0.8082307432870961,
      "learning_rate": 8.606594547008636e-06,
      "loss": 0.35011341571807864,
      "num_input_tokens_seen": 230632600,
      "step": 7235,
      "train_runtime": 18954.7226,
      "train_tokens_per_second": 12167.553
    },
    {
      "epoch": 0.4874764341502828,
      "grad_norm": 1.0017943008042627,
      "learning_rate": 8.604762743354536e-06,
      "loss": 0.40049118995666505,
      "num_input_tokens_seen": 230796440,
      "step": 7240,
      "train_runtime": 18967.1762,
      "train_tokens_per_second": 12168.202
    },
    {
      "epoch": 0.48781308914624294,
      "grad_norm": 0.7981797628269263,
      "learning_rate": 8.602929931640029e-06,
      "loss": 0.41891989707946775,
      "num_input_tokens_seen": 230952464,
      "step": 7245,
      "train_runtime": 18981.4591,
      "train_tokens_per_second": 12167.266
    },
    {
      "epoch": 0.4881497441422031,
      "grad_norm": 0.9095281408060589,
      "learning_rate": 8.601096112377659e-06,
      "loss": 0.4262809753417969,
      "num_input_tokens_seen": 231109384,
      "step": 7250,
      "train_runtime": 18994.0573,
      "train_tokens_per_second": 12167.457
    },
    {
      "epoch": 0.4884863991381632,
      "grad_norm": 0.7643689769995177,
      "learning_rate": 8.59926128608024e-06,
      "loss": 0.38933048248291013,
      "num_input_tokens_seen": 231269392,
      "step": 7255,
      "train_runtime": 19006.7388,
      "train_tokens_per_second": 12167.758
    },
    {
      "epoch": 0.48882305413412336,
      "grad_norm": 0.7592561526267536,
      "learning_rate": 8.597425453260881e-06,
      "loss": 0.38649346828460696,
      "num_input_tokens_seen": 231424080,
      "step": 7260,
      "train_runtime": 19019.8184,
      "train_tokens_per_second": 12167.523
    },
    {
      "epoch": 0.4891597091300835,
      "grad_norm": 0.7925823797575888,
      "learning_rate": 8.595588614432967e-06,
      "loss": 0.38105430603027346,
      "num_input_tokens_seen": 231577080,
      "step": 7265,
      "train_runtime": 19032.8116,
      "train_tokens_per_second": 12167.255
    },
    {
      "epoch": 0.48949636412604364,
      "grad_norm": 0.7428295823783329,
      "learning_rate": 8.59375077011016e-06,
      "loss": 0.3714127540588379,
      "num_input_tokens_seen": 231740040,
      "step": 7270,
      "train_runtime": 19045.6479,
      "train_tokens_per_second": 12167.611
    },
    {
      "epoch": 0.4898330191220038,
      "grad_norm": 0.8435109599364333,
      "learning_rate": 8.591911920806412e-06,
      "loss": 0.4041157245635986,
      "num_input_tokens_seen": 231902720,
      "step": 7275,
      "train_runtime": 19058.3506,
      "train_tokens_per_second": 12168.037
    },
    {
      "epoch": 0.4901696741179639,
      "grad_norm": 0.8322023688385077,
      "learning_rate": 8.590072067035947e-06,
      "loss": 0.4000819206237793,
      "num_input_tokens_seen": 232064472,
      "step": 7280,
      "train_runtime": 19071.5335,
      "train_tokens_per_second": 12168.108
    },
    {
      "epoch": 0.49050632911392406,
      "grad_norm": 0.745203809756682,
      "learning_rate": 8.588231209313278e-06,
      "loss": 0.3859596252441406,
      "num_input_tokens_seen": 232228312,
      "step": 7285,
      "train_runtime": 19083.9793,
      "train_tokens_per_second": 12168.757
    },
    {
      "epoch": 0.4908429841098842,
      "grad_norm": 0.9308728658219169,
      "learning_rate": 8.586389348153192e-06,
      "loss": 0.3805895566940308,
      "num_input_tokens_seen": 232386608,
      "step": 7290,
      "train_runtime": 19097.3855,
      "train_tokens_per_second": 12168.504
    },
    {
      "epoch": 0.49117963910584433,
      "grad_norm": 0.8355659832138728,
      "learning_rate": 8.584546484070762e-06,
      "loss": 0.39642109870910647,
      "num_input_tokens_seen": 232545848,
      "step": 7295,
      "train_runtime": 19110.2701,
      "train_tokens_per_second": 12168.632
    },
    {
      "epoch": 0.4915162941018045,
      "grad_norm": 0.8025104909986821,
      "learning_rate": 8.582702617581338e-06,
      "loss": 0.3807774543762207,
      "num_input_tokens_seen": 232708040,
      "step": 7300,
      "train_runtime": 19124.206,
      "train_tokens_per_second": 12168.246
    },
    {
      "epoch": 0.4918529490977646,
      "grad_norm": 0.829236226042725,
      "learning_rate": 8.58085774920055e-06,
      "loss": 0.3820274829864502,
      "num_input_tokens_seen": 232863752,
      "step": 7305,
      "train_runtime": 19137.4555,
      "train_tokens_per_second": 12167.958
    },
    {
      "epoch": 0.49218960409372475,
      "grad_norm": 0.8240924008864927,
      "learning_rate": 8.579011879444313e-06,
      "loss": 0.43933906555175783,
      "num_input_tokens_seen": 233027224,
      "step": 7310,
      "train_runtime": 19150.7035,
      "train_tokens_per_second": 12168.076
    },
    {
      "epoch": 0.4925262590896849,
      "grad_norm": 0.7568917401896091,
      "learning_rate": 8.577165008828817e-06,
      "loss": 0.36581594944000245,
      "num_input_tokens_seen": 233182440,
      "step": 7315,
      "train_runtime": 19164.479,
      "train_tokens_per_second": 12167.429
    },
    {
      "epoch": 0.49286291408564503,
      "grad_norm": 0.8032131779692756,
      "learning_rate": 8.57531713787053e-06,
      "loss": 0.39217257499694824,
      "num_input_tokens_seen": 233336664,
      "step": 7320,
      "train_runtime": 19177.6754,
      "train_tokens_per_second": 12167.098
    },
    {
      "epoch": 0.49319956908160517,
      "grad_norm": 0.8942793216104644,
      "learning_rate": 8.573468267086208e-06,
      "loss": 0.3723256826400757,
      "num_input_tokens_seen": 233495256,
      "step": 7325,
      "train_runtime": 19190.9738,
      "train_tokens_per_second": 12166.931
    },
    {
      "epoch": 0.4935362240775653,
      "grad_norm": 0.7747893411559189,
      "learning_rate": 8.57161839699288e-06,
      "loss": 0.371173357963562,
      "num_input_tokens_seen": 233656224,
      "step": 7330,
      "train_runtime": 19204.4715,
      "train_tokens_per_second": 12166.762
    },
    {
      "epoch": 0.49387287907352545,
      "grad_norm": 0.9068913623850248,
      "learning_rate": 8.569767528107857e-06,
      "loss": 0.3850739002227783,
      "num_input_tokens_seen": 233814584,
      "step": 7335,
      "train_runtime": 19217.1072,
      "train_tokens_per_second": 12167.002
    },
    {
      "epoch": 0.4942095340694856,
      "grad_norm": 0.7385027125017917,
      "learning_rate": 8.567915660948727e-06,
      "loss": 0.37473487854003906,
      "num_input_tokens_seen": 233973864,
      "step": 7340,
      "train_runtime": 19229.6046,
      "train_tokens_per_second": 12167.378
    },
    {
      "epoch": 0.49454618906544573,
      "grad_norm": 0.824237938872014,
      "learning_rate": 8.566062796033358e-06,
      "loss": 0.41674509048461916,
      "num_input_tokens_seen": 234132952,
      "step": 7345,
      "train_runtime": 19242.5975,
      "train_tokens_per_second": 12167.43
    },
    {
      "epoch": 0.49488284406140587,
      "grad_norm": 0.8680983633290126,
      "learning_rate": 8.564208933879903e-06,
      "loss": 0.3666702747344971,
      "num_input_tokens_seen": 234286280,
      "step": 7350,
      "train_runtime": 19255.0868,
      "train_tokens_per_second": 12167.501
    },
    {
      "epoch": 0.495219499057366,
      "grad_norm": 0.7612132445523104,
      "learning_rate": 8.562354075006782e-06,
      "loss": 0.360690975189209,
      "num_input_tokens_seen": 234446784,
      "step": 7355,
      "train_runtime": 19269.2341,
      "train_tokens_per_second": 12166.897
    },
    {
      "epoch": 0.49555615405332615,
      "grad_norm": 0.802037477303204,
      "learning_rate": 8.560498219932706e-06,
      "loss": 0.4306445598602295,
      "num_input_tokens_seen": 234607880,
      "step": 7360,
      "train_runtime": 19282.9528,
      "train_tokens_per_second": 12166.595
    },
    {
      "epoch": 0.4958928090492863,
      "grad_norm": 0.742109746389664,
      "learning_rate": 8.558641369176654e-06,
      "loss": 0.32950568199157715,
      "num_input_tokens_seen": 234762864,
      "step": 7365,
      "train_runtime": 19296.163,
      "train_tokens_per_second": 12166.298
    },
    {
      "epoch": 0.4962294640452464,
      "grad_norm": 0.9235646081545151,
      "learning_rate": 8.556783523257892e-06,
      "loss": 0.40966596603393557,
      "num_input_tokens_seen": 234920088,
      "step": 7370,
      "train_runtime": 19309.0039,
      "train_tokens_per_second": 12166.349
    },
    {
      "epoch": 0.49656611904120657,
      "grad_norm": 0.7244504718388206,
      "learning_rate": 8.554924682695959e-06,
      "loss": 0.3682614803314209,
      "num_input_tokens_seen": 235075224,
      "step": 7375,
      "train_runtime": 19322.4372,
      "train_tokens_per_second": 12165.92
    },
    {
      "epoch": 0.4969027740371667,
      "grad_norm": 0.7768787755664207,
      "learning_rate": 8.553064848010677e-06,
      "loss": 0.4121041774749756,
      "num_input_tokens_seen": 235231432,
      "step": 7380,
      "train_runtime": 19336.0195,
      "train_tokens_per_second": 12165.453
    },
    {
      "epoch": 0.49723942903312685,
      "grad_norm": 0.7685923527523802,
      "learning_rate": 8.551204019722137e-06,
      "loss": 0.3862525701522827,
      "num_input_tokens_seen": 235387816,
      "step": 7385,
      "train_runtime": 19348.6684,
      "train_tokens_per_second": 12165.582
    },
    {
      "epoch": 0.497576084029087,
      "grad_norm": 0.9050937914697457,
      "learning_rate": 8.549342198350718e-06,
      "loss": 0.4252310752868652,
      "num_input_tokens_seen": 235548672,
      "step": 7390,
      "train_runtime": 19361.527,
      "train_tokens_per_second": 12165.811
    },
    {
      "epoch": 0.4979127390250471,
      "grad_norm": 0.7720720987560661,
      "learning_rate": 8.54747938441707e-06,
      "loss": 0.37248151302337645,
      "num_input_tokens_seen": 235709512,
      "step": 7395,
      "train_runtime": 19375.1405,
      "train_tokens_per_second": 12165.564
    },
    {
      "epoch": 0.49824939402100726,
      "grad_norm": 0.8291025850724041,
      "learning_rate": 8.545615578442126e-06,
      "loss": 0.36538872718811033,
      "num_input_tokens_seen": 235870312,
      "step": 7400,
      "train_runtime": 19388.0634,
      "train_tokens_per_second": 12165.749
    },
    {
      "epoch": 0.4985860490169674,
      "grad_norm": 0.910963815850178,
      "learning_rate": 8.54375078094709e-06,
      "loss": 0.37861878871917726,
      "num_input_tokens_seen": 236028208,
      "step": 7405,
      "train_runtime": 19400.6204,
      "train_tokens_per_second": 12166.013
    },
    {
      "epoch": 0.49892270401292754,
      "grad_norm": 0.7748510620627111,
      "learning_rate": 8.541884992453449e-06,
      "loss": 0.37966189384460447,
      "num_input_tokens_seen": 236188904,
      "step": 7410,
      "train_runtime": 19413.4046,
      "train_tokens_per_second": 12166.279
    },
    {
      "epoch": 0.4992593590088877,
      "grad_norm": 0.7848473061975617,
      "learning_rate": 8.540018213482966e-06,
      "loss": 0.40282344818115234,
      "num_input_tokens_seen": 236351696,
      "step": 7415,
      "train_runtime": 19426.1495,
      "train_tokens_per_second": 12166.677
    },
    {
      "epoch": 0.4995960140048478,
      "grad_norm": 0.7683440760802511,
      "learning_rate": 8.538150444557676e-06,
      "loss": 0.38904709815979005,
      "num_input_tokens_seen": 236508368,
      "step": 7420,
      "train_runtime": 19439.6034,
      "train_tokens_per_second": 12166.317
    },
    {
      "epoch": 0.49993266900080796,
      "grad_norm": 0.902781369044073,
      "learning_rate": 8.536281686199896e-06,
      "loss": 0.39225544929504397,
      "num_input_tokens_seen": 236667672,
      "step": 7425,
      "train_runtime": 19452.0455,
      "train_tokens_per_second": 12166.724
    },
    {
      "epoch": 0.5002693239967682,
      "grad_norm": 0.7970050827980126,
      "learning_rate": 8.534411938932218e-06,
      "loss": 0.38663678169250487,
      "num_input_tokens_seen": 236827488,
      "step": 7430,
      "train_runtime": 19464.4942,
      "train_tokens_per_second": 12167.153
    },
    {
      "epoch": 0.5006059789927283,
      "grad_norm": 0.7669132812395764,
      "learning_rate": 8.532541203277515e-06,
      "loss": 0.36896088123321535,
      "num_input_tokens_seen": 236988456,
      "step": 7435,
      "train_runtime": 19477.6134,
      "train_tokens_per_second": 12167.223
    },
    {
      "epoch": 0.5009426339886884,
      "grad_norm": 0.8878355004903203,
      "learning_rate": 8.530669479758926e-06,
      "loss": 0.3835675001144409,
      "num_input_tokens_seen": 237148488,
      "step": 7440,
      "train_runtime": 19490.297,
      "train_tokens_per_second": 12167.515
    },
    {
      "epoch": 0.5012792889846486,
      "grad_norm": 0.8489847968410151,
      "learning_rate": 8.528796768899878e-06,
      "loss": 0.38852581977844236,
      "num_input_tokens_seen": 237303120,
      "step": 7445,
      "train_runtime": 19503.635,
      "train_tokens_per_second": 12167.123
    },
    {
      "epoch": 0.5016159439806087,
      "grad_norm": 0.7422409117601764,
      "learning_rate": 8.526923071224064e-06,
      "loss": 0.3939798355102539,
      "num_input_tokens_seen": 237464112,
      "step": 7450,
      "train_runtime": 19517.2973,
      "train_tokens_per_second": 12166.854
    },
    {
      "epoch": 0.5019525989765689,
      "grad_norm": 0.7608285116002562,
      "learning_rate": 8.525048387255461e-06,
      "loss": 0.4041768550872803,
      "num_input_tokens_seen": 237626928,
      "step": 7455,
      "train_runtime": 19529.9549,
      "train_tokens_per_second": 12167.305
    },
    {
      "epoch": 0.502289253972529,
      "grad_norm": 0.8380333530892272,
      "learning_rate": 8.523172717518315e-06,
      "loss": 0.3676736354827881,
      "num_input_tokens_seen": 237783856,
      "step": 7460,
      "train_runtime": 19542.9495,
      "train_tokens_per_second": 12167.245
    },
    {
      "epoch": 0.5026259089684891,
      "grad_norm": 0.8189479656433616,
      "learning_rate": 8.521296062537156e-06,
      "loss": 0.389082670211792,
      "num_input_tokens_seen": 237943672,
      "step": 7465,
      "train_runtime": 19556.3916,
      "train_tokens_per_second": 12167.054
    },
    {
      "epoch": 0.5029625639644493,
      "grad_norm": 0.9315388775794754,
      "learning_rate": 8.51941842283678e-06,
      "loss": 0.3850302457809448,
      "num_input_tokens_seen": 238101752,
      "step": 7470,
      "train_runtime": 19570.342,
      "train_tokens_per_second": 12166.458
    },
    {
      "epoch": 0.5032992189604094,
      "grad_norm": 0.8967299187583033,
      "learning_rate": 8.517539798942265e-06,
      "loss": 0.38015069961547854,
      "num_input_tokens_seen": 238262696,
      "step": 7475,
      "train_runtime": 19583.1735,
      "train_tokens_per_second": 12166.705
    },
    {
      "epoch": 0.5036358739563696,
      "grad_norm": 0.7083454489790483,
      "learning_rate": 8.515660191378962e-06,
      "loss": 0.3841752529144287,
      "num_input_tokens_seen": 238423472,
      "step": 7480,
      "train_runtime": 19596.3601,
      "train_tokens_per_second": 12166.722
    },
    {
      "epoch": 0.5039725289523297,
      "grad_norm": 0.8069264290032326,
      "learning_rate": 8.513779600672495e-06,
      "loss": 0.3637341022491455,
      "num_input_tokens_seen": 238586368,
      "step": 7485,
      "train_runtime": 19609.8881,
      "train_tokens_per_second": 12166.636
    },
    {
      "epoch": 0.5043091839482898,
      "grad_norm": 0.7564689029445455,
      "learning_rate": 8.511898027348771e-06,
      "loss": 0.3830281734466553,
      "num_input_tokens_seen": 238739184,
      "step": 7490,
      "train_runtime": 19623.3643,
      "train_tokens_per_second": 12166.068
    },
    {
      "epoch": 0.50464583894425,
      "grad_norm": 0.7794347900353182,
      "learning_rate": 8.51001547193396e-06,
      "loss": 0.37814195156097413,
      "num_input_tokens_seen": 238899384,
      "step": 7495,
      "train_runtime": 19636.1538,
      "train_tokens_per_second": 12166.302
    },
    {
      "epoch": 0.5049824939402101,
      "grad_norm": 0.8619149041371794,
      "learning_rate": 8.508131934954515e-06,
      "loss": 0.37680203914642335,
      "num_input_tokens_seen": 239053992,
      "step": 7500,
      "train_runtime": 19648.6489,
      "train_tokens_per_second": 12166.434
    },
    {
      "epoch": 0.5053191489361702,
      "grad_norm": 0.8338488966440819,
      "learning_rate": 8.50624741693716e-06,
      "loss": 0.39595775604248046,
      "num_input_tokens_seen": 239208000,
      "step": 7505,
      "train_runtime": 19661.908,
      "train_tokens_per_second": 12166.062
    },
    {
      "epoch": 0.5056558039321304,
      "grad_norm": 0.9679525675145325,
      "learning_rate": 8.5043619184089e-06,
      "loss": 0.4429058074951172,
      "num_input_tokens_seen": 239365952,
      "step": 7510,
      "train_runtime": 19674.7092,
      "train_tokens_per_second": 12166.175
    },
    {
      "epoch": 0.5059924589280905,
      "grad_norm": 0.8648709784716618,
      "learning_rate": 8.502475439897e-06,
      "loss": 0.4197493553161621,
      "num_input_tokens_seen": 239526976,
      "step": 7515,
      "train_runtime": 19688.2169,
      "train_tokens_per_second": 12166.007
    },
    {
      "epoch": 0.5063291139240507,
      "grad_norm": 0.6912066851641233,
      "learning_rate": 8.500587981929013e-06,
      "loss": 0.3921358585357666,
      "num_input_tokens_seen": 239685984,
      "step": 7520,
      "train_runtime": 19701.1855,
      "train_tokens_per_second": 12166.069
    },
    {
      "epoch": 0.5066657689200108,
      "grad_norm": 0.8549956604670848,
      "learning_rate": 8.498699545032758e-06,
      "loss": 0.3827868938446045,
      "num_input_tokens_seen": 239846936,
      "step": 7525,
      "train_runtime": 19713.6453,
      "train_tokens_per_second": 12166.544
    },
    {
      "epoch": 0.507002423915971,
      "grad_norm": 0.7730257566710779,
      "learning_rate": 8.496810129736334e-06,
      "loss": 0.4004085063934326,
      "num_input_tokens_seen": 240009304,
      "step": 7530,
      "train_runtime": 19726.9965,
      "train_tokens_per_second": 12166.541
    },
    {
      "epoch": 0.5073390789119311,
      "grad_norm": 0.8578613508801158,
      "learning_rate": 8.494919736568105e-06,
      "loss": 0.3898079633712769,
      "num_input_tokens_seen": 240166072,
      "step": 7535,
      "train_runtime": 19739.9661,
      "train_tokens_per_second": 12166.489
    },
    {
      "epoch": 0.5076757339078912,
      "grad_norm": 0.8847458062741189,
      "learning_rate": 8.49302836605672e-06,
      "loss": 0.3955669403076172,
      "num_input_tokens_seen": 240329200,
      "step": 7540,
      "train_runtime": 19753.5471,
      "train_tokens_per_second": 12166.382
    },
    {
      "epoch": 0.5080123889038514,
      "grad_norm": 0.7306983596069232,
      "learning_rate": 8.491136018731088e-06,
      "loss": 0.377892804145813,
      "num_input_tokens_seen": 240491352,
      "step": 7545,
      "train_runtime": 19766.2004,
      "train_tokens_per_second": 12166.797
    },
    {
      "epoch": 0.5083490438998115,
      "grad_norm": 0.8312924055029284,
      "learning_rate": 8.4892426951204e-06,
      "loss": 0.37731308937072755,
      "num_input_tokens_seen": 240648120,
      "step": 7550,
      "train_runtime": 19778.9615,
      "train_tokens_per_second": 12166.873
    },
    {
      "epoch": 0.5086856988957716,
      "grad_norm": 0.762426867682343,
      "learning_rate": 8.48734839575412e-06,
      "loss": 0.3856975078582764,
      "num_input_tokens_seen": 240808192,
      "step": 7555,
      "train_runtime": 19791.5374,
      "train_tokens_per_second": 12167.23
    },
    {
      "epoch": 0.5090223538917318,
      "grad_norm": 0.7172912578935791,
      "learning_rate": 8.485453121161983e-06,
      "loss": 0.3787505626678467,
      "num_input_tokens_seen": 240969544,
      "step": 7560,
      "train_runtime": 19803.9573,
      "train_tokens_per_second": 12167.747
    },
    {
      "epoch": 0.5093590088876919,
      "grad_norm": 0.9498827879975054,
      "learning_rate": 8.483556871873993e-06,
      "loss": 0.44745187759399413,
      "num_input_tokens_seen": 241126864,
      "step": 7565,
      "train_runtime": 19816.4176,
      "train_tokens_per_second": 12168.035
    },
    {
      "epoch": 0.5096956638836521,
      "grad_norm": 0.7758613005293536,
      "learning_rate": 8.481659648420433e-06,
      "loss": 0.394504714012146,
      "num_input_tokens_seen": 241289048,
      "step": 7570,
      "train_runtime": 19829.6676,
      "train_tokens_per_second": 12168.083
    },
    {
      "epoch": 0.5100323188796122,
      "grad_norm": 0.8266814657366572,
      "learning_rate": 8.479761451331855e-06,
      "loss": 0.3816620588302612,
      "num_input_tokens_seen": 241439040,
      "step": 7575,
      "train_runtime": 19843.4902,
      "train_tokens_per_second": 12167.166
    },
    {
      "epoch": 0.5103689738755723,
      "grad_norm": 1.0558269134572658,
      "learning_rate": 8.477862281139082e-06,
      "loss": 0.3548101902008057,
      "num_input_tokens_seen": 241601992,
      "step": 7580,
      "train_runtime": 19856.9542,
      "train_tokens_per_second": 12167.122
    },
    {
      "epoch": 0.5107056288715325,
      "grad_norm": 0.9328783809010911,
      "learning_rate": 8.475962138373212e-06,
      "loss": 0.4315321445465088,
      "num_input_tokens_seen": 241754552,
      "step": 7585,
      "train_runtime": 19870.1777,
      "train_tokens_per_second": 12166.703
    },
    {
      "epoch": 0.5110422838674926,
      "grad_norm": 0.8412357848336036,
      "learning_rate": 8.474061023565614e-06,
      "loss": 0.39973859786987304,
      "num_input_tokens_seen": 241914200,
      "step": 7590,
      "train_runtime": 19882.949,
      "train_tokens_per_second": 12166.917
    },
    {
      "epoch": 0.5113789388634528,
      "grad_norm": 0.7726466280220472,
      "learning_rate": 8.472158937247931e-06,
      "loss": 0.41996207237243655,
      "num_input_tokens_seen": 242074056,
      "step": 7595,
      "train_runtime": 19896.7215,
      "train_tokens_per_second": 12166.53
    },
    {
      "epoch": 0.5117155938594129,
      "grad_norm": 0.8497524440338405,
      "learning_rate": 8.47025587995207e-06,
      "loss": 0.3897047758102417,
      "num_input_tokens_seen": 242229744,
      "step": 7600,
      "train_runtime": 19909.1637,
      "train_tokens_per_second": 12166.746
    },
    {
      "epoch": 0.512052248855373,
      "grad_norm": 0.8792368480569257,
      "learning_rate": 8.46835185221022e-06,
      "loss": 0.3975832939147949,
      "num_input_tokens_seen": 242385768,
      "step": 7605,
      "train_runtime": 19921.7492,
      "train_tokens_per_second": 12166.892
    },
    {
      "epoch": 0.5123889038513332,
      "grad_norm": 0.8754946329275609,
      "learning_rate": 8.466446854554835e-06,
      "loss": 0.36615700721740724,
      "num_input_tokens_seen": 242540456,
      "step": 7610,
      "train_runtime": 19934.5433,
      "train_tokens_per_second": 12166.843
    },
    {
      "epoch": 0.5127255588472933,
      "grad_norm": 0.7136502747491608,
      "learning_rate": 8.464540887518638e-06,
      "loss": 0.37505927085876467,
      "num_input_tokens_seen": 242696552,
      "step": 7615,
      "train_runtime": 19947.1594,
      "train_tokens_per_second": 12166.973
    },
    {
      "epoch": 0.5130622138432535,
      "grad_norm": 0.7829888877987689,
      "learning_rate": 8.462633951634631e-06,
      "loss": 0.3695961952209473,
      "num_input_tokens_seen": 242857896,
      "step": 7620,
      "train_runtime": 19960.4643,
      "train_tokens_per_second": 12166.946
    },
    {
      "epoch": 0.5133988688392136,
      "grad_norm": 0.7068907817328869,
      "learning_rate": 8.46072604743608e-06,
      "loss": 0.3502632141113281,
      "num_input_tokens_seen": 243016816,
      "step": 7625,
      "train_runtime": 19973.9017,
      "train_tokens_per_second": 12166.717
    },
    {
      "epoch": 0.5137355238351737,
      "grad_norm": 0.8506792473687397,
      "learning_rate": 8.458817175456528e-06,
      "loss": 0.40693235397338867,
      "num_input_tokens_seen": 243174048,
      "step": 7630,
      "train_runtime": 19987.1906,
      "train_tokens_per_second": 12166.495
    },
    {
      "epoch": 0.5140721788311339,
      "grad_norm": 0.8127883060542206,
      "learning_rate": 8.45690733622978e-06,
      "loss": 0.37832767963409425,
      "num_input_tokens_seen": 243333912,
      "step": 7635,
      "train_runtime": 19999.6168,
      "train_tokens_per_second": 12166.929
    },
    {
      "epoch": 0.514408833827094,
      "grad_norm": 0.8968057976755485,
      "learning_rate": 8.45499653028992e-06,
      "loss": 0.3742717981338501,
      "num_input_tokens_seen": 243493224,
      "step": 7640,
      "train_runtime": 20012.9478,
      "train_tokens_per_second": 12166.785
    },
    {
      "epoch": 0.5147454888230542,
      "grad_norm": 0.7569776972548438,
      "learning_rate": 8.453084758171297e-06,
      "loss": 0.3892237663269043,
      "num_input_tokens_seen": 243654280,
      "step": 7645,
      "train_runtime": 20025.4141,
      "train_tokens_per_second": 12167.253
    },
    {
      "epoch": 0.5150821438190143,
      "grad_norm": 0.778873899046102,
      "learning_rate": 8.451172020408532e-06,
      "loss": 0.39396061897277834,
      "num_input_tokens_seen": 243813424,
      "step": 7650,
      "train_runtime": 20038.4698,
      "train_tokens_per_second": 12167.268
    },
    {
      "epoch": 0.5154187988149744,
      "grad_norm": 0.7767233641799279,
      "learning_rate": 8.449258317536517e-06,
      "loss": 0.3607063293457031,
      "num_input_tokens_seen": 243974976,
      "step": 7655,
      "train_runtime": 20052.1346,
      "train_tokens_per_second": 12167.033
    },
    {
      "epoch": 0.5157554538109346,
      "grad_norm": 0.7795016196075504,
      "learning_rate": 8.447343650090412e-06,
      "loss": 0.3432271480560303,
      "num_input_tokens_seen": 244134640,
      "step": 7660,
      "train_runtime": 20065.2402,
      "train_tokens_per_second": 12167.043
    },
    {
      "epoch": 0.5160921088068947,
      "grad_norm": 0.7782332883208828,
      "learning_rate": 8.445428018605647e-06,
      "loss": 0.41146144866943357,
      "num_input_tokens_seen": 244297536,
      "step": 7665,
      "train_runtime": 20078.7232,
      "train_tokens_per_second": 12166.986
    },
    {
      "epoch": 0.5164287638028549,
      "grad_norm": 0.7414761064456209,
      "learning_rate": 8.443511423617924e-06,
      "loss": 0.35539951324462893,
      "num_input_tokens_seen": 244458608,
      "step": 7670,
      "train_runtime": 20091.7448,
      "train_tokens_per_second": 12167.117
    },
    {
      "epoch": 0.516765418798815,
      "grad_norm": 0.7815287739903983,
      "learning_rate": 8.441593865663213e-06,
      "loss": 0.4221913814544678,
      "num_input_tokens_seen": 244620464,
      "step": 7675,
      "train_runtime": 20104.9014,
      "train_tokens_per_second": 12167.205
    },
    {
      "epoch": 0.5171020737947751,
      "grad_norm": 0.7440374984488749,
      "learning_rate": 8.439675345277751e-06,
      "loss": 0.4159858703613281,
      "num_input_tokens_seen": 244782280,
      "step": 7680,
      "train_runtime": 20119.1111,
      "train_tokens_per_second": 12166.655
    },
    {
      "epoch": 0.5174387287907353,
      "grad_norm": 0.8014815644288046,
      "learning_rate": 8.437755862998047e-06,
      "loss": 0.39878134727478026,
      "num_input_tokens_seen": 244941944,
      "step": 7685,
      "train_runtime": 20132.4215,
      "train_tokens_per_second": 12166.542
    },
    {
      "epoch": 0.5177753837866954,
      "grad_norm": 0.7111023492510823,
      "learning_rate": 8.435835419360878e-06,
      "loss": 0.41121768951416016,
      "num_input_tokens_seen": 245103320,
      "step": 7690,
      "train_runtime": 20144.9166,
      "train_tokens_per_second": 12167.006
    },
    {
      "epoch": 0.5181120387826555,
      "grad_norm": 0.8123618429863106,
      "learning_rate": 8.433914014903291e-06,
      "loss": 0.4353185653686523,
      "num_input_tokens_seen": 245260784,
      "step": 7695,
      "train_runtime": 20157.9165,
      "train_tokens_per_second": 12166.971
    },
    {
      "epoch": 0.5184486937786157,
      "grad_norm": 0.79179170382196,
      "learning_rate": 8.431991650162599e-06,
      "loss": 0.37683467864990233,
      "num_input_tokens_seen": 245419456,
      "step": 7700,
      "train_runtime": 20171.232,
      "train_tokens_per_second": 12166.805
    },
    {
      "epoch": 0.5187853487745758,
      "grad_norm": 0.8091620590462157,
      "learning_rate": 8.430068325676386e-06,
      "loss": 0.3776644229888916,
      "num_input_tokens_seen": 245580880,
      "step": 7705,
      "train_runtime": 20185.2724,
      "train_tokens_per_second": 12166.34
    },
    {
      "epoch": 0.519122003770536,
      "grad_norm": 0.7498266146947521,
      "learning_rate": 8.428144041982502e-06,
      "loss": 0.40474853515625,
      "num_input_tokens_seen": 245738352,
      "step": 7710,
      "train_runtime": 20198.2936,
      "train_tokens_per_second": 12166.293
    },
    {
      "epoch": 0.5194586587664961,
      "grad_norm": 0.7138303446787942,
      "learning_rate": 8.42621879961907e-06,
      "loss": 0.37674245834350584,
      "num_input_tokens_seen": 245899776,
      "step": 7715,
      "train_runtime": 20210.7724,
      "train_tokens_per_second": 12166.768
    },
    {
      "epoch": 0.5197953137624562,
      "grad_norm": 0.7767770915280027,
      "learning_rate": 8.424292599124476e-06,
      "loss": 0.3740605592727661,
      "num_input_tokens_seen": 246061600,
      "step": 7720,
      "train_runtime": 20224.546,
      "train_tokens_per_second": 12166.483
    },
    {
      "epoch": 0.5201319687584164,
      "grad_norm": 0.9248925753056944,
      "learning_rate": 8.422365441037374e-06,
      "loss": 0.3987976789474487,
      "num_input_tokens_seen": 246221432,
      "step": 7725,
      "train_runtime": 20236.9882,
      "train_tokens_per_second": 12166.901
    },
    {
      "epoch": 0.5204686237543765,
      "grad_norm": 0.8200949380573862,
      "learning_rate": 8.420437325896692e-06,
      "loss": 0.40408992767333984,
      "num_input_tokens_seen": 246382304,
      "step": 7730,
      "train_runtime": 20249.9735,
      "train_tokens_per_second": 12167.043
    },
    {
      "epoch": 0.5208052787503367,
      "grad_norm": 0.8598375346576372,
      "learning_rate": 8.418508254241617e-06,
      "loss": 0.40617833137512205,
      "num_input_tokens_seen": 246542040,
      "step": 7735,
      "train_runtime": 20263.483,
      "train_tokens_per_second": 12166.815
    },
    {
      "epoch": 0.5211419337462968,
      "grad_norm": 0.7810428539264734,
      "learning_rate": 8.416578226611612e-06,
      "loss": 0.38080477714538574,
      "num_input_tokens_seen": 246700464,
      "step": 7740,
      "train_runtime": 20277.1689,
      "train_tokens_per_second": 12166.416
    },
    {
      "epoch": 0.5214785887422569,
      "grad_norm": 1.010595620595859,
      "learning_rate": 8.414647243546398e-06,
      "loss": 0.384970498085022,
      "num_input_tokens_seen": 246860784,
      "step": 7745,
      "train_runtime": 20290.5191,
      "train_tokens_per_second": 12166.312
    },
    {
      "epoch": 0.5218152437382171,
      "grad_norm": 0.7240710878438316,
      "learning_rate": 8.412715305585973e-06,
      "loss": 0.38633999824523924,
      "num_input_tokens_seen": 247021592,
      "step": 7750,
      "train_runtime": 20302.958,
      "train_tokens_per_second": 12166.778
    },
    {
      "epoch": 0.5221518987341772,
      "grad_norm": 0.7049935231055876,
      "learning_rate": 8.410782413270594e-06,
      "loss": 0.39238436222076417,
      "num_input_tokens_seen": 247185144,
      "step": 7755,
      "train_runtime": 20316.6095,
      "train_tokens_per_second": 12166.653
    },
    {
      "epoch": 0.5224885537301374,
      "grad_norm": 0.841964335887728,
      "learning_rate": 8.408848567140787e-06,
      "loss": 0.3709267616271973,
      "num_input_tokens_seen": 247335832,
      "step": 7760,
      "train_runtime": 20330.4889,
      "train_tokens_per_second": 12165.759
    },
    {
      "epoch": 0.5228252087260975,
      "grad_norm": 0.788929782171271,
      "learning_rate": 8.406913767737353e-06,
      "loss": 0.3649853229522705,
      "num_input_tokens_seen": 247495408,
      "step": 7765,
      "train_runtime": 20343.0868,
      "train_tokens_per_second": 12166.07
    },
    {
      "epoch": 0.5231618637220576,
      "grad_norm": 0.7624272221215714,
      "learning_rate": 8.404978015601344e-06,
      "loss": 0.3929929256439209,
      "num_input_tokens_seen": 247655200,
      "step": 7770,
      "train_runtime": 20355.5281,
      "train_tokens_per_second": 12166.484
    },
    {
      "epoch": 0.5234985187180178,
      "grad_norm": 0.7396381636756648,
      "learning_rate": 8.403041311274091e-06,
      "loss": 0.4017702579498291,
      "num_input_tokens_seen": 247817440,
      "step": 7775,
      "train_runtime": 20368.7975,
      "train_tokens_per_second": 12166.523
    },
    {
      "epoch": 0.5238351737139779,
      "grad_norm": 0.8371009855425124,
      "learning_rate": 8.401103655297188e-06,
      "loss": 0.38645689487457274,
      "num_input_tokens_seen": 247977560,
      "step": 7780,
      "train_runtime": 20382.3991,
      "train_tokens_per_second": 12166.26
    },
    {
      "epoch": 0.5241718287099381,
      "grad_norm": 0.7046093758482901,
      "learning_rate": 8.399165048212494e-06,
      "loss": 0.35761566162109376,
      "num_input_tokens_seen": 248133968,
      "step": 7785,
      "train_runtime": 20395.539,
      "train_tokens_per_second": 12166.09
    },
    {
      "epoch": 0.5245084837058982,
      "grad_norm": 0.816213114840001,
      "learning_rate": 8.397225490562131e-06,
      "loss": 0.41030006408691405,
      "num_input_tokens_seen": 248293760,
      "step": 7790,
      "train_runtime": 20408.6403,
      "train_tokens_per_second": 12166.11
    },
    {
      "epoch": 0.5248451387018583,
      "grad_norm": 0.833352904084274,
      "learning_rate": 8.395284982888494e-06,
      "loss": 0.3800528049468994,
      "num_input_tokens_seen": 248456616,
      "step": 7795,
      "train_runtime": 20422.0732,
      "train_tokens_per_second": 12166.082
    },
    {
      "epoch": 0.5251817936978185,
      "grad_norm": 0.9607875155809358,
      "learning_rate": 8.393343525734239e-06,
      "loss": 0.3983436346054077,
      "num_input_tokens_seen": 248618224,
      "step": 7800,
      "train_runtime": 20434.9908,
      "train_tokens_per_second": 12166.3
    },
    {
      "epoch": 0.5255184486937786,
      "grad_norm": 0.9147627066521322,
      "learning_rate": 8.391401119642286e-06,
      "loss": 0.3930044651031494,
      "num_input_tokens_seen": 248769512,
      "step": 7805,
      "train_runtime": 20448.1038,
      "train_tokens_per_second": 12165.896
    },
    {
      "epoch": 0.5258551036897388,
      "grad_norm": 0.7843803427735412,
      "learning_rate": 8.389457765155824e-06,
      "loss": 0.4199055671691895,
      "num_input_tokens_seen": 248924376,
      "step": 7810,
      "train_runtime": 20461.3065,
      "train_tokens_per_second": 12165.615
    },
    {
      "epoch": 0.5261917586856989,
      "grad_norm": 0.7646396350868451,
      "learning_rate": 8.387513462818309e-06,
      "loss": 0.3531463623046875,
      "num_input_tokens_seen": 249083936,
      "step": 7815,
      "train_runtime": 20473.8857,
      "train_tokens_per_second": 12165.934
    },
    {
      "epoch": 0.526528413681659,
      "grad_norm": 0.7768900595010912,
      "learning_rate": 8.385568213173453e-06,
      "loss": 0.3795043468475342,
      "num_input_tokens_seen": 249243224,
      "step": 7820,
      "train_runtime": 20488.0505,
      "train_tokens_per_second": 12165.297
    },
    {
      "epoch": 0.5268650686776192,
      "grad_norm": 0.7794627164528214,
      "learning_rate": 8.383622016765242e-06,
      "loss": 0.3782067060470581,
      "num_input_tokens_seen": 249405808,
      "step": 7825,
      "train_runtime": 20501.3971,
      "train_tokens_per_second": 12165.308
    },
    {
      "epoch": 0.5272017236735793,
      "grad_norm": 0.8385909978428195,
      "learning_rate": 8.381674874137926e-06,
      "loss": 0.403814697265625,
      "num_input_tokens_seen": 249564352,
      "step": 7830,
      "train_runtime": 20515.0477,
      "train_tokens_per_second": 12164.941
    },
    {
      "epoch": 0.5275383786695395,
      "grad_norm": 1.0122437021354653,
      "learning_rate": 8.379726785836013e-06,
      "loss": 0.361968994140625,
      "num_input_tokens_seen": 249726520,
      "step": 7835,
      "train_runtime": 20527.6346,
      "train_tokens_per_second": 12165.382
    },
    {
      "epoch": 0.5278750336654996,
      "grad_norm": 0.7678244244105163,
      "learning_rate": 8.377777752404283e-06,
      "loss": 0.38558340072631836,
      "num_input_tokens_seen": 249890360,
      "step": 7840,
      "train_runtime": 20540.6656,
      "train_tokens_per_second": 12165.641
    },
    {
      "epoch": 0.5282116886614597,
      "grad_norm": 0.8052327288520001,
      "learning_rate": 8.375827774387774e-06,
      "loss": 0.3559632062911987,
      "num_input_tokens_seen": 250043112,
      "step": 7845,
      "train_runtime": 20554.3468,
      "train_tokens_per_second": 12164.975
    },
    {
      "epoch": 0.5285483436574199,
      "grad_norm": 0.725806198965624,
      "learning_rate": 8.373876852331793e-06,
      "loss": 0.366229772567749,
      "num_input_tokens_seen": 250205824,
      "step": 7850,
      "train_runtime": 20567.0496,
      "train_tokens_per_second": 12165.373
    },
    {
      "epoch": 0.52888499865338,
      "grad_norm": 0.8520018416014798,
      "learning_rate": 8.371924986781909e-06,
      "loss": 0.40759763717651365,
      "num_input_tokens_seen": 250363328,
      "step": 7855,
      "train_runtime": 20580.2525,
      "train_tokens_per_second": 12165.221
    },
    {
      "epoch": 0.5292216536493402,
      "grad_norm": 0.8288492403392017,
      "learning_rate": 8.369972178283955e-06,
      "loss": 0.36505701541900637,
      "num_input_tokens_seen": 250526008,
      "step": 7860,
      "train_runtime": 20592.9496,
      "train_tokens_per_second": 12165.62
    },
    {
      "epoch": 0.5295583086453003,
      "grad_norm": 1.0179341615443855,
      "learning_rate": 8.368018427384027e-06,
      "loss": 0.3974146366119385,
      "num_input_tokens_seen": 250687128,
      "step": 7865,
      "train_runtime": 20605.8176,
      "train_tokens_per_second": 12165.842
    },
    {
      "epoch": 0.5298949636412604,
      "grad_norm": 0.8117119583218815,
      "learning_rate": 8.366063734628485e-06,
      "loss": 0.37839417457580565,
      "num_input_tokens_seen": 250846720,
      "step": 7870,
      "train_runtime": 20619.0419,
      "train_tokens_per_second": 12165.78
    },
    {
      "epoch": 0.5302316186372206,
      "grad_norm": 0.8170966197918764,
      "learning_rate": 8.364108100563955e-06,
      "loss": 0.39380273818969724,
      "num_input_tokens_seen": 251008152,
      "step": 7875,
      "train_runtime": 20632.5666,
      "train_tokens_per_second": 12165.629
    },
    {
      "epoch": 0.5305682736331807,
      "grad_norm": 0.8255489516799258,
      "learning_rate": 8.362151525737321e-06,
      "loss": 0.43450231552124025,
      "num_input_tokens_seen": 251168832,
      "step": 7880,
      "train_runtime": 20644.9953,
      "train_tokens_per_second": 12166.088
    },
    {
      "epoch": 0.5309049286291408,
      "grad_norm": 0.8077496068379889,
      "learning_rate": 8.360194010695733e-06,
      "loss": 0.36590213775634767,
      "num_input_tokens_seen": 251328480,
      "step": 7885,
      "train_runtime": 20658.4251,
      "train_tokens_per_second": 12165.907
    },
    {
      "epoch": 0.531241583625101,
      "grad_norm": 0.7633973042774351,
      "learning_rate": 8.358235555986607e-06,
      "loss": 0.39882049560546873,
      "num_input_tokens_seen": 251487248,
      "step": 7890,
      "train_runtime": 20671.0092,
      "train_tokens_per_second": 12166.181
    },
    {
      "epoch": 0.5315782386210611,
      "grad_norm": 0.8283702885559234,
      "learning_rate": 8.356276162157613e-06,
      "loss": 0.36642003059387207,
      "num_input_tokens_seen": 251648360,
      "step": 7895,
      "train_runtime": 20683.57,
      "train_tokens_per_second": 12166.582
    },
    {
      "epoch": 0.5319148936170213,
      "grad_norm": 0.7739911236375652,
      "learning_rate": 8.354315829756696e-06,
      "loss": 0.3816749095916748,
      "num_input_tokens_seen": 251811016,
      "step": 7900,
      "train_runtime": 20696.8503,
      "train_tokens_per_second": 12166.635
    },
    {
      "epoch": 0.5322515486129814,
      "grad_norm": 0.7413155522951834,
      "learning_rate": 8.352354559332053e-06,
      "loss": 0.3656564235687256,
      "num_input_tokens_seen": 251967920,
      "step": 7905,
      "train_runtime": 20709.649,
      "train_tokens_per_second": 12166.692
    },
    {
      "epoch": 0.5325882036089415,
      "grad_norm": 0.7414665289152079,
      "learning_rate": 8.350392351432146e-06,
      "loss": 0.3953502893447876,
      "num_input_tokens_seen": 252123224,
      "step": 7910,
      "train_runtime": 20722.866,
      "train_tokens_per_second": 12166.426
    },
    {
      "epoch": 0.5329248586049017,
      "grad_norm": 0.8209203115906711,
      "learning_rate": 8.348429206605702e-06,
      "loss": 0.377488112449646,
      "num_input_tokens_seen": 252287064,
      "step": 7915,
      "train_runtime": 20735.2911,
      "train_tokens_per_second": 12167.037
    },
    {
      "epoch": 0.5332615136008618,
      "grad_norm": 0.9159843598480798,
      "learning_rate": 8.346465125401706e-06,
      "loss": 0.42546706199645995,
      "num_input_tokens_seen": 252444448,
      "step": 7920,
      "train_runtime": 20748.3336,
      "train_tokens_per_second": 12166.975
    },
    {
      "epoch": 0.533598168596822,
      "grad_norm": 0.7761276140410946,
      "learning_rate": 8.344500108369412e-06,
      "loss": 0.3723264217376709,
      "num_input_tokens_seen": 252607952,
      "step": 7925,
      "train_runtime": 20762.0531,
      "train_tokens_per_second": 12166.81
    },
    {
      "epoch": 0.5339348235927821,
      "grad_norm": 0.8042724923676196,
      "learning_rate": 8.342534156058326e-06,
      "loss": 0.4076098442077637,
      "num_input_tokens_seen": 252768032,
      "step": 7930,
      "train_runtime": 20775.4302,
      "train_tokens_per_second": 12166.681
    },
    {
      "epoch": 0.5342714785887422,
      "grad_norm": 0.8605619741915326,
      "learning_rate": 8.34056726901822e-06,
      "loss": 0.39628233909606936,
      "num_input_tokens_seen": 252928512,
      "step": 7935,
      "train_runtime": 20788.7648,
      "train_tokens_per_second": 12166.596
    },
    {
      "epoch": 0.5346081335847024,
      "grad_norm": 0.7876151311590377,
      "learning_rate": 8.338599447799132e-06,
      "loss": 0.3728104829788208,
      "num_input_tokens_seen": 253080656,
      "step": 7940,
      "train_runtime": 20801.8402,
      "train_tokens_per_second": 12166.263
    },
    {
      "epoch": 0.5349447885806625,
      "grad_norm": 0.8682662531395543,
      "learning_rate": 8.336630692951353e-06,
      "loss": 0.397021222114563,
      "num_input_tokens_seen": 253241440,
      "step": 7945,
      "train_runtime": 20814.8163,
      "train_tokens_per_second": 12166.403
    },
    {
      "epoch": 0.5352814435766227,
      "grad_norm": 0.8177289067366508,
      "learning_rate": 8.334661005025441e-06,
      "loss": 0.40717229843139646,
      "num_input_tokens_seen": 253399952,
      "step": 7950,
      "train_runtime": 20827.8131,
      "train_tokens_per_second": 12166.421
    },
    {
      "epoch": 0.5356180985725828,
      "grad_norm": 0.8782876265032664,
      "learning_rate": 8.332690384572213e-06,
      "loss": 0.4144033432006836,
      "num_input_tokens_seen": 253558104,
      "step": 7955,
      "train_runtime": 20841.1597,
      "train_tokens_per_second": 12166.219
    },
    {
      "epoch": 0.5359547535685429,
      "grad_norm": 0.7365889529271511,
      "learning_rate": 8.330718832142746e-06,
      "loss": 0.3852388381958008,
      "num_input_tokens_seen": 253721936,
      "step": 7960,
      "train_runtime": 20854.2739,
      "train_tokens_per_second": 12166.424
    },
    {
      "epoch": 0.5362914085645031,
      "grad_norm": 0.7807756560653858,
      "learning_rate": 8.328746348288379e-06,
      "loss": 0.3837946891784668,
      "num_input_tokens_seen": 253879912,
      "step": 7965,
      "train_runtime": 20867.9567,
      "train_tokens_per_second": 12166.017
    },
    {
      "epoch": 0.5366280635604632,
      "grad_norm": 0.8000760057512833,
      "learning_rate": 8.32677293356071e-06,
      "loss": 0.4331723690032959,
      "num_input_tokens_seen": 254043312,
      "step": 7970,
      "train_runtime": 20880.9469,
      "train_tokens_per_second": 12166.274
    },
    {
      "epoch": 0.5369647185564234,
      "grad_norm": 0.7930234163833042,
      "learning_rate": 8.324798588511598e-06,
      "loss": 0.38106508255004884,
      "num_input_tokens_seen": 254201456,
      "step": 7975,
      "train_runtime": 20894.9202,
      "train_tokens_per_second": 12165.706
    },
    {
      "epoch": 0.5373013735523835,
      "grad_norm": 0.7903696270192934,
      "learning_rate": 8.322823313693162e-06,
      "loss": 0.39886314868927003,
      "num_input_tokens_seen": 254357256,
      "step": 7980,
      "train_runtime": 20907.6205,
      "train_tokens_per_second": 12165.768
    },
    {
      "epoch": 0.5376380285483436,
      "grad_norm": 0.7829286849514997,
      "learning_rate": 8.320847109657782e-06,
      "loss": 0.38727149963378904,
      "num_input_tokens_seen": 254518128,
      "step": 7985,
      "train_runtime": 20920.4799,
      "train_tokens_per_second": 12165.979
    },
    {
      "epoch": 0.5379746835443038,
      "grad_norm": 0.7351681774217423,
      "learning_rate": 8.318869976958099e-06,
      "loss": 0.34882638454437254,
      "num_input_tokens_seen": 254680256,
      "step": 7990,
      "train_runtime": 20933.3133,
      "train_tokens_per_second": 12166.266
    },
    {
      "epoch": 0.5383113385402639,
      "grad_norm": 0.8033044200957941,
      "learning_rate": 8.316891916147007e-06,
      "loss": 0.42199172973632815,
      "num_input_tokens_seen": 254839392,
      "step": 7995,
      "train_runtime": 20946.1093,
      "train_tokens_per_second": 12166.431
    },
    {
      "epoch": 0.5386479935362241,
      "grad_norm": 0.8416378503675388,
      "learning_rate": 8.314912927777671e-06,
      "loss": 0.3883615016937256,
      "num_input_tokens_seen": 254997072,
      "step": 8000,
      "train_runtime": 20958.6464,
      "train_tokens_per_second": 12166.677
    },
    {
      "epoch": 0.5389846485321842,
      "grad_norm": 0.8290087239270112,
      "learning_rate": 8.312933012403503e-06,
      "loss": 0.38657207489013673,
      "num_input_tokens_seen": 255149992,
      "step": 8005,
      "train_runtime": 20971.6178,
      "train_tokens_per_second": 12166.443
    },
    {
      "epoch": 0.5393213035281443,
      "grad_norm": 0.7456774013033717,
      "learning_rate": 8.310952170578184e-06,
      "loss": 0.39329233169555666,
      "num_input_tokens_seen": 255305648,
      "step": 8010,
      "train_runtime": 20984.7562,
      "train_tokens_per_second": 12166.243
    },
    {
      "epoch": 0.5396579585241045,
      "grad_norm": 0.8638884996816859,
      "learning_rate": 8.308970402855647e-06,
      "loss": 0.36403968334198,
      "num_input_tokens_seen": 255459520,
      "step": 8015,
      "train_runtime": 20997.9482,
      "train_tokens_per_second": 12165.928
    },
    {
      "epoch": 0.5399946135200646,
      "grad_norm": 0.8921896241768411,
      "learning_rate": 8.306987709790087e-06,
      "loss": 0.36961803436279295,
      "num_input_tokens_seen": 255617080,
      "step": 8020,
      "train_runtime": 21010.6946,
      "train_tokens_per_second": 12166.046
    },
    {
      "epoch": 0.5403312685160248,
      "grad_norm": 0.7884543600476868,
      "learning_rate": 8.305004091935962e-06,
      "loss": 0.38249356746673585,
      "num_input_tokens_seen": 255779960,
      "step": 8025,
      "train_runtime": 21023.8214,
      "train_tokens_per_second": 12166.197
    },
    {
      "epoch": 0.5406679235119849,
      "grad_norm": 0.8576962211180145,
      "learning_rate": 8.303019549847979e-06,
      "loss": 0.4121133804321289,
      "num_input_tokens_seen": 255939616,
      "step": 8030,
      "train_runtime": 21036.9758,
      "train_tokens_per_second": 12166.179
    },
    {
      "epoch": 0.541004578507945,
      "grad_norm": 0.8077388305870126,
      "learning_rate": 8.301034084081114e-06,
      "loss": 0.366147518157959,
      "num_input_tokens_seen": 256101712,
      "step": 8035,
      "train_runtime": 21050.2145,
      "train_tokens_per_second": 12166.228
    },
    {
      "epoch": 0.5413412335039052,
      "grad_norm": 0.9355436403969812,
      "learning_rate": 8.299047695190592e-06,
      "loss": 0.41315793991088867,
      "num_input_tokens_seen": 256260216,
      "step": 8040,
      "train_runtime": 21062.8995,
      "train_tokens_per_second": 12166.426
    },
    {
      "epoch": 0.5416778884998653,
      "grad_norm": 0.8029423840121809,
      "learning_rate": 8.297060383731903e-06,
      "loss": 0.39433751106262205,
      "num_input_tokens_seen": 256419136,
      "step": 8045,
      "train_runtime": 21075.9222,
      "train_tokens_per_second": 12166.449
    },
    {
      "epoch": 0.5420145434958255,
      "grad_norm": 0.7976789965945154,
      "learning_rate": 8.29507215026079e-06,
      "loss": 0.42723779678344725,
      "num_input_tokens_seen": 256578472,
      "step": 8050,
      "train_runtime": 21089.028,
      "train_tokens_per_second": 12166.444
    },
    {
      "epoch": 0.5423511984917856,
      "grad_norm": 0.9088627087336268,
      "learning_rate": 8.293082995333257e-06,
      "loss": 0.37396821975708006,
      "num_input_tokens_seen": 256738496,
      "step": 8055,
      "train_runtime": 21101.7723,
      "train_tokens_per_second": 12166.679
    },
    {
      "epoch": 0.5426878534877457,
      "grad_norm": 0.7719170522386951,
      "learning_rate": 8.291092919505568e-06,
      "loss": 0.35263934135437014,
      "num_input_tokens_seen": 256895880,
      "step": 8060,
      "train_runtime": 21114.8952,
      "train_tokens_per_second": 12166.571
    },
    {
      "epoch": 0.5430245084837059,
      "grad_norm": 0.799782147518969,
      "learning_rate": 8.289101923334237e-06,
      "loss": 0.3940428733825684,
      "num_input_tokens_seen": 257057392,
      "step": 8065,
      "train_runtime": 21128.5026,
      "train_tokens_per_second": 12166.38
    },
    {
      "epoch": 0.543361163479666,
      "grad_norm": 0.8608741107233793,
      "learning_rate": 8.28711000737604e-06,
      "loss": 0.38349123001098634,
      "num_input_tokens_seen": 257216648,
      "step": 8070,
      "train_runtime": 21142.0974,
      "train_tokens_per_second": 12166.089
    },
    {
      "epoch": 0.5436978184756261,
      "grad_norm": 0.8124190970204751,
      "learning_rate": 8.285117172188013e-06,
      "loss": 0.4074282646179199,
      "num_input_tokens_seen": 257379272,
      "step": 8075,
      "train_runtime": 21154.8627,
      "train_tokens_per_second": 12166.435
    },
    {
      "epoch": 0.5440344734715863,
      "grad_norm": 0.9612535327891233,
      "learning_rate": 8.283123418327443e-06,
      "loss": 0.4188706398010254,
      "num_input_tokens_seen": 257538168,
      "step": 8080,
      "train_runtime": 21167.4838,
      "train_tokens_per_second": 12166.688
    },
    {
      "epoch": 0.5443711284675464,
      "grad_norm": 0.7760901263834276,
      "learning_rate": 8.281128746351878e-06,
      "loss": 0.3785744905471802,
      "num_input_tokens_seen": 257699888,
      "step": 8085,
      "train_runtime": 21180.4504,
      "train_tokens_per_second": 12166.875
    },
    {
      "epoch": 0.5447077834635066,
      "grad_norm": 0.7636941815034735,
      "learning_rate": 8.279133156819123e-06,
      "loss": 0.41728720664978025,
      "num_input_tokens_seen": 257857016,
      "step": 8090,
      "train_runtime": 21193.4221,
      "train_tokens_per_second": 12166.842
    },
    {
      "epoch": 0.5450444384594667,
      "grad_norm": 0.7904308387858431,
      "learning_rate": 8.277136650287237e-06,
      "loss": 0.36835551261901855,
      "num_input_tokens_seen": 258020800,
      "step": 8095,
      "train_runtime": 21206.5193,
      "train_tokens_per_second": 12167.051
    },
    {
      "epoch": 0.5453810934554268,
      "grad_norm": 0.8753312039925617,
      "learning_rate": 8.275139227314537e-06,
      "loss": 0.38594019412994385,
      "num_input_tokens_seen": 258179016,
      "step": 8100,
      "train_runtime": 21219.9491,
      "train_tokens_per_second": 12166.807
    },
    {
      "epoch": 0.545717748451387,
      "grad_norm": 0.876517530377018,
      "learning_rate": 8.273140888459594e-06,
      "loss": 0.4352447509765625,
      "num_input_tokens_seen": 258339720,
      "step": 8105,
      "train_runtime": 21232.578,
      "train_tokens_per_second": 12167.139
    },
    {
      "epoch": 0.5460544034473471,
      "grad_norm": 0.8571059235986455,
      "learning_rate": 8.27114163428124e-06,
      "loss": 0.3966317653656006,
      "num_input_tokens_seen": 258496768,
      "step": 8110,
      "train_runtime": 21245.3685,
      "train_tokens_per_second": 12167.206
    },
    {
      "epoch": 0.5463910584433073,
      "grad_norm": 0.8191605218307668,
      "learning_rate": 8.26914146533856e-06,
      "loss": 0.3764180660247803,
      "num_input_tokens_seen": 258659288,
      "step": 8115,
      "train_runtime": 21258.05,
      "train_tokens_per_second": 12167.592
    },
    {
      "epoch": 0.5467277134392674,
      "grad_norm": 0.7076217618322249,
      "learning_rate": 8.267140382190892e-06,
      "loss": 0.38625895977020264,
      "num_input_tokens_seen": 258822600,
      "step": 8120,
      "train_runtime": 21270.9505,
      "train_tokens_per_second": 12167.891
    },
    {
      "epoch": 0.5470643684352275,
      "grad_norm": 0.6873517575005611,
      "learning_rate": 8.265138385397835e-06,
      "loss": 0.38873815536499023,
      "num_input_tokens_seen": 258984208,
      "step": 8125,
      "train_runtime": 21283.9922,
      "train_tokens_per_second": 12168.028
    },
    {
      "epoch": 0.5474010234311877,
      "grad_norm": 0.7065896847522521,
      "learning_rate": 8.263135475519241e-06,
      "loss": 0.3519615173339844,
      "num_input_tokens_seen": 259147520,
      "step": 8130,
      "train_runtime": 21296.9705,
      "train_tokens_per_second": 12168.281
    },
    {
      "epoch": 0.5477376784271478,
      "grad_norm": 0.7478595008927922,
      "learning_rate": 8.261131653115216e-06,
      "loss": 0.41793012619018555,
      "num_input_tokens_seen": 259309360,
      "step": 8135,
      "train_runtime": 21309.6629,
      "train_tokens_per_second": 12168.628
    },
    {
      "epoch": 0.548074333423108,
      "grad_norm": 0.7429522179656578,
      "learning_rate": 8.259126918746122e-06,
      "loss": 0.3670862436294556,
      "num_input_tokens_seen": 259468888,
      "step": 8140,
      "train_runtime": 21322.1509,
      "train_tokens_per_second": 12168.983
    },
    {
      "epoch": 0.5484109884190681,
      "grad_norm": 0.7661716108119748,
      "learning_rate": 8.25712127297258e-06,
      "loss": 0.4059457778930664,
      "num_input_tokens_seen": 259630856,
      "step": 8145,
      "train_runtime": 21335.3619,
      "train_tokens_per_second": 12169.039
    },
    {
      "epoch": 0.5487476434150282,
      "grad_norm": 0.8370165993133727,
      "learning_rate": 8.255114716355459e-06,
      "loss": 0.41359848976135255,
      "num_input_tokens_seen": 259793512,
      "step": 8150,
      "train_runtime": 21348.073,
      "train_tokens_per_second": 12169.413
    },
    {
      "epoch": 0.5490842984109884,
      "grad_norm": 0.7143909078266771,
      "learning_rate": 8.253107249455887e-06,
      "loss": 0.40358386039733884,
      "num_input_tokens_seen": 259952888,
      "step": 8155,
      "train_runtime": 21361.2066,
      "train_tokens_per_second": 12169.392
    },
    {
      "epoch": 0.5494209534069485,
      "grad_norm": 0.7882241759978242,
      "learning_rate": 8.251098872835247e-06,
      "loss": 0.39182052612304685,
      "num_input_tokens_seen": 260113960,
      "step": 8160,
      "train_runtime": 21373.6647,
      "train_tokens_per_second": 12169.834
    },
    {
      "epoch": 0.5497576084029087,
      "grad_norm": 0.6747458733466716,
      "learning_rate": 8.249089587055175e-06,
      "loss": 0.35428686141967775,
      "num_input_tokens_seen": 260271968,
      "step": 8165,
      "train_runtime": 21386.5977,
      "train_tokens_per_second": 12169.863
    },
    {
      "epoch": 0.5500942633988688,
      "grad_norm": 0.7128552139516158,
      "learning_rate": 8.247079392677558e-06,
      "loss": 0.38996999263763427,
      "num_input_tokens_seen": 260431992,
      "step": 8170,
      "train_runtime": 21399.4532,
      "train_tokens_per_second": 12170.03
    },
    {
      "epoch": 0.5504309183948289,
      "grad_norm": 0.7517140615498464,
      "learning_rate": 8.245068290264544e-06,
      "loss": 0.3815065860748291,
      "num_input_tokens_seen": 260592392,
      "step": 8175,
      "train_runtime": 21411.8849,
      "train_tokens_per_second": 12170.455
    },
    {
      "epoch": 0.5507675733907891,
      "grad_norm": 0.8416414791589104,
      "learning_rate": 8.243056280378531e-06,
      "loss": 0.39004452228546144,
      "num_input_tokens_seen": 260755024,
      "step": 8180,
      "train_runtime": 21424.8351,
      "train_tokens_per_second": 12170.69
    },
    {
      "epoch": 0.5511042283867492,
      "grad_norm": 0.7468734694875759,
      "learning_rate": 8.241043363582171e-06,
      "loss": 0.35741243362426756,
      "num_input_tokens_seen": 260917784,
      "step": 8185,
      "train_runtime": 21437.3025,
      "train_tokens_per_second": 12171.204
    },
    {
      "epoch": 0.5514408833827094,
      "grad_norm": 0.7103716877935099,
      "learning_rate": 8.239029540438369e-06,
      "loss": 0.35527615547180175,
      "num_input_tokens_seen": 261080520,
      "step": 8190,
      "train_runtime": 21450.0333,
      "train_tokens_per_second": 12171.567
    },
    {
      "epoch": 0.5517775383786695,
      "grad_norm": 0.7554093517707741,
      "learning_rate": 8.237014811510284e-06,
      "loss": 0.3753568172454834,
      "num_input_tokens_seen": 261242352,
      "step": 8195,
      "train_runtime": 21463.3012,
      "train_tokens_per_second": 12171.583
    },
    {
      "epoch": 0.5521141933746296,
      "grad_norm": 0.8373822534978808,
      "learning_rate": 8.234999177361329e-06,
      "loss": 0.3857575416564941,
      "num_input_tokens_seen": 261404840,
      "step": 8200,
      "train_runtime": 21476.1197,
      "train_tokens_per_second": 12171.884
    },
    {
      "epoch": 0.5524508483705898,
      "grad_norm": 0.802756688777733,
      "learning_rate": 8.232982638555169e-06,
      "loss": 0.4095576286315918,
      "num_input_tokens_seen": 261565560,
      "step": 8205,
      "train_runtime": 21489.4729,
      "train_tokens_per_second": 12171.8
    },
    {
      "epoch": 0.5527875033665499,
      "grad_norm": 0.722754241987713,
      "learning_rate": 8.230965195655724e-06,
      "loss": 0.3943202972412109,
      "num_input_tokens_seen": 261729400,
      "step": 8210,
      "train_runtime": 21501.9135,
      "train_tokens_per_second": 12172.377
    },
    {
      "epoch": 0.55312415836251,
      "grad_norm": 0.8695634418914352,
      "learning_rate": 8.228946849227167e-06,
      "loss": 0.4108874320983887,
      "num_input_tokens_seen": 261893072,
      "step": 8215,
      "train_runtime": 21514.9726,
      "train_tokens_per_second": 12172.596
    },
    {
      "epoch": 0.5534608133584702,
      "grad_norm": 0.7716735366927598,
      "learning_rate": 8.226927599833916e-06,
      "loss": 0.4123076438903809,
      "num_input_tokens_seen": 262056072,
      "step": 8220,
      "train_runtime": 21527.8144,
      "train_tokens_per_second": 12172.907
    },
    {
      "epoch": 0.5537974683544303,
      "grad_norm": 0.8315520614140791,
      "learning_rate": 8.224907448040655e-06,
      "loss": 0.38757858276367185,
      "num_input_tokens_seen": 262219288,
      "step": 8225,
      "train_runtime": 21540.7157,
      "train_tokens_per_second": 12173.193
    },
    {
      "epoch": 0.5541341233503905,
      "grad_norm": 0.734912972299796,
      "learning_rate": 8.222886394412309e-06,
      "loss": 0.40882530212402346,
      "num_input_tokens_seen": 262379536,
      "step": 8230,
      "train_runtime": 21554.7448,
      "train_tokens_per_second": 12172.704
    },
    {
      "epoch": 0.5544707783463506,
      "grad_norm": 0.7564290211402108,
      "learning_rate": 8.220864439514057e-06,
      "loss": 0.3858541488647461,
      "num_input_tokens_seen": 262540024,
      "step": 8235,
      "train_runtime": 21567.7596,
      "train_tokens_per_second": 12172.8
    },
    {
      "epoch": 0.5548074333423108,
      "grad_norm": 0.8271864431813577,
      "learning_rate": 8.218841583911338e-06,
      "loss": 0.36765623092651367,
      "num_input_tokens_seen": 262703544,
      "step": 8240,
      "train_runtime": 21581.4534,
      "train_tokens_per_second": 12172.653
    },
    {
      "epoch": 0.5551440883382709,
      "grad_norm": 0.7594293827762876,
      "learning_rate": 8.216817828169834e-06,
      "loss": 0.41515312194824217,
      "num_input_tokens_seen": 262862680,
      "step": 8245,
      "train_runtime": 21594.5782,
      "train_tokens_per_second": 12172.624
    },
    {
      "epoch": 0.555480743334231,
      "grad_norm": 0.9611220663082597,
      "learning_rate": 8.214793172855481e-06,
      "loss": 0.3621303796768188,
      "num_input_tokens_seen": 263024248,
      "step": 8250,
      "train_runtime": 21608.3432,
      "train_tokens_per_second": 12172.347
    },
    {
      "epoch": 0.5558173983301912,
      "grad_norm": 0.6898171571625179,
      "learning_rate": 8.21276761853447e-06,
      "loss": 0.3664684534072876,
      "num_input_tokens_seen": 263185984,
      "step": 8255,
      "train_runtime": 21621.8566,
      "train_tokens_per_second": 12172.219
    },
    {
      "epoch": 0.5561540533261513,
      "grad_norm": 0.8783686407005666,
      "learning_rate": 8.210741165773238e-06,
      "loss": 0.39526987075805664,
      "num_input_tokens_seen": 263348280,
      "step": 8260,
      "train_runtime": 21634.5167,
      "train_tokens_per_second": 12172.598
    },
    {
      "epoch": 0.5564907083221114,
      "grad_norm": 0.7790868136221593,
      "learning_rate": 8.208713815138476e-06,
      "loss": 0.3606027364730835,
      "num_input_tokens_seen": 263511104,
      "step": 8265,
      "train_runtime": 21647.272,
      "train_tokens_per_second": 12172.947
    },
    {
      "epoch": 0.5568273633180716,
      "grad_norm": 0.7176895296935734,
      "learning_rate": 8.20668556719713e-06,
      "loss": 0.36252636909484864,
      "num_input_tokens_seen": 263667440,
      "step": 8270,
      "train_runtime": 21661.1473,
      "train_tokens_per_second": 12172.367
    },
    {
      "epoch": 0.5571640183140317,
      "grad_norm": 0.8265667546030943,
      "learning_rate": 8.204656422516388e-06,
      "loss": 0.3659949541091919,
      "num_input_tokens_seen": 263822664,
      "step": 8275,
      "train_runtime": 21673.9371,
      "train_tokens_per_second": 12172.346
    },
    {
      "epoch": 0.5575006733099919,
      "grad_norm": 0.8184529864280441,
      "learning_rate": 8.2026263816637e-06,
      "loss": 0.3745150566101074,
      "num_input_tokens_seen": 263977896,
      "step": 8280,
      "train_runtime": 21687.1232,
      "train_tokens_per_second": 12172.103
    },
    {
      "epoch": 0.557837328305952,
      "grad_norm": 0.8032973510381942,
      "learning_rate": 8.200595445206753e-06,
      "loss": 0.3884341955184937,
      "num_input_tokens_seen": 264137152,
      "step": 8285,
      "train_runtime": 21700.3594,
      "train_tokens_per_second": 12172.017
    },
    {
      "epoch": 0.5581739833019121,
      "grad_norm": 0.8485317228465236,
      "learning_rate": 8.198563613713497e-06,
      "loss": 0.37020292282104494,
      "num_input_tokens_seen": 264292776,
      "step": 8290,
      "train_runtime": 21713.2594,
      "train_tokens_per_second": 12171.953
    },
    {
      "epoch": 0.5585106382978723,
      "grad_norm": 0.8949440536013235,
      "learning_rate": 8.196530887752127e-06,
      "loss": 0.3872734546661377,
      "num_input_tokens_seen": 264454128,
      "step": 8295,
      "train_runtime": 21726.9007,
      "train_tokens_per_second": 12171.737
    },
    {
      "epoch": 0.5588472932938324,
      "grad_norm": 0.8109015537228611,
      "learning_rate": 8.194497267891086e-06,
      "loss": 0.3815859317779541,
      "num_input_tokens_seen": 264616936,
      "step": 8300,
      "train_runtime": 21739.7694,
      "train_tokens_per_second": 12172.021
    },
    {
      "epoch": 0.5591839482897926,
      "grad_norm": 0.8124142643204788,
      "learning_rate": 8.19246275469907e-06,
      "loss": 0.3873375415802002,
      "num_input_tokens_seen": 264779592,
      "step": 8305,
      "train_runtime": 21752.4712,
      "train_tokens_per_second": 12172.391
    },
    {
      "epoch": 0.5595206032857528,
      "grad_norm": 0.756754880722595,
      "learning_rate": 8.190427348745027e-06,
      "loss": 0.3874617338180542,
      "num_input_tokens_seen": 264938912,
      "step": 8310,
      "train_runtime": 21765.3057,
      "train_tokens_per_second": 12172.533
    },
    {
      "epoch": 0.559857258281713,
      "grad_norm": 0.8239454837314649,
      "learning_rate": 8.188391050598146e-06,
      "loss": 0.3656805753707886,
      "num_input_tokens_seen": 265102752,
      "step": 8315,
      "train_runtime": 21777.7362,
      "train_tokens_per_second": 12173.109
    },
    {
      "epoch": 0.5601939132776731,
      "grad_norm": 0.8185450853395417,
      "learning_rate": 8.186353860827878e-06,
      "loss": 0.37709674835205076,
      "num_input_tokens_seen": 265257952,
      "step": 8320,
      "train_runtime": 21791.0779,
      "train_tokens_per_second": 12172.778
    },
    {
      "epoch": 0.5605305682736332,
      "grad_norm": 0.7124691940502255,
      "learning_rate": 8.184315780003911e-06,
      "loss": 0.39144511222839357,
      "num_input_tokens_seen": 265419888,
      "step": 8325,
      "train_runtime": 21804.2674,
      "train_tokens_per_second": 12172.841
    },
    {
      "epoch": 0.5608672232695934,
      "grad_norm": 0.7229200696217318,
      "learning_rate": 8.182276808696193e-06,
      "loss": 0.4190876007080078,
      "num_input_tokens_seen": 265573272,
      "step": 8330,
      "train_runtime": 21817.4744,
      "train_tokens_per_second": 12172.503
    },
    {
      "epoch": 0.5612038782655535,
      "grad_norm": 0.6850738096416372,
      "learning_rate": 8.18023694747491e-06,
      "loss": 0.3996566295623779,
      "num_input_tokens_seen": 265735536,
      "step": 8335,
      "train_runtime": 21830.2524,
      "train_tokens_per_second": 12172.811
    },
    {
      "epoch": 0.5615405332615137,
      "grad_norm": 0.9012775793624178,
      "learning_rate": 8.178196196910508e-06,
      "loss": 0.37130112648010255,
      "num_input_tokens_seen": 265898448,
      "step": 8340,
      "train_runtime": 21843.0524,
      "train_tokens_per_second": 12173.136
    },
    {
      "epoch": 0.5618771882574738,
      "grad_norm": 0.8285357952897106,
      "learning_rate": 8.176154557573672e-06,
      "loss": 0.41409831047058104,
      "num_input_tokens_seen": 266059472,
      "step": 8345,
      "train_runtime": 21855.5106,
      "train_tokens_per_second": 12173.565
    },
    {
      "epoch": 0.5622138432534339,
      "grad_norm": 0.8202784248577092,
      "learning_rate": 8.174112030035344e-06,
      "loss": 0.3550588130950928,
      "num_input_tokens_seen": 266221024,
      "step": 8350,
      "train_runtime": 21868.5725,
      "train_tokens_per_second": 12173.681
    },
    {
      "epoch": 0.5625504982493941,
      "grad_norm": 0.7442094656739202,
      "learning_rate": 8.172068614866707e-06,
      "loss": 0.3731220245361328,
      "num_input_tokens_seen": 266373080,
      "step": 8355,
      "train_runtime": 21881.6034,
      "train_tokens_per_second": 12173.38
    },
    {
      "epoch": 0.5628871532453542,
      "grad_norm": 0.8153921212340124,
      "learning_rate": 8.170024312639197e-06,
      "loss": 0.4146949291229248,
      "num_input_tokens_seen": 266535376,
      "step": 8360,
      "train_runtime": 21894.812,
      "train_tokens_per_second": 12173.449
    },
    {
      "epoch": 0.5632238082413143,
      "grad_norm": 0.7276035779833736,
      "learning_rate": 8.167979123924496e-06,
      "loss": 0.4149846076965332,
      "num_input_tokens_seen": 266694712,
      "step": 8365,
      "train_runtime": 21907.581,
      "train_tokens_per_second": 12173.627
    },
    {
      "epoch": 0.5635604632372745,
      "grad_norm": 0.835081908225235,
      "learning_rate": 8.165933049294537e-06,
      "loss": 0.4069093704223633,
      "num_input_tokens_seen": 266852256,
      "step": 8370,
      "train_runtime": 21920.7325,
      "train_tokens_per_second": 12173.51
    },
    {
      "epoch": 0.5638971182332346,
      "grad_norm": 0.8827699456387667,
      "learning_rate": 8.163886089321493e-06,
      "loss": 0.3603796482086182,
      "num_input_tokens_seen": 267014304,
      "step": 8375,
      "train_runtime": 21933.2709,
      "train_tokens_per_second": 12173.939
    },
    {
      "epoch": 0.5642337732291948,
      "grad_norm": 0.7270849273987028,
      "learning_rate": 8.161838244577795e-06,
      "loss": 0.3624864101409912,
      "num_input_tokens_seen": 267175920,
      "step": 8380,
      "train_runtime": 21945.9983,
      "train_tokens_per_second": 12174.243
    },
    {
      "epoch": 0.5645704282251549,
      "grad_norm": 0.7370969189073313,
      "learning_rate": 8.159789515636114e-06,
      "loss": 0.3571066379547119,
      "num_input_tokens_seen": 267331512,
      "step": 8385,
      "train_runtime": 21958.7734,
      "train_tokens_per_second": 12174.246
    },
    {
      "epoch": 0.564907083221115,
      "grad_norm": 0.8381292867158753,
      "learning_rate": 8.15773990306937e-06,
      "loss": 0.39977755546569826,
      "num_input_tokens_seen": 267485704,
      "step": 8390,
      "train_runtime": 21972.0191,
      "train_tokens_per_second": 12173.925
    },
    {
      "epoch": 0.5652437382170752,
      "grad_norm": 0.943740758025792,
      "learning_rate": 8.155689407450732e-06,
      "loss": 0.39592530727386477,
      "num_input_tokens_seen": 267648568,
      "step": 8395,
      "train_runtime": 21984.7736,
      "train_tokens_per_second": 12174.27
    },
    {
      "epoch": 0.5655803932130353,
      "grad_norm": 0.8043708769303783,
      "learning_rate": 8.153638029353615e-06,
      "loss": 0.3897716999053955,
      "num_input_tokens_seen": 267807064,
      "step": 8400,
      "train_runtime": 21998.2899,
      "train_tokens_per_second": 12173.995
    },
    {
      "epoch": 0.5659170482089955,
      "grad_norm": 0.8193110586615127,
      "learning_rate": 8.151585769351678e-06,
      "loss": 0.4180576324462891,
      "num_input_tokens_seen": 267967016,
      "step": 8405,
      "train_runtime": 22011.5405,
      "train_tokens_per_second": 12173.933
    },
    {
      "epoch": 0.5662537032049556,
      "grad_norm": 0.8163507788864877,
      "learning_rate": 8.149532628018833e-06,
      "loss": 0.3977893114089966,
      "num_input_tokens_seen": 268121784,
      "step": 8410,
      "train_runtime": 22024.0712,
      "train_tokens_per_second": 12174.034
    },
    {
      "epoch": 0.5665903582009157,
      "grad_norm": 0.8364728254577559,
      "learning_rate": 8.147478605929228e-06,
      "loss": 0.38081111907958987,
      "num_input_tokens_seen": 268282952,
      "step": 8415,
      "train_runtime": 22036.5487,
      "train_tokens_per_second": 12174.454
    },
    {
      "epoch": 0.5669270131968759,
      "grad_norm": 0.8680009193804383,
      "learning_rate": 8.145423703657269e-06,
      "loss": 0.35738024711608884,
      "num_input_tokens_seen": 268444200,
      "step": 8420,
      "train_runtime": 22049.6609,
      "train_tokens_per_second": 12174.527
    },
    {
      "epoch": 0.567263668192836,
      "grad_norm": 0.8056087281466204,
      "learning_rate": 8.143367921777601e-06,
      "loss": 0.4098641395568848,
      "num_input_tokens_seen": 268606664,
      "step": 8425,
      "train_runtime": 22062.7745,
      "train_tokens_per_second": 12174.655
    },
    {
      "epoch": 0.5676003231887962,
      "grad_norm": 0.8504922477420707,
      "learning_rate": 8.141311260865117e-06,
      "loss": 0.4474315643310547,
      "num_input_tokens_seen": 268767816,
      "step": 8430,
      "train_runtime": 22076.0916,
      "train_tokens_per_second": 12174.61
    },
    {
      "epoch": 0.5679369781847563,
      "grad_norm": 0.8444411391348389,
      "learning_rate": 8.139253721494956e-06,
      "loss": 0.4369081974029541,
      "num_input_tokens_seen": 268926840,
      "step": 8435,
      "train_runtime": 22089.8305,
      "train_tokens_per_second": 12174.237
    },
    {
      "epoch": 0.5682736331807164,
      "grad_norm": 0.7355145625285048,
      "learning_rate": 8.137195304242502e-06,
      "loss": 0.36676859855651855,
      "num_input_tokens_seen": 269088248,
      "step": 8440,
      "train_runtime": 22103.139,
      "train_tokens_per_second": 12174.21
    },
    {
      "epoch": 0.5686102881766766,
      "grad_norm": 0.690829657473422,
      "learning_rate": 8.135136009683387e-06,
      "loss": 0.3793917179107666,
      "num_input_tokens_seen": 269245512,
      "step": 8445,
      "train_runtime": 22116.3118,
      "train_tokens_per_second": 12174.069
    },
    {
      "epoch": 0.5689469431726367,
      "grad_norm": 0.7234183447556742,
      "learning_rate": 8.13307583839348e-06,
      "loss": 0.3861642360687256,
      "num_input_tokens_seen": 269408320,
      "step": 8450,
      "train_runtime": 22129.7623,
      "train_tokens_per_second": 12174.027
    },
    {
      "epoch": 0.5692835981685969,
      "grad_norm": 0.7854405385581302,
      "learning_rate": 8.131014790948908e-06,
      "loss": 0.4038506507873535,
      "num_input_tokens_seen": 269571032,
      "step": 8455,
      "train_runtime": 22143.1297,
      "train_tokens_per_second": 12174.026
    },
    {
      "epoch": 0.569620253164557,
      "grad_norm": 0.8161949815873275,
      "learning_rate": 8.128952867926035e-06,
      "loss": 0.4216309070587158,
      "num_input_tokens_seen": 269726688,
      "step": 8460,
      "train_runtime": 22156.3727,
      "train_tokens_per_second": 12173.775
    },
    {
      "epoch": 0.5699569081605171,
      "grad_norm": 0.7836594801256083,
      "learning_rate": 8.126890069901468e-06,
      "loss": 0.3847153425216675,
      "num_input_tokens_seen": 269881936,
      "step": 8465,
      "train_runtime": 22168.9263,
      "train_tokens_per_second": 12173.884
    },
    {
      "epoch": 0.5702935631564773,
      "grad_norm": 0.82059150815824,
      "learning_rate": 8.124826397452064e-06,
      "loss": 0.38450558185577394,
      "num_input_tokens_seen": 270045416,
      "step": 8470,
      "train_runtime": 22181.9378,
      "train_tokens_per_second": 12174.113
    },
    {
      "epoch": 0.5706302181524374,
      "grad_norm": 0.7167766567984177,
      "learning_rate": 8.122761851154923e-06,
      "loss": 0.38742387294769287,
      "num_input_tokens_seen": 270202728,
      "step": 8475,
      "train_runtime": 22195.1913,
      "train_tokens_per_second": 12173.931
    },
    {
      "epoch": 0.5709668731483976,
      "grad_norm": 0.9276943461747567,
      "learning_rate": 8.120696431587387e-06,
      "loss": 0.39433815479278567,
      "num_input_tokens_seen": 270363168,
      "step": 8480,
      "train_runtime": 22208.5952,
      "train_tokens_per_second": 12173.808
    },
    {
      "epoch": 0.5713035281443577,
      "grad_norm": 0.776133607676395,
      "learning_rate": 8.118630139327046e-06,
      "loss": 0.370235538482666,
      "num_input_tokens_seen": 270522464,
      "step": 8485,
      "train_runtime": 22221.7427,
      "train_tokens_per_second": 12173.774
    },
    {
      "epoch": 0.5716401831403178,
      "grad_norm": 0.7156344523220457,
      "learning_rate": 8.116562974951731e-06,
      "loss": 0.41245274543762206,
      "num_input_tokens_seen": 270684136,
      "step": 8490,
      "train_runtime": 22234.2535,
      "train_tokens_per_second": 12174.195
    },
    {
      "epoch": 0.571976838136278,
      "grad_norm": 0.7804048892656817,
      "learning_rate": 8.11449493903952e-06,
      "loss": 0.39719586372375487,
      "num_input_tokens_seen": 270847400,
      "step": 8495,
      "train_runtime": 22247.094,
      "train_tokens_per_second": 12174.507
    },
    {
      "epoch": 0.5723134931322381,
      "grad_norm": 0.7207332804711001,
      "learning_rate": 8.112426032168727e-06,
      "loss": 0.3941131591796875,
      "num_input_tokens_seen": 271005608,
      "step": 8500,
      "train_runtime": 22260.2787,
      "train_tokens_per_second": 12174.403
    },
    {
      "epoch": 0.5726501481281983,
      "grad_norm": 0.6967662545242425,
      "learning_rate": 8.11035625491792e-06,
      "loss": 0.35590691566467286,
      "num_input_tokens_seen": 271160672,
      "step": 8505,
      "train_runtime": 22273.9817,
      "train_tokens_per_second": 12173.875
    },
    {
      "epoch": 0.5729868031241584,
      "grad_norm": 0.8131319355102123,
      "learning_rate": 8.108285607865907e-06,
      "loss": 0.39660770893096925,
      "num_input_tokens_seen": 271319104,
      "step": 8510,
      "train_runtime": 22287.6069,
      "train_tokens_per_second": 12173.541
    },
    {
      "epoch": 0.5733234581201185,
      "grad_norm": 0.7575521275020969,
      "learning_rate": 8.106214091591731e-06,
      "loss": 0.3723886251449585,
      "num_input_tokens_seen": 271482944,
      "step": 8515,
      "train_runtime": 22300.0239,
      "train_tokens_per_second": 12174.11
    },
    {
      "epoch": 0.5736601131160787,
      "grad_norm": 0.7830147335499614,
      "learning_rate": 8.104141706674693e-06,
      "loss": 0.3499229192733765,
      "num_input_tokens_seen": 271644256,
      "step": 8520,
      "train_runtime": 22312.9529,
      "train_tokens_per_second": 12174.285
    },
    {
      "epoch": 0.5739967681120388,
      "grad_norm": 0.8800645904563533,
      "learning_rate": 8.102068453694324e-06,
      "loss": 0.40448813438415526,
      "num_input_tokens_seen": 271802664,
      "step": 8525,
      "train_runtime": 22325.9414,
      "train_tokens_per_second": 12174.298
    },
    {
      "epoch": 0.574333423107999,
      "grad_norm": 0.8439399507971489,
      "learning_rate": 8.099994333230404e-06,
      "loss": 0.4061802864074707,
      "num_input_tokens_seen": 271957560,
      "step": 8530,
      "train_runtime": 22338.7315,
      "train_tokens_per_second": 12174.262
    },
    {
      "epoch": 0.5746700781039591,
      "grad_norm": 0.9612036758130611,
      "learning_rate": 8.097919345862952e-06,
      "loss": 0.4010190963745117,
      "num_input_tokens_seen": 272111792,
      "step": 8535,
      "train_runtime": 22352.4003,
      "train_tokens_per_second": 12173.717
    },
    {
      "epoch": 0.5750067330999192,
      "grad_norm": 0.7399560763202631,
      "learning_rate": 8.095843492172235e-06,
      "loss": 0.3945467948913574,
      "num_input_tokens_seen": 272269928,
      "step": 8540,
      "train_runtime": 22365.0601,
      "train_tokens_per_second": 12173.897
    },
    {
      "epoch": 0.5753433880958794,
      "grad_norm": 0.9785278522948974,
      "learning_rate": 8.093766772738758e-06,
      "loss": 0.32863426208496094,
      "num_input_tokens_seen": 272427928,
      "step": 8545,
      "train_runtime": 22378.469,
      "train_tokens_per_second": 12173.662
    },
    {
      "epoch": 0.5756800430918395,
      "grad_norm": 0.8298269615242244,
      "learning_rate": 8.091689188143267e-06,
      "loss": 0.4153585910797119,
      "num_input_tokens_seen": 272585816,
      "step": 8550,
      "train_runtime": 22392.0085,
      "train_tokens_per_second": 12173.353
    },
    {
      "epoch": 0.5760166980877997,
      "grad_norm": 0.8011625906672519,
      "learning_rate": 8.089610738966754e-06,
      "loss": 0.39707293510437014,
      "num_input_tokens_seen": 272742840,
      "step": 8555,
      "train_runtime": 22404.855,
      "train_tokens_per_second": 12173.381
    },
    {
      "epoch": 0.5763533530837598,
      "grad_norm": 0.8746069596813106,
      "learning_rate": 8.08753142579045e-06,
      "loss": 0.4332891464233398,
      "num_input_tokens_seen": 272902336,
      "step": 8560,
      "train_runtime": 22418.6766,
      "train_tokens_per_second": 12172.99
    },
    {
      "epoch": 0.5766900080797199,
      "grad_norm": 0.7512638221808229,
      "learning_rate": 8.085451249195828e-06,
      "loss": 0.3762501239776611,
      "num_input_tokens_seen": 273065176,
      "step": 8565,
      "train_runtime": 22431.4598,
      "train_tokens_per_second": 12173.313
    },
    {
      "epoch": 0.5770266630756801,
      "grad_norm": 0.808258096006063,
      "learning_rate": 8.083370209764603e-06,
      "loss": 0.38850204944610595,
      "num_input_tokens_seen": 273223568,
      "step": 8570,
      "train_runtime": 22443.9159,
      "train_tokens_per_second": 12173.614
    },
    {
      "epoch": 0.5773633180716402,
      "grad_norm": 0.7890239528574572,
      "learning_rate": 8.081288308078731e-06,
      "loss": 0.39879584312438965,
      "num_input_tokens_seen": 273387000,
      "step": 8575,
      "train_runtime": 22456.9007,
      "train_tokens_per_second": 12173.853
    },
    {
      "epoch": 0.5776999730676003,
      "grad_norm": 0.7147821954311969,
      "learning_rate": 8.07920554472041e-06,
      "loss": 0.37086896896362304,
      "num_input_tokens_seen": 273550696,
      "step": 8580,
      "train_runtime": 22469.9443,
      "train_tokens_per_second": 12174.071
    },
    {
      "epoch": 0.5780366280635605,
      "grad_norm": 0.7777265304649422,
      "learning_rate": 8.077121920272075e-06,
      "loss": 0.3966056346893311,
      "num_input_tokens_seen": 273709456,
      "step": 8585,
      "train_runtime": 22482.5209,
      "train_tokens_per_second": 12174.322
    },
    {
      "epoch": 0.5783732830595206,
      "grad_norm": 0.721689173604904,
      "learning_rate": 8.07503743531641e-06,
      "loss": 0.3649152278900146,
      "num_input_tokens_seen": 273871032,
      "step": 8590,
      "train_runtime": 22496.9472,
      "train_tokens_per_second": 12173.698
    },
    {
      "epoch": 0.5787099380554808,
      "grad_norm": 0.6952559696965347,
      "learning_rate": 8.072952090436328e-06,
      "loss": 0.3712308406829834,
      "num_input_tokens_seen": 274030920,
      "step": 8595,
      "train_runtime": 22509.5918,
      "train_tokens_per_second": 12173.962
    },
    {
      "epoch": 0.5790465930514409,
      "grad_norm": 0.8158945573185497,
      "learning_rate": 8.070865886214996e-06,
      "loss": 0.3915894031524658,
      "num_input_tokens_seen": 274188688,
      "step": 8600,
      "train_runtime": 22522.6309,
      "train_tokens_per_second": 12173.919
    },
    {
      "epoch": 0.579383248047401,
      "grad_norm": 0.8636587925759287,
      "learning_rate": 8.06877882323581e-06,
      "loss": 0.3938615798950195,
      "num_input_tokens_seen": 274344832,
      "step": 8605,
      "train_runtime": 22535.0767,
      "train_tokens_per_second": 12174.125
    },
    {
      "epoch": 0.5797199030433612,
      "grad_norm": 0.8244518270421494,
      "learning_rate": 8.066690902082413e-06,
      "loss": 0.3691457748413086,
      "num_input_tokens_seen": 274505952,
      "step": 8610,
      "train_runtime": 22548.0573,
      "train_tokens_per_second": 12174.262
    },
    {
      "epoch": 0.5800565580393213,
      "grad_norm": 0.8122949601775046,
      "learning_rate": 8.064602123338683e-06,
      "loss": 0.4260343074798584,
      "num_input_tokens_seen": 274668272,
      "step": 8615,
      "train_runtime": 22560.662,
      "train_tokens_per_second": 12174.655
    },
    {
      "epoch": 0.5803932130352815,
      "grad_norm": 0.8227638337281615,
      "learning_rate": 8.06251248758874e-06,
      "loss": 0.38289215564727785,
      "num_input_tokens_seen": 274830080,
      "step": 8620,
      "train_runtime": 22573.6823,
      "train_tokens_per_second": 12174.801
    },
    {
      "epoch": 0.5807298680312416,
      "grad_norm": 0.775636416833392,
      "learning_rate": 8.060421995416949e-06,
      "loss": 0.41685895919799804,
      "num_input_tokens_seen": 274990488,
      "step": 8625,
      "train_runtime": 22587.0817,
      "train_tokens_per_second": 12174.68
    },
    {
      "epoch": 0.5810665230272017,
      "grad_norm": 0.8042505540363598,
      "learning_rate": 8.058330647407904e-06,
      "loss": 0.37619171142578123,
      "num_input_tokens_seen": 275147760,
      "step": 8630,
      "train_runtime": 22599.7347,
      "train_tokens_per_second": 12174.823
    },
    {
      "epoch": 0.5814031780231619,
      "grad_norm": 0.9291989077120335,
      "learning_rate": 8.056238444146445e-06,
      "loss": 0.3848073959350586,
      "num_input_tokens_seen": 275304784,
      "step": 8635,
      "train_runtime": 22612.9196,
      "train_tokens_per_second": 12174.668
    },
    {
      "epoch": 0.581739833019122,
      "grad_norm": 0.800751675448568,
      "learning_rate": 8.054145386217652e-06,
      "loss": 0.3641780138015747,
      "num_input_tokens_seen": 275467216,
      "step": 8640,
      "train_runtime": 22625.963,
      "train_tokens_per_second": 12174.828
    },
    {
      "epoch": 0.5820764880150822,
      "grad_norm": 0.7486964706026495,
      "learning_rate": 8.05205147420684e-06,
      "loss": 0.36989707946777345,
      "num_input_tokens_seen": 275626184,
      "step": 8645,
      "train_runtime": 22639.411,
      "train_tokens_per_second": 12174.618
    },
    {
      "epoch": 0.5824131430110423,
      "grad_norm": 0.8941003159988418,
      "learning_rate": 8.049956708699567e-06,
      "loss": 0.4291075706481934,
      "num_input_tokens_seen": 275784608,
      "step": 8650,
      "train_runtime": 22652.7497,
      "train_tokens_per_second": 12174.443
    },
    {
      "epoch": 0.5827497980070024,
      "grad_norm": 0.9339586414219014,
      "learning_rate": 8.047861090281623e-06,
      "loss": 0.39731745719909667,
      "num_input_tokens_seen": 275941720,
      "step": 8655,
      "train_runtime": 22665.576,
      "train_tokens_per_second": 12174.485
    },
    {
      "epoch": 0.5830864530029626,
      "grad_norm": 0.7109480431725202,
      "learning_rate": 8.045764619539047e-06,
      "loss": 0.37222120761871336,
      "num_input_tokens_seen": 276103832,
      "step": 8660,
      "train_runtime": 22678.5989,
      "train_tokens_per_second": 12174.642
    },
    {
      "epoch": 0.5834231079989227,
      "grad_norm": 0.7106880449575651,
      "learning_rate": 8.043667297058106e-06,
      "loss": 0.3681588888168335,
      "num_input_tokens_seen": 276260800,
      "step": 8665,
      "train_runtime": 22691.9031,
      "train_tokens_per_second": 12174.422
    },
    {
      "epoch": 0.5837597629948829,
      "grad_norm": 0.7149673404758417,
      "learning_rate": 8.04156912342531e-06,
      "loss": 0.3993093252182007,
      "num_input_tokens_seen": 276420256,
      "step": 8670,
      "train_runtime": 22705.3574,
      "train_tokens_per_second": 12174.231
    },
    {
      "epoch": 0.584096417990843,
      "grad_norm": 0.8914321219093229,
      "learning_rate": 8.03947009922741e-06,
      "loss": 0.3780981540679932,
      "num_input_tokens_seen": 276581744,
      "step": 8675,
      "train_runtime": 22718.4773,
      "train_tokens_per_second": 12174.308
    },
    {
      "epoch": 0.5844330729868031,
      "grad_norm": 0.8359863707888998,
      "learning_rate": 8.037370225051387e-06,
      "loss": 0.43466815948486326,
      "num_input_tokens_seen": 276739832,
      "step": 8680,
      "train_runtime": 22731.9708,
      "train_tokens_per_second": 12174.036
    },
    {
      "epoch": 0.5847697279827633,
      "grad_norm": 0.7874980324200367,
      "learning_rate": 8.035269501484467e-06,
      "loss": 0.37740159034729004,
      "num_input_tokens_seen": 276900544,
      "step": 8685,
      "train_runtime": 22745.3541,
      "train_tokens_per_second": 12173.939
    },
    {
      "epoch": 0.5851063829787234,
      "grad_norm": 0.8961160981913266,
      "learning_rate": 8.03316792911411e-06,
      "loss": 0.42656307220458983,
      "num_input_tokens_seen": 277056840,
      "step": 8690,
      "train_runtime": 22758.6526,
      "train_tokens_per_second": 12173.693
    },
    {
      "epoch": 0.5854430379746836,
      "grad_norm": 0.7127667643073454,
      "learning_rate": 8.031065508528011e-06,
      "loss": 0.3715827226638794,
      "num_input_tokens_seen": 277218544,
      "step": 8695,
      "train_runtime": 22771.7111,
      "train_tokens_per_second": 12173.813
    },
    {
      "epoch": 0.5857796929706437,
      "grad_norm": 0.7420706976281825,
      "learning_rate": 8.028962240314111e-06,
      "loss": 0.37357990741729735,
      "num_input_tokens_seen": 277380800,
      "step": 8700,
      "train_runtime": 22784.3049,
      "train_tokens_per_second": 12174.205
    },
    {
      "epoch": 0.5861163479666038,
      "grad_norm": 0.9183448741636746,
      "learning_rate": 8.026858125060577e-06,
      "loss": 0.4012912750244141,
      "num_input_tokens_seen": 277543976,
      "step": 8705,
      "train_runtime": 22797.2611,
      "train_tokens_per_second": 12174.444
    },
    {
      "epoch": 0.586453002962564,
      "grad_norm": 0.8040404792936039,
      "learning_rate": 8.024753163355821e-06,
      "loss": 0.43096342086791994,
      "num_input_tokens_seen": 277706592,
      "step": 8710,
      "train_runtime": 22810.6207,
      "train_tokens_per_second": 12174.443
    },
    {
      "epoch": 0.5867896579585241,
      "grad_norm": 0.7334247851298463,
      "learning_rate": 8.022647355788489e-06,
      "loss": 0.3863797903060913,
      "num_input_tokens_seen": 277869440,
      "step": 8715,
      "train_runtime": 22823.6524,
      "train_tokens_per_second": 12174.626
    },
    {
      "epoch": 0.5871263129544843,
      "grad_norm": 0.8634058779630797,
      "learning_rate": 8.02054070294746e-06,
      "loss": 0.4158777236938477,
      "num_input_tokens_seen": 278030304,
      "step": 8720,
      "train_runtime": 22836.8703,
      "train_tokens_per_second": 12174.624
    },
    {
      "epoch": 0.5874629679504444,
      "grad_norm": 0.988254186329373,
      "learning_rate": 8.018433205421859e-06,
      "loss": 0.37461717128753663,
      "num_input_tokens_seen": 278191104,
      "step": 8725,
      "train_runtime": 22849.7108,
      "train_tokens_per_second": 12174.819
    },
    {
      "epoch": 0.5877996229464045,
      "grad_norm": 0.8286663324292157,
      "learning_rate": 8.016324863801036e-06,
      "loss": 0.39465899467468263,
      "num_input_tokens_seen": 278352680,
      "step": 8730,
      "train_runtime": 22862.7181,
      "train_tokens_per_second": 12174.96
    },
    {
      "epoch": 0.5881362779423647,
      "grad_norm": 0.7949949062878114,
      "learning_rate": 8.014215678674581e-06,
      "loss": 0.41408047676086424,
      "num_input_tokens_seen": 278516472,
      "step": 8735,
      "train_runtime": 22875.7993,
      "train_tokens_per_second": 12175.158
    },
    {
      "epoch": 0.5884729329383248,
      "grad_norm": 0.6753628598508605,
      "learning_rate": 8.012105650632325e-06,
      "loss": 0.3678672075271606,
      "num_input_tokens_seen": 278675872,
      "step": 8740,
      "train_runtime": 22889.4729,
      "train_tokens_per_second": 12174.849
    },
    {
      "epoch": 0.588809587934285,
      "grad_norm": 0.736473655899873,
      "learning_rate": 8.009994780264328e-06,
      "loss": 0.41208972930908205,
      "num_input_tokens_seen": 278826360,
      "step": 8745,
      "train_runtime": 22902.7511,
      "train_tokens_per_second": 12174.361
    },
    {
      "epoch": 0.5891462429302451,
      "grad_norm": 0.8168118745280201,
      "learning_rate": 8.00788306816089e-06,
      "loss": 0.3884446382522583,
      "num_input_tokens_seen": 278984896,
      "step": 8750,
      "train_runtime": 22915.6377,
      "train_tokens_per_second": 12174.433
    },
    {
      "epoch": 0.5894828979262052,
      "grad_norm": 0.8489483429047614,
      "learning_rate": 8.005770514912543e-06,
      "loss": 0.39844307899475095,
      "num_input_tokens_seen": 279148736,
      "step": 8755,
      "train_runtime": 22928.0541,
      "train_tokens_per_second": 12174.986
    },
    {
      "epoch": 0.5898195529221654,
      "grad_norm": 0.7941232601197789,
      "learning_rate": 8.003657121110055e-06,
      "loss": 0.38739628791809083,
      "num_input_tokens_seen": 279312576,
      "step": 8760,
      "train_runtime": 22940.4675,
      "train_tokens_per_second": 12175.54
    },
    {
      "epoch": 0.5901562079181255,
      "grad_norm": 0.8409881816856485,
      "learning_rate": 8.001542887344432e-06,
      "loss": 0.35993902683258056,
      "num_input_tokens_seen": 279470760,
      "step": 8765,
      "train_runtime": 22953.4012,
      "train_tokens_per_second": 12175.571
    },
    {
      "epoch": 0.5904928629140856,
      "grad_norm": 0.7491800421278206,
      "learning_rate": 7.999427814206911e-06,
      "loss": 0.3962878942489624,
      "num_input_tokens_seen": 279629592,
      "step": 8770,
      "train_runtime": 22966.3941,
      "train_tokens_per_second": 12175.598
    },
    {
      "epoch": 0.5908295179100458,
      "grad_norm": 0.8087658568362288,
      "learning_rate": 7.997311902288968e-06,
      "loss": 0.34962759017944334,
      "num_input_tokens_seen": 279790416,
      "step": 8775,
      "train_runtime": 22979.2247,
      "train_tokens_per_second": 12175.799
    },
    {
      "epoch": 0.5911661729060059,
      "grad_norm": 0.7103341446273141,
      "learning_rate": 7.995195152182307e-06,
      "loss": 0.347735857963562,
      "num_input_tokens_seen": 279948192,
      "step": 8780,
      "train_runtime": 22993.027,
      "train_tokens_per_second": 12175.352
    },
    {
      "epoch": 0.5915028279019661,
      "grad_norm": 0.7451778041182738,
      "learning_rate": 7.993077564478875e-06,
      "loss": 0.3549297332763672,
      "num_input_tokens_seen": 280110600,
      "step": 8785,
      "train_runtime": 23005.6833,
      "train_tokens_per_second": 12175.713
    },
    {
      "epoch": 0.5918394828979262,
      "grad_norm": 0.8300167940942154,
      "learning_rate": 7.990959139770844e-06,
      "loss": 0.3999783039093018,
      "num_input_tokens_seen": 280273760,
      "step": 8790,
      "train_runtime": 23018.5371,
      "train_tokens_per_second": 12176.002
    },
    {
      "epoch": 0.5921761378938863,
      "grad_norm": 0.8143973170883196,
      "learning_rate": 7.988839878650628e-06,
      "loss": 0.3855675935745239,
      "num_input_tokens_seen": 280436736,
      "step": 8795,
      "train_runtime": 23031.3402,
      "train_tokens_per_second": 12176.31
    },
    {
      "epoch": 0.5925127928898465,
      "grad_norm": 0.8448232082458818,
      "learning_rate": 7.986719781710872e-06,
      "loss": 0.40084547996520997,
      "num_input_tokens_seen": 280599328,
      "step": 8800,
      "train_runtime": 23044.7132,
      "train_tokens_per_second": 12176.299
    },
    {
      "epoch": 0.5928494478858066,
      "grad_norm": 0.8602322554365683,
      "learning_rate": 7.984598849544452e-06,
      "loss": 0.39586241245269777,
      "num_input_tokens_seen": 280759112,
      "step": 8805,
      "train_runtime": 23057.9959,
      "train_tokens_per_second": 12176.215
    },
    {
      "epoch": 0.5931861028817668,
      "grad_norm": 0.7899005891649498,
      "learning_rate": 7.98247708274448e-06,
      "loss": 0.39266557693481446,
      "num_input_tokens_seen": 280918480,
      "step": 8810,
      "train_runtime": 23071.6304,
      "train_tokens_per_second": 12175.927
    },
    {
      "epoch": 0.5935227578777269,
      "grad_norm": 0.7050545158480707,
      "learning_rate": 7.980354481904303e-06,
      "loss": 0.41337099075317385,
      "num_input_tokens_seen": 281079096,
      "step": 8815,
      "train_runtime": 23084.6181,
      "train_tokens_per_second": 12176.034
    },
    {
      "epoch": 0.593859412873687,
      "grad_norm": 0.7923408727023673,
      "learning_rate": 7.978231047617498e-06,
      "loss": 0.3995427131652832,
      "num_input_tokens_seen": 281237712,
      "step": 8820,
      "train_runtime": 23097.7902,
      "train_tokens_per_second": 12175.957
    },
    {
      "epoch": 0.5941960678696472,
      "grad_norm": 0.7871674414211024,
      "learning_rate": 7.976106780477878e-06,
      "loss": 0.38719174861907957,
      "num_input_tokens_seen": 281398936,
      "step": 8825,
      "train_runtime": 23110.8706,
      "train_tokens_per_second": 12176.042
    },
    {
      "epoch": 0.5945327228656073,
      "grad_norm": 0.8253029439980832,
      "learning_rate": 7.973981681079487e-06,
      "loss": 0.3673261642456055,
      "num_input_tokens_seen": 281558368,
      "step": 8830,
      "train_runtime": 23124.2971,
      "train_tokens_per_second": 12175.867
    },
    {
      "epoch": 0.5948693778615675,
      "grad_norm": 0.804235107789454,
      "learning_rate": 7.971855750016603e-06,
      "loss": 0.3686270475387573,
      "num_input_tokens_seen": 281718672,
      "step": 8835,
      "train_runtime": 23137.6598,
      "train_tokens_per_second": 12175.763
    },
    {
      "epoch": 0.5952060328575276,
      "grad_norm": 0.8452937818474988,
      "learning_rate": 7.969728987883734e-06,
      "loss": 0.4011039733886719,
      "num_input_tokens_seen": 281877344,
      "step": 8840,
      "train_runtime": 23150.2095,
      "train_tokens_per_second": 12176.017
    },
    {
      "epoch": 0.5955426878534877,
      "grad_norm": 0.8362119491621816,
      "learning_rate": 7.967601395275624e-06,
      "loss": 0.3875253438949585,
      "num_input_tokens_seen": 282037776,
      "step": 8845,
      "train_runtime": 23162.6057,
      "train_tokens_per_second": 12176.427
    },
    {
      "epoch": 0.5958793428494479,
      "grad_norm": 0.6566904327209583,
      "learning_rate": 7.965472972787247e-06,
      "loss": 0.3690237522125244,
      "num_input_tokens_seen": 282198616,
      "step": 8850,
      "train_runtime": 23175.6661,
      "train_tokens_per_second": 12176.505
    },
    {
      "epoch": 0.596215997845408,
      "grad_norm": 0.7347838525342171,
      "learning_rate": 7.963343721013808e-06,
      "loss": 0.3724492073059082,
      "num_input_tokens_seen": 282360136,
      "step": 8855,
      "train_runtime": 23188.1491,
      "train_tokens_per_second": 12176.916
    },
    {
      "epoch": 0.5965526528413682,
      "grad_norm": 0.7364155953264305,
      "learning_rate": 7.961213640550748e-06,
      "loss": 0.3840902328491211,
      "num_input_tokens_seen": 282522184,
      "step": 8860,
      "train_runtime": 23200.595,
      "train_tokens_per_second": 12177.368
    },
    {
      "epoch": 0.5968893078373283,
      "grad_norm": 0.7080907547368479,
      "learning_rate": 7.959082731993736e-06,
      "loss": 0.35625221729278567,
      "num_input_tokens_seen": 282682128,
      "step": 8865,
      "train_runtime": 23213.6346,
      "train_tokens_per_second": 12177.418
    },
    {
      "epoch": 0.5972259628332884,
      "grad_norm": 0.8238642332231467,
      "learning_rate": 7.956950995938673e-06,
      "loss": 0.3702124834060669,
      "num_input_tokens_seen": 282839920,
      "step": 8870,
      "train_runtime": 23226.8434,
      "train_tokens_per_second": 12177.286
    },
    {
      "epoch": 0.5975626178292486,
      "grad_norm": 0.7849987649911606,
      "learning_rate": 7.954818432981694e-06,
      "loss": 0.4011682987213135,
      "num_input_tokens_seen": 282997704,
      "step": 8875,
      "train_runtime": 23240.0692,
      "train_tokens_per_second": 12177.146
    },
    {
      "epoch": 0.5978992728252087,
      "grad_norm": 0.7736111628596385,
      "learning_rate": 7.952685043719162e-06,
      "loss": 0.36965084075927734,
      "num_input_tokens_seen": 283157928,
      "step": 8880,
      "train_runtime": 23253.0856,
      "train_tokens_per_second": 12177.22
    },
    {
      "epoch": 0.5982359278211689,
      "grad_norm": 0.8666893211101858,
      "learning_rate": 7.950550828747672e-06,
      "loss": 0.3772313117980957,
      "num_input_tokens_seen": 283320040,
      "step": 8885,
      "train_runtime": 23265.5242,
      "train_tokens_per_second": 12177.677
    },
    {
      "epoch": 0.598572582817129,
      "grad_norm": 0.7647879799162872,
      "learning_rate": 7.948415788664052e-06,
      "loss": 0.4188659191131592,
      "num_input_tokens_seen": 283478352,
      "step": 8890,
      "train_runtime": 23278.6013,
      "train_tokens_per_second": 12177.637
    },
    {
      "epoch": 0.5989092378130891,
      "grad_norm": 0.8678610623407963,
      "learning_rate": 7.946279924065359e-06,
      "loss": 0.3735698699951172,
      "num_input_tokens_seen": 283635440,
      "step": 8895,
      "train_runtime": 23291.851,
      "train_tokens_per_second": 12177.454
    },
    {
      "epoch": 0.5992458928090493,
      "grad_norm": 0.8213500777997411,
      "learning_rate": 7.94414323554888e-06,
      "loss": 0.4037994384765625,
      "num_input_tokens_seen": 283798128,
      "step": 8900,
      "train_runtime": 23304.8949,
      "train_tokens_per_second": 12177.619
    },
    {
      "epoch": 0.5995825478050094,
      "grad_norm": 0.7857615760198975,
      "learning_rate": 7.942005723712135e-06,
      "loss": 0.3853250503540039,
      "num_input_tokens_seen": 283960768,
      "step": 8905,
      "train_runtime": 23317.5692,
      "train_tokens_per_second": 12177.975
    },
    {
      "epoch": 0.5999192028009696,
      "grad_norm": 0.7497548305321684,
      "learning_rate": 7.939867389152868e-06,
      "loss": 0.3475773334503174,
      "num_input_tokens_seen": 284119888,
      "step": 8910,
      "train_runtime": 23330.3746,
      "train_tokens_per_second": 12178.111
    },
    {
      "epoch": 0.6002558577969297,
      "grad_norm": 0.7819093336243684,
      "learning_rate": 7.937728232469062e-06,
      "loss": 0.4126457214355469,
      "num_input_tokens_seen": 284279320,
      "step": 8915,
      "train_runtime": 23343.5346,
      "train_tokens_per_second": 12178.075
    },
    {
      "epoch": 0.6005925127928898,
      "grad_norm": 0.7601000914299808,
      "learning_rate": 7.935588254258921e-06,
      "loss": 0.4114988803863525,
      "num_input_tokens_seen": 284435456,
      "step": 8920,
      "train_runtime": 23357.2799,
      "train_tokens_per_second": 12177.593
    },
    {
      "epoch": 0.60092916778885,
      "grad_norm": 0.7900040179056541,
      "learning_rate": 7.933447455120889e-06,
      "loss": 0.37104811668396,
      "num_input_tokens_seen": 284594008,
      "step": 8925,
      "train_runtime": 23370.0561,
      "train_tokens_per_second": 12177.72
    },
    {
      "epoch": 0.6012658227848101,
      "grad_norm": 0.8022451977677605,
      "learning_rate": 7.931305835653629e-06,
      "loss": 0.3637326955795288,
      "num_input_tokens_seen": 284755288,
      "step": 8930,
      "train_runtime": 23382.4795,
      "train_tokens_per_second": 12178.148
    },
    {
      "epoch": 0.6016024777807703,
      "grad_norm": 0.7375049038345973,
      "learning_rate": 7.929163396456038e-06,
      "loss": 0.40918521881103515,
      "num_input_tokens_seen": 284913176,
      "step": 8935,
      "train_runtime": 23395.3012,
      "train_tokens_per_second": 12178.222
    },
    {
      "epoch": 0.6019391327767304,
      "grad_norm": 0.8072144346128841,
      "learning_rate": 7.927020138127245e-06,
      "loss": 0.38272120952606203,
      "num_input_tokens_seen": 285073000,
      "step": 8940,
      "train_runtime": 23407.9168,
      "train_tokens_per_second": 12178.487
    },
    {
      "epoch": 0.6022757877726905,
      "grad_norm": 0.7882028314213002,
      "learning_rate": 7.924876061266603e-06,
      "loss": 0.3714626312255859,
      "num_input_tokens_seen": 285234480,
      "step": 8945,
      "train_runtime": 23420.4564,
      "train_tokens_per_second": 12178.861
    },
    {
      "epoch": 0.6026124427686507,
      "grad_norm": 0.8754163575029166,
      "learning_rate": 7.922731166473698e-06,
      "loss": 0.40097503662109374,
      "num_input_tokens_seen": 285394888,
      "step": 8950,
      "train_runtime": 23433.9563,
      "train_tokens_per_second": 12178.69
    },
    {
      "epoch": 0.6029490977646108,
      "grad_norm": 0.8508960258467497,
      "learning_rate": 7.920585454348341e-06,
      "loss": 0.37189421653747556,
      "num_input_tokens_seen": 285555016,
      "step": 8955,
      "train_runtime": 23447.8168,
      "train_tokens_per_second": 12178.32
    },
    {
      "epoch": 0.603285752760571,
      "grad_norm": 0.8362788484676088,
      "learning_rate": 7.918438925490578e-06,
      "loss": 0.35410442352294924,
      "num_input_tokens_seen": 285717328,
      "step": 8960,
      "train_runtime": 23461.1444,
      "train_tokens_per_second": 12178.32
    },
    {
      "epoch": 0.6036224077565311,
      "grad_norm": 0.8386698450276354,
      "learning_rate": 7.916291580500672e-06,
      "loss": 0.37752423286437986,
      "num_input_tokens_seen": 285876280,
      "step": 8965,
      "train_runtime": 23473.7226,
      "train_tokens_per_second": 12178.566
    },
    {
      "epoch": 0.6039590627524912,
      "grad_norm": 0.8045346955682096,
      "learning_rate": 7.914143419979127e-06,
      "loss": 0.35655837059020995,
      "num_input_tokens_seen": 286038840,
      "step": 8970,
      "train_runtime": 23486.717,
      "train_tokens_per_second": 12178.749
    },
    {
      "epoch": 0.6042957177484514,
      "grad_norm": 0.7017855589952327,
      "learning_rate": 7.911994444526667e-06,
      "loss": 0.34722366333007815,
      "num_input_tokens_seen": 286199160,
      "step": 8975,
      "train_runtime": 23500.129,
      "train_tokens_per_second": 12178.621
    },
    {
      "epoch": 0.6046323727444115,
      "grad_norm": 0.6849381736555221,
      "learning_rate": 7.909844654744246e-06,
      "loss": 0.3659569263458252,
      "num_input_tokens_seen": 286360816,
      "step": 8980,
      "train_runtime": 23513.3141,
      "train_tokens_per_second": 12178.667
    },
    {
      "epoch": 0.6049690277403716,
      "grad_norm": 0.8095170848709679,
      "learning_rate": 7.907694051233044e-06,
      "loss": 0.4197356700897217,
      "num_input_tokens_seen": 286519024,
      "step": 8985,
      "train_runtime": 23526.823,
      "train_tokens_per_second": 12178.398
    },
    {
      "epoch": 0.6053056827363318,
      "grad_norm": 0.6597064582321633,
      "learning_rate": 7.905542634594476e-06,
      "loss": 0.3881105899810791,
      "num_input_tokens_seen": 286677648,
      "step": 8990,
      "train_runtime": 23539.3624,
      "train_tokens_per_second": 12178.65
    },
    {
      "epoch": 0.6056423377322919,
      "grad_norm": 0.7848981795311314,
      "learning_rate": 7.903390405430174e-06,
      "loss": 0.3517893314361572,
      "num_input_tokens_seen": 286837536,
      "step": 8995,
      "train_runtime": 23552.9672,
      "train_tokens_per_second": 12178.403
    },
    {
      "epoch": 0.6059789927282521,
      "grad_norm": 0.8482209975508648,
      "learning_rate": 7.901237364342003e-06,
      "loss": 0.36460878849029543,
      "num_input_tokens_seen": 286988872,
      "step": 9000,
      "train_runtime": 23565.9246,
      "train_tokens_per_second": 12178.129
    },
    {
      "epoch": 0.6063156477242122,
      "grad_norm": 0.7955616993774309,
      "learning_rate": 7.899083511932053e-06,
      "loss": 0.3517482042312622,
      "num_input_tokens_seen": 287147792,
      "step": 9005,
      "train_runtime": 23578.9165,
      "train_tokens_per_second": 12178.159
    },
    {
      "epoch": 0.6066523027201723,
      "grad_norm": 0.7697559834799271,
      "learning_rate": 7.896928848802645e-06,
      "loss": 0.3766965389251709,
      "num_input_tokens_seen": 287311632,
      "step": 9010,
      "train_runtime": 23591.3251,
      "train_tokens_per_second": 12178.698
    },
    {
      "epoch": 0.6069889577161325,
      "grad_norm": 0.7199901453444618,
      "learning_rate": 7.894773375556322e-06,
      "loss": 0.3919867753982544,
      "num_input_tokens_seen": 287465672,
      "step": 9015,
      "train_runtime": 23604.5386,
      "train_tokens_per_second": 12178.407
    },
    {
      "epoch": 0.6073256127120926,
      "grad_norm": 0.8101691652911277,
      "learning_rate": 7.892617092795855e-06,
      "loss": 0.42398991584777834,
      "num_input_tokens_seen": 287623632,
      "step": 9020,
      "train_runtime": 23617.7205,
      "train_tokens_per_second": 12178.298
    },
    {
      "epoch": 0.6076622677080528,
      "grad_norm": 0.7689033259377673,
      "learning_rate": 7.890460001124242e-06,
      "loss": 0.3903092861175537,
      "num_input_tokens_seen": 287781584,
      "step": 9025,
      "train_runtime": 23631.0546,
      "train_tokens_per_second": 12178.11
    },
    {
      "epoch": 0.6079989227040129,
      "grad_norm": 0.8003076154440676,
      "learning_rate": 7.888302101144707e-06,
      "loss": 0.39416003227233887,
      "num_input_tokens_seen": 287939840,
      "step": 9030,
      "train_runtime": 23643.7573,
      "train_tokens_per_second": 12178.261
    },
    {
      "epoch": 0.608335577699973,
      "grad_norm": 0.7865560672753751,
      "learning_rate": 7.886143393460699e-06,
      "loss": 0.41128716468811033,
      "num_input_tokens_seen": 288101744,
      "step": 9035,
      "train_runtime": 23656.3058,
      "train_tokens_per_second": 12178.645
    },
    {
      "epoch": 0.6086722326959332,
      "grad_norm": 0.7875181115627478,
      "learning_rate": 7.883983878675895e-06,
      "loss": 0.37975549697875977,
      "num_input_tokens_seen": 288265584,
      "step": 9040,
      "train_runtime": 23668.7304,
      "train_tokens_per_second": 12179.174
    },
    {
      "epoch": 0.6090088876918933,
      "grad_norm": 0.7582833487953359,
      "learning_rate": 7.881823557394194e-06,
      "loss": 0.39029483795166015,
      "num_input_tokens_seen": 288429328,
      "step": 9045,
      "train_runtime": 23681.8035,
      "train_tokens_per_second": 12179.365
    },
    {
      "epoch": 0.6093455426878535,
      "grad_norm": 0.749531808486419,
      "learning_rate": 7.879662430219728e-06,
      "loss": 0.36229681968688965,
      "num_input_tokens_seen": 288584096,
      "step": 9050,
      "train_runtime": 23695.2908,
      "train_tokens_per_second": 12178.964
    },
    {
      "epoch": 0.6096821976838136,
      "grad_norm": 0.8493674933295446,
      "learning_rate": 7.877500497756843e-06,
      "loss": 0.4147617340087891,
      "num_input_tokens_seen": 288745008,
      "step": 9055,
      "train_runtime": 23708.329,
      "train_tokens_per_second": 12179.054
    },
    {
      "epoch": 0.6100188526797737,
      "grad_norm": 0.9887661985930278,
      "learning_rate": 7.87533776061012e-06,
      "loss": 0.38833179473876955,
      "num_input_tokens_seen": 288904576,
      "step": 9060,
      "train_runtime": 23721.5335,
      "train_tokens_per_second": 12179.001
    },
    {
      "epoch": 0.6103555076757339,
      "grad_norm": 0.7686606359638662,
      "learning_rate": 7.873174219384362e-06,
      "loss": 0.413944149017334,
      "num_input_tokens_seen": 289065528,
      "step": 9065,
      "train_runtime": 23734.2953,
      "train_tokens_per_second": 12179.234
    },
    {
      "epoch": 0.610692162671694,
      "grad_norm": 0.8110899315310492,
      "learning_rate": 7.871009874684595e-06,
      "loss": 0.4094103813171387,
      "num_input_tokens_seen": 289226656,
      "step": 9070,
      "train_runtime": 23746.8077,
      "train_tokens_per_second": 12179.602
    },
    {
      "epoch": 0.6110288176676542,
      "grad_norm": 0.9484941214828831,
      "learning_rate": 7.868844727116072e-06,
      "loss": 0.40843710899353025,
      "num_input_tokens_seen": 289384936,
      "step": 9075,
      "train_runtime": 23759.9151,
      "train_tokens_per_second": 12179.544
    },
    {
      "epoch": 0.6113654726636143,
      "grad_norm": 0.7242308894248829,
      "learning_rate": 7.866678777284267e-06,
      "loss": 0.3753045558929443,
      "num_input_tokens_seen": 289545840,
      "step": 9080,
      "train_runtime": 23773.3948,
      "train_tokens_per_second": 12179.407
    },
    {
      "epoch": 0.6117021276595744,
      "grad_norm": 0.7845727151169357,
      "learning_rate": 7.864512025794886e-06,
      "loss": 0.38861889839172364,
      "num_input_tokens_seen": 289706912,
      "step": 9085,
      "train_runtime": 23786.1323,
      "train_tokens_per_second": 12179.656
    },
    {
      "epoch": 0.6120387826555346,
      "grad_norm": 0.9692023666159564,
      "learning_rate": 7.86234447325385e-06,
      "loss": 0.35805387496948243,
      "num_input_tokens_seen": 289865864,
      "step": 9090,
      "train_runtime": 23798.7783,
      "train_tokens_per_second": 12179.863
    },
    {
      "epoch": 0.6123754376514947,
      "grad_norm": 0.9122754061754776,
      "learning_rate": 7.860176120267309e-06,
      "loss": 0.42113037109375,
      "num_input_tokens_seen": 290027224,
      "step": 9095,
      "train_runtime": 23811.3137,
      "train_tokens_per_second": 12180.228
    },
    {
      "epoch": 0.6127120926474549,
      "grad_norm": 0.7268134512640009,
      "learning_rate": 7.858006967441637e-06,
      "loss": 0.3743340492248535,
      "num_input_tokens_seen": 290184216,
      "step": 9100,
      "train_runtime": 23824.8737,
      "train_tokens_per_second": 12179.885
    },
    {
      "epoch": 0.613048747643415,
      "grad_norm": 0.8719108636652475,
      "learning_rate": 7.85583701538343e-06,
      "loss": 0.4194091796875,
      "num_input_tokens_seen": 290346312,
      "step": 9105,
      "train_runtime": 23837.4548,
      "train_tokens_per_second": 12180.256
    },
    {
      "epoch": 0.6133854026393751,
      "grad_norm": 0.8611505214761725,
      "learning_rate": 7.853666264699506e-06,
      "loss": 0.3680274963378906,
      "num_input_tokens_seen": 290505432,
      "step": 9110,
      "train_runtime": 23850.9427,
      "train_tokens_per_second": 12180.04
    },
    {
      "epoch": 0.6137220576353353,
      "grad_norm": 0.7512745837140431,
      "learning_rate": 7.851494715996914e-06,
      "loss": 0.4184731006622314,
      "num_input_tokens_seen": 290664120,
      "step": 9115,
      "train_runtime": 23863.6889,
      "train_tokens_per_second": 12180.184
    },
    {
      "epoch": 0.6140587126312954,
      "grad_norm": 0.8121069178417053,
      "learning_rate": 7.849322369882915e-06,
      "loss": 0.36065824031829835,
      "num_input_tokens_seen": 290825632,
      "step": 9120,
      "train_runtime": 23876.1368,
      "train_tokens_per_second": 12180.598
    },
    {
      "epoch": 0.6143953676272556,
      "grad_norm": 0.7577998032576303,
      "learning_rate": 7.847149226965003e-06,
      "loss": 0.3561420440673828,
      "num_input_tokens_seen": 290981504,
      "step": 9125,
      "train_runtime": 23888.7392,
      "train_tokens_per_second": 12180.697
    },
    {
      "epoch": 0.6147320226232157,
      "grad_norm": 0.9172271248760583,
      "learning_rate": 7.844975287850886e-06,
      "loss": 0.3470319271087646,
      "num_input_tokens_seen": 291142904,
      "step": 9130,
      "train_runtime": 23902.4085,
      "train_tokens_per_second": 12180.484
    },
    {
      "epoch": 0.6150686776191758,
      "grad_norm": 0.719020617041841,
      "learning_rate": 7.842800553148506e-06,
      "loss": 0.4091758728027344,
      "num_input_tokens_seen": 291300552,
      "step": 9135,
      "train_runtime": 23916.2763,
      "train_tokens_per_second": 12180.013
    },
    {
      "epoch": 0.615405332615136,
      "grad_norm": 0.7404291565440465,
      "learning_rate": 7.840625023466014e-06,
      "loss": 0.37669970989227297,
      "num_input_tokens_seen": 291464384,
      "step": 9140,
      "train_runtime": 23929.4137,
      "train_tokens_per_second": 12180.172
    },
    {
      "epoch": 0.6157419876110961,
      "grad_norm": 0.8431832118210815,
      "learning_rate": 7.838448699411792e-06,
      "loss": 0.36565518379211426,
      "num_input_tokens_seen": 291625040,
      "step": 9145,
      "train_runtime": 23942.5143,
      "train_tokens_per_second": 12180.218
    },
    {
      "epoch": 0.6160786426070562,
      "grad_norm": 0.7662486188210009,
      "learning_rate": 7.836271581594442e-06,
      "loss": 0.4152923583984375,
      "num_input_tokens_seen": 291788592,
      "step": 9150,
      "train_runtime": 23955.5591,
      "train_tokens_per_second": 12180.413
    },
    {
      "epoch": 0.6164152976030164,
      "grad_norm": 0.8101167711617063,
      "learning_rate": 7.834093670622789e-06,
      "loss": 0.40111422538757324,
      "num_input_tokens_seen": 291945824,
      "step": 9155,
      "train_runtime": 23968.7824,
      "train_tokens_per_second": 12180.253
    },
    {
      "epoch": 0.6167519525989765,
      "grad_norm": 0.7762839315864667,
      "learning_rate": 7.831914967105878e-06,
      "loss": 0.4347250938415527,
      "num_input_tokens_seen": 292106656,
      "step": 9160,
      "train_runtime": 23981.2091,
      "train_tokens_per_second": 12180.648
    },
    {
      "epoch": 0.6170886075949367,
      "grad_norm": 0.7125706272776811,
      "learning_rate": 7.829735471652978e-06,
      "loss": 0.35070099830627444,
      "num_input_tokens_seen": 292265792,
      "step": 9165,
      "train_runtime": 23994.3153,
      "train_tokens_per_second": 12180.626
    },
    {
      "epoch": 0.6174252625908968,
      "grad_norm": 0.9294847354353997,
      "learning_rate": 7.827555184873575e-06,
      "loss": 0.4055934906005859,
      "num_input_tokens_seen": 292425880,
      "step": 9170,
      "train_runtime": 24007.2315,
      "train_tokens_per_second": 12180.741
    },
    {
      "epoch": 0.617761917586857,
      "grad_norm": 0.7492561996379158,
      "learning_rate": 7.825374107377381e-06,
      "loss": 0.3657411813735962,
      "num_input_tokens_seen": 292586440,
      "step": 9175,
      "train_runtime": 24019.8859,
      "train_tokens_per_second": 12181.009
    },
    {
      "epoch": 0.6180985725828171,
      "grad_norm": 0.8712370340218983,
      "learning_rate": 7.823192239774327e-06,
      "loss": 0.3919386863708496,
      "num_input_tokens_seen": 292745960,
      "step": 9180,
      "train_runtime": 24033.4372,
      "train_tokens_per_second": 12180.778
    },
    {
      "epoch": 0.6184352275787772,
      "grad_norm": 0.7540690727529309,
      "learning_rate": 7.821009582674567e-06,
      "loss": 0.39654107093811036,
      "num_input_tokens_seen": 292904952,
      "step": 9185,
      "train_runtime": 24046.5699,
      "train_tokens_per_second": 12180.737
    },
    {
      "epoch": 0.6187718825747374,
      "grad_norm": 0.7841719577989008,
      "learning_rate": 7.818826136688473e-06,
      "loss": 0.41560792922973633,
      "num_input_tokens_seen": 293067016,
      "step": 9190,
      "train_runtime": 24059.0237,
      "train_tokens_per_second": 12181.168
    },
    {
      "epoch": 0.6191085375706975,
      "grad_norm": 0.8190941515639358,
      "learning_rate": 7.816641902426638e-06,
      "loss": 0.40740361213684084,
      "num_input_tokens_seen": 293226608,
      "step": 9195,
      "train_runtime": 24071.4867,
      "train_tokens_per_second": 12181.491
    },
    {
      "epoch": 0.6194451925666576,
      "grad_norm": 0.8743595905110048,
      "learning_rate": 7.814456880499877e-06,
      "loss": 0.3890812873840332,
      "num_input_tokens_seen": 293384608,
      "step": 9200,
      "train_runtime": 24085.1341,
      "train_tokens_per_second": 12181.149
    },
    {
      "epoch": 0.6197818475626178,
      "grad_norm": 0.928077575479999,
      "learning_rate": 7.812271071519222e-06,
      "loss": 0.4053468704223633,
      "num_input_tokens_seen": 293541656,
      "step": 9205,
      "train_runtime": 24098.7869,
      "train_tokens_per_second": 12180.765
    },
    {
      "epoch": 0.6201185025585779,
      "grad_norm": 0.73520523977525,
      "learning_rate": 7.810084476095932e-06,
      "loss": 0.35848202705383303,
      "num_input_tokens_seen": 293702368,
      "step": 9210,
      "train_runtime": 24112.1333,
      "train_tokens_per_second": 12180.688
    },
    {
      "epoch": 0.6204551575545381,
      "grad_norm": 0.780470601554464,
      "learning_rate": 7.807897094841476e-06,
      "loss": 0.39025588035583497,
      "num_input_tokens_seen": 293859200,
      "step": 9215,
      "train_runtime": 24125.2113,
      "train_tokens_per_second": 12180.586
    },
    {
      "epoch": 0.6207918125504982,
      "grad_norm": 0.7618089512461381,
      "learning_rate": 7.805708928367555e-06,
      "loss": 0.37273664474487306,
      "num_input_tokens_seen": 294013792,
      "step": 9220,
      "train_runtime": 24139.0048,
      "train_tokens_per_second": 12180.03
    },
    {
      "epoch": 0.6211284675464583,
      "grad_norm": 0.7819775526912318,
      "learning_rate": 7.803519977286075e-06,
      "loss": 0.41295676231384276,
      "num_input_tokens_seen": 294168016,
      "step": 9225,
      "train_runtime": 24151.6562,
      "train_tokens_per_second": 12180.035
    },
    {
      "epoch": 0.6214651225424185,
      "grad_norm": 0.7487835022848892,
      "learning_rate": 7.801330242209176e-06,
      "loss": 0.3901520729064941,
      "num_input_tokens_seen": 294321824,
      "step": 9230,
      "train_runtime": 24164.0927,
      "train_tokens_per_second": 12180.131
    },
    {
      "epoch": 0.6218017775383786,
      "grad_norm": 0.8670256618164732,
      "learning_rate": 7.799139723749207e-06,
      "loss": 0.404341983795166,
      "num_input_tokens_seen": 294475952,
      "step": 9235,
      "train_runtime": 24177.1819,
      "train_tokens_per_second": 12179.912
    },
    {
      "epoch": 0.6221384325343388,
      "grad_norm": 0.8435578140362983,
      "learning_rate": 7.79694842251874e-06,
      "loss": 0.4066864013671875,
      "num_input_tokens_seen": 294636528,
      "step": 9240,
      "train_runtime": 24189.6066,
      "train_tokens_per_second": 12180.294
    },
    {
      "epoch": 0.6224750875302989,
      "grad_norm": 0.7215396442858609,
      "learning_rate": 7.794756339130565e-06,
      "loss": 0.38472962379455566,
      "num_input_tokens_seen": 294799728,
      "step": 9245,
      "train_runtime": 24202.4406,
      "train_tokens_per_second": 12180.579
    },
    {
      "epoch": 0.622811742526259,
      "grad_norm": 0.7765244248354118,
      "learning_rate": 7.792563474197692e-06,
      "loss": 0.3896779537200928,
      "num_input_tokens_seen": 294955560,
      "step": 9250,
      "train_runtime": 24216.2673,
      "train_tokens_per_second": 12180.059
    },
    {
      "epoch": 0.6231483975222192,
      "grad_norm": 0.7285516595637883,
      "learning_rate": 7.79036982833335e-06,
      "loss": 0.409562349319458,
      "num_input_tokens_seen": 295117640,
      "step": 9255,
      "train_runtime": 24228.858,
      "train_tokens_per_second": 12180.419
    },
    {
      "epoch": 0.6234850525181793,
      "grad_norm": 0.802454557810152,
      "learning_rate": 7.78817540215098e-06,
      "loss": 0.3786600351333618,
      "num_input_tokens_seen": 295278544,
      "step": 9260,
      "train_runtime": 24241.9456,
      "train_tokens_per_second": 12180.48
    },
    {
      "epoch": 0.6238217075141395,
      "grad_norm": 0.7418031601802797,
      "learning_rate": 7.78598019626425e-06,
      "loss": 0.3960576057434082,
      "num_input_tokens_seen": 295439992,
      "step": 9265,
      "train_runtime": 24254.4825,
      "train_tokens_per_second": 12180.841
    },
    {
      "epoch": 0.6241583625100996,
      "grad_norm": 0.7436413209478754,
      "learning_rate": 7.783784211287044e-06,
      "loss": 0.3982872486114502,
      "num_input_tokens_seen": 295591312,
      "step": 9270,
      "train_runtime": 24267.9091,
      "train_tokens_per_second": 12180.337
    },
    {
      "epoch": 0.6244950175060597,
      "grad_norm": 0.8225552651030371,
      "learning_rate": 7.78158744783346e-06,
      "loss": 0.4066149711608887,
      "num_input_tokens_seen": 295751928,
      "step": 9275,
      "train_runtime": 24281.8287,
      "train_tokens_per_second": 12179.969
    },
    {
      "epoch": 0.6248316725020199,
      "grad_norm": 0.8776165063676942,
      "learning_rate": 7.779389906517813e-06,
      "loss": 0.36040706634521485,
      "num_input_tokens_seen": 295912176,
      "step": 9280,
      "train_runtime": 24294.8564,
      "train_tokens_per_second": 12180.034
    },
    {
      "epoch": 0.6251683274979801,
      "grad_norm": 0.8096869951293024,
      "learning_rate": 7.777191587954645e-06,
      "loss": 0.3809802532196045,
      "num_input_tokens_seen": 296066848,
      "step": 9285,
      "train_runtime": 24307.744,
      "train_tokens_per_second": 12179.939
    },
    {
      "epoch": 0.6255049824939403,
      "grad_norm": 0.8494302477707368,
      "learning_rate": 7.774992492758704e-06,
      "loss": 0.37774038314819336,
      "num_input_tokens_seen": 296223360,
      "step": 9290,
      "train_runtime": 24320.5823,
      "train_tokens_per_second": 12179.945
    },
    {
      "epoch": 0.6258416374899004,
      "grad_norm": 0.7589427998234359,
      "learning_rate": 7.77279262154496e-06,
      "loss": 0.3866282939910889,
      "num_input_tokens_seen": 296384384,
      "step": 9295,
      "train_runtime": 24333.8313,
      "train_tokens_per_second": 12179.931
    },
    {
      "epoch": 0.6261782924858605,
      "grad_norm": 0.8752839079239674,
      "learning_rate": 7.770591974928604e-06,
      "loss": 0.42908878326416017,
      "num_input_tokens_seen": 296545152,
      "step": 9300,
      "train_runtime": 24346.2857,
      "train_tokens_per_second": 12180.304
    },
    {
      "epoch": 0.6265149474818207,
      "grad_norm": 0.7120098656109812,
      "learning_rate": 7.768390553525034e-06,
      "loss": 0.37722671031951904,
      "num_input_tokens_seen": 296708992,
      "step": 9305,
      "train_runtime": 24358.7147,
      "train_tokens_per_second": 12180.815
    },
    {
      "epoch": 0.6268516024777808,
      "grad_norm": 0.7880343935520849,
      "learning_rate": 7.766188357949875e-06,
      "loss": 0.37330379486083987,
      "num_input_tokens_seen": 296867896,
      "step": 9310,
      "train_runtime": 24371.6229,
      "train_tokens_per_second": 12180.883
    },
    {
      "epoch": 0.627188257473741,
      "grad_norm": 0.8277986295011157,
      "learning_rate": 7.763985388818963e-06,
      "loss": 0.3654982566833496,
      "num_input_tokens_seen": 297026312,
      "step": 9315,
      "train_runtime": 24384.6928,
      "train_tokens_per_second": 12180.851
    },
    {
      "epoch": 0.6275249124697011,
      "grad_norm": 0.8374240498833566,
      "learning_rate": 7.76178164674835e-06,
      "loss": 0.3375447988510132,
      "num_input_tokens_seen": 297183520,
      "step": 9320,
      "train_runtime": 24398.2991,
      "train_tokens_per_second": 12180.502
    },
    {
      "epoch": 0.6278615674656612,
      "grad_norm": 0.7816731165558601,
      "learning_rate": 7.759577132354305e-06,
      "loss": 0.38550314903259275,
      "num_input_tokens_seen": 297345448,
      "step": 9325,
      "train_runtime": 24410.9048,
      "train_tokens_per_second": 12180.845
    },
    {
      "epoch": 0.6281982224616214,
      "grad_norm": 0.7578409675591835,
      "learning_rate": 7.757371846253315e-06,
      "loss": 0.34258270263671875,
      "num_input_tokens_seen": 297501024,
      "step": 9330,
      "train_runtime": 24423.649,
      "train_tokens_per_second": 12180.859
    },
    {
      "epoch": 0.6285348774575815,
      "grad_norm": 0.7505307392363114,
      "learning_rate": 7.755165789062083e-06,
      "loss": 0.4085099220275879,
      "num_input_tokens_seen": 297661824,
      "step": 9335,
      "train_runtime": 24436.2613,
      "train_tokens_per_second": 12181.152
    },
    {
      "epoch": 0.6288715324535417,
      "grad_norm": 0.7448974634311154,
      "learning_rate": 7.752958961397522e-06,
      "loss": 0.38463964462280276,
      "num_input_tokens_seen": 297821848,
      "step": 9340,
      "train_runtime": 24448.6821,
      "train_tokens_per_second": 12181.509
    },
    {
      "epoch": 0.6292081874495018,
      "grad_norm": 0.7353586561374738,
      "learning_rate": 7.750751363876765e-06,
      "loss": 0.3922884941101074,
      "num_input_tokens_seen": 297979992,
      "step": 9345,
      "train_runtime": 24461.5405,
      "train_tokens_per_second": 12181.571
    },
    {
      "epoch": 0.6295448424454619,
      "grad_norm": 0.7840945054857366,
      "learning_rate": 7.748542997117162e-06,
      "loss": 0.3817403316497803,
      "num_input_tokens_seen": 298141720,
      "step": 9350,
      "train_runtime": 24475.0162,
      "train_tokens_per_second": 12181.472
    },
    {
      "epoch": 0.6298814974414221,
      "grad_norm": 0.7872579958313309,
      "learning_rate": 7.746333861736275e-06,
      "loss": 0.35845060348510743,
      "num_input_tokens_seen": 298298488,
      "step": 9355,
      "train_runtime": 24488.9948,
      "train_tokens_per_second": 12180.92
    },
    {
      "epoch": 0.6302181524373822,
      "grad_norm": 0.772511553990588,
      "learning_rate": 7.74412395835188e-06,
      "loss": 0.39458537101745605,
      "num_input_tokens_seen": 298457392,
      "step": 9360,
      "train_runtime": 24501.5503,
      "train_tokens_per_second": 12181.164
    },
    {
      "epoch": 0.6305548074333424,
      "grad_norm": 0.7394341024220819,
      "learning_rate": 7.74191328758197e-06,
      "loss": 0.3871213674545288,
      "num_input_tokens_seen": 298615648,
      "step": 9365,
      "train_runtime": 24514.4067,
      "train_tokens_per_second": 12181.231
    },
    {
      "epoch": 0.6308914624293025,
      "grad_norm": 0.9128110585746823,
      "learning_rate": 7.739701850044752e-06,
      "loss": 0.4151636600494385,
      "num_input_tokens_seen": 298775352,
      "step": 9370,
      "train_runtime": 24527.7166,
      "train_tokens_per_second": 12181.132
    },
    {
      "epoch": 0.6312281174252626,
      "grad_norm": 0.7729114399406369,
      "learning_rate": 7.73748964635865e-06,
      "loss": 0.3765444040298462,
      "num_input_tokens_seen": 298936488,
      "step": 9375,
      "train_runtime": 24540.7166,
      "train_tokens_per_second": 12181.245
    },
    {
      "epoch": 0.6315647724212228,
      "grad_norm": 0.8021515237596587,
      "learning_rate": 7.735276677142297e-06,
      "loss": 0.34388697147369385,
      "num_input_tokens_seen": 299088880,
      "step": 9380,
      "train_runtime": 24554.2559,
      "train_tokens_per_second": 12180.735
    },
    {
      "epoch": 0.6319014274171829,
      "grad_norm": 0.7845684536905075,
      "learning_rate": 7.733062943014542e-06,
      "loss": 0.4200117111206055,
      "num_input_tokens_seen": 299251040,
      "step": 9385,
      "train_runtime": 24567.0177,
      "train_tokens_per_second": 12181.008
    },
    {
      "epoch": 0.632238082413143,
      "grad_norm": 0.7580483437022177,
      "learning_rate": 7.730848444594452e-06,
      "loss": 0.3718400478363037,
      "num_input_tokens_seen": 299412360,
      "step": 9390,
      "train_runtime": 24579.5433,
      "train_tokens_per_second": 12181.364
    },
    {
      "epoch": 0.6325747374091032,
      "grad_norm": 0.7644558018026806,
      "learning_rate": 7.7286331825013e-06,
      "loss": 0.36536850929260256,
      "num_input_tokens_seen": 299569992,
      "step": 9395,
      "train_runtime": 24592.5727,
      "train_tokens_per_second": 12181.32
    },
    {
      "epoch": 0.6329113924050633,
      "grad_norm": 0.8855408815221194,
      "learning_rate": 7.72641715735458e-06,
      "loss": 0.38566153049468993,
      "num_input_tokens_seen": 299728600,
      "step": 9400,
      "train_runtime": 24605.4558,
      "train_tokens_per_second": 12181.388
    },
    {
      "epoch": 0.6332480474010235,
      "grad_norm": 0.860610224670672,
      "learning_rate": 7.724200369773998e-06,
      "loss": 0.4158731460571289,
      "num_input_tokens_seen": 299891664,
      "step": 9405,
      "train_runtime": 24618.6065,
      "train_tokens_per_second": 12181.504
    },
    {
      "epoch": 0.6335847023969836,
      "grad_norm": 0.7674006834943989,
      "learning_rate": 7.721982820379467e-06,
      "loss": 0.374387526512146,
      "num_input_tokens_seen": 300050872,
      "step": 9410,
      "train_runtime": 24632.172,
      "train_tokens_per_second": 12181.259
    },
    {
      "epoch": 0.6339213573929438,
      "grad_norm": 0.8370842010257241,
      "learning_rate": 7.719764509791123e-06,
      "loss": 0.3891880989074707,
      "num_input_tokens_seen": 300200416,
      "step": 9415,
      "train_runtime": 24645.6499,
      "train_tokens_per_second": 12180.665
    },
    {
      "epoch": 0.6342580123889039,
      "grad_norm": 0.8418827874617654,
      "learning_rate": 7.717545438629303e-06,
      "loss": 0.4424652099609375,
      "num_input_tokens_seen": 300364096,
      "step": 9420,
      "train_runtime": 24658.7003,
      "train_tokens_per_second": 12180.857
    },
    {
      "epoch": 0.634594667384864,
      "grad_norm": 0.8044552876573974,
      "learning_rate": 7.71532560751457e-06,
      "loss": 0.3684035301208496,
      "num_input_tokens_seen": 300524592,
      "step": 9425,
      "train_runtime": 24672.1734,
      "train_tokens_per_second": 12180.71
    },
    {
      "epoch": 0.6349313223808242,
      "grad_norm": 0.7916068717380753,
      "learning_rate": 7.713105017067686e-06,
      "loss": 0.3946441411972046,
      "num_input_tokens_seen": 300688432,
      "step": 9430,
      "train_runtime": 24684.5941,
      "train_tokens_per_second": 12181.218
    },
    {
      "epoch": 0.6352679773767843,
      "grad_norm": 0.7835795313529306,
      "learning_rate": 7.710883667909639e-06,
      "loss": 0.37479429244995116,
      "num_input_tokens_seen": 300846544,
      "step": 9435,
      "train_runtime": 24697.5913,
      "train_tokens_per_second": 12181.21
    },
    {
      "epoch": 0.6356046323727444,
      "grad_norm": 0.8073490872966396,
      "learning_rate": 7.708661560661617e-06,
      "loss": 0.4086909294128418,
      "num_input_tokens_seen": 301007096,
      "step": 9440,
      "train_runtime": 24710.0148,
      "train_tokens_per_second": 12181.583
    },
    {
      "epoch": 0.6359412873687046,
      "grad_norm": 0.9170585496865793,
      "learning_rate": 7.706438695945026e-06,
      "loss": 0.3982698917388916,
      "num_input_tokens_seen": 301166248,
      "step": 9445,
      "train_runtime": 24722.6192,
      "train_tokens_per_second": 12181.81
    },
    {
      "epoch": 0.6362779423646647,
      "grad_norm": 0.8099390338660851,
      "learning_rate": 7.704215074381484e-06,
      "loss": 0.38959312438964844,
      "num_input_tokens_seen": 301315752,
      "step": 9450,
      "train_runtime": 24735.2575,
      "train_tokens_per_second": 12181.63
    },
    {
      "epoch": 0.6366145973606249,
      "grad_norm": 0.8832774892333866,
      "learning_rate": 7.70199069659282e-06,
      "loss": 0.3853009223937988,
      "num_input_tokens_seen": 301473864,
      "step": 9455,
      "train_runtime": 24748.4005,
      "train_tokens_per_second": 12181.549
    },
    {
      "epoch": 0.636951252356585,
      "grad_norm": 0.7629907383605017,
      "learning_rate": 7.699765563201075e-06,
      "loss": 0.3948886156082153,
      "num_input_tokens_seen": 301633592,
      "step": 9460,
      "train_runtime": 24760.9519,
      "train_tokens_per_second": 12181.825
    },
    {
      "epoch": 0.6372879073525451,
      "grad_norm": 0.7332082529886655,
      "learning_rate": 7.697539674828498e-06,
      "loss": 0.38031625747680664,
      "num_input_tokens_seen": 301797336,
      "step": 9465,
      "train_runtime": 24774.0405,
      "train_tokens_per_second": 12181.999
    },
    {
      "epoch": 0.6376245623485053,
      "grad_norm": 0.7664332509282752,
      "learning_rate": 7.695313032097554e-06,
      "loss": 0.3588172197341919,
      "num_input_tokens_seen": 301956024,
      "step": 9470,
      "train_runtime": 24787.0756,
      "train_tokens_per_second": 12181.995
    },
    {
      "epoch": 0.6379612173444654,
      "grad_norm": 0.7416778875937772,
      "learning_rate": 7.693085635630913e-06,
      "loss": 0.38371524810791013,
      "num_input_tokens_seen": 302109184,
      "step": 9475,
      "train_runtime": 24799.9308,
      "train_tokens_per_second": 12181.856
    },
    {
      "epoch": 0.6382978723404256,
      "grad_norm": 0.8303200197730307,
      "learning_rate": 7.690857486051463e-06,
      "loss": 0.37233662605285645,
      "num_input_tokens_seen": 302271160,
      "step": 9480,
      "train_runtime": 24812.5754,
      "train_tokens_per_second": 12182.176
    },
    {
      "epoch": 0.6386345273363857,
      "grad_norm": 0.7166377093512025,
      "learning_rate": 7.688628583982298e-06,
      "loss": 0.3579227924346924,
      "num_input_tokens_seen": 302424952,
      "step": 9485,
      "train_runtime": 24825.1099,
      "train_tokens_per_second": 12182.22
    },
    {
      "epoch": 0.6389711823323458,
      "grad_norm": 0.7657009096898025,
      "learning_rate": 7.68639893004672e-06,
      "loss": 0.36395127773284913,
      "num_input_tokens_seen": 302587216,
      "step": 9490,
      "train_runtime": 24838.3303,
      "train_tokens_per_second": 12182.269
    },
    {
      "epoch": 0.639307837328306,
      "grad_norm": 0.7886960877465337,
      "learning_rate": 7.684168524868253e-06,
      "loss": 0.36876890659332273,
      "num_input_tokens_seen": 302751056,
      "step": 9495,
      "train_runtime": 24850.7543,
      "train_tokens_per_second": 12182.771
    },
    {
      "epoch": 0.6396444923242661,
      "grad_norm": 0.7671192065472773,
      "learning_rate": 7.681937369070612e-06,
      "loss": 0.39136033058166503,
      "num_input_tokens_seen": 302907128,
      "step": 9500,
      "train_runtime": 24864.212,
      "train_tokens_per_second": 12182.454
    },
    {
      "epoch": 0.6399811473202263,
      "grad_norm": 0.8186722468718604,
      "learning_rate": 7.67970546327774e-06,
      "loss": 0.3906740665435791,
      "num_input_tokens_seen": 303068112,
      "step": 9505,
      "train_runtime": 24877.0468,
      "train_tokens_per_second": 12182.64
    },
    {
      "epoch": 0.6403178023161864,
      "grad_norm": 1.0085198271159794,
      "learning_rate": 7.67747280811378e-06,
      "loss": 0.3567544460296631,
      "num_input_tokens_seen": 303226688,
      "step": 9510,
      "train_runtime": 24889.935,
      "train_tokens_per_second": 12182.703
    },
    {
      "epoch": 0.6406544573121465,
      "grad_norm": 0.672436175512182,
      "learning_rate": 7.675239404203088e-06,
      "loss": 0.34552655220031736,
      "num_input_tokens_seen": 303385680,
      "step": 9515,
      "train_runtime": 24902.4084,
      "train_tokens_per_second": 12182.985
    },
    {
      "epoch": 0.6409911123081067,
      "grad_norm": 0.8385608894310054,
      "learning_rate": 7.673005252170226e-06,
      "loss": 0.40718975067138674,
      "num_input_tokens_seen": 303548088,
      "step": 9520,
      "train_runtime": 24915.0188,
      "train_tokens_per_second": 12183.338
    },
    {
      "epoch": 0.6413277673040668,
      "grad_norm": 0.8078099550398448,
      "learning_rate": 7.670770352639968e-06,
      "loss": 0.36879839897155764,
      "num_input_tokens_seen": 303708288,
      "step": 9525,
      "train_runtime": 24929.3781,
      "train_tokens_per_second": 12182.746
    },
    {
      "epoch": 0.641664422300027,
      "grad_norm": 0.7597356786701748,
      "learning_rate": 7.668534706237299e-06,
      "loss": 0.3803065299987793,
      "num_input_tokens_seen": 303866264,
      "step": 9530,
      "train_runtime": 24942.5705,
      "train_tokens_per_second": 12182.636
    },
    {
      "epoch": 0.6420010772959871,
      "grad_norm": 0.8329749626046767,
      "learning_rate": 7.666298313587409e-06,
      "loss": 0.4222110748291016,
      "num_input_tokens_seen": 304026192,
      "step": 9535,
      "train_runtime": 24955.753,
      "train_tokens_per_second": 12182.609
    },
    {
      "epoch": 0.6423377322919472,
      "grad_norm": 0.8170117359261758,
      "learning_rate": 7.664061175315696e-06,
      "loss": 0.369573712348938,
      "num_input_tokens_seen": 304182296,
      "step": 9540,
      "train_runtime": 24970.481,
      "train_tokens_per_second": 12181.675
    },
    {
      "epoch": 0.6426743872879074,
      "grad_norm": 0.7612852091189225,
      "learning_rate": 7.661823292047771e-06,
      "loss": 0.3984743595123291,
      "num_input_tokens_seen": 304339032,
      "step": 9545,
      "train_runtime": 24983.3821,
      "train_tokens_per_second": 12181.659
    },
    {
      "epoch": 0.6430110422838675,
      "grad_norm": 0.8527087021681153,
      "learning_rate": 7.659584664409452e-06,
      "loss": 0.3907004833221436,
      "num_input_tokens_seen": 304501992,
      "step": 9550,
      "train_runtime": 24996.8685,
      "train_tokens_per_second": 12181.606
    },
    {
      "epoch": 0.6433476972798277,
      "grad_norm": 0.8144762067752043,
      "learning_rate": 7.65734529302676e-06,
      "loss": 0.34315185546875,
      "num_input_tokens_seen": 304655232,
      "step": 9555,
      "train_runtime": 25009.934,
      "train_tokens_per_second": 12181.369
    },
    {
      "epoch": 0.6436843522757878,
      "grad_norm": 0.7886409332528834,
      "learning_rate": 7.655105178525932e-06,
      "loss": 0.3718048095703125,
      "num_input_tokens_seen": 304816944,
      "step": 9560,
      "train_runtime": 25022.8235,
      "train_tokens_per_second": 12181.557
    },
    {
      "epoch": 0.6440210072717479,
      "grad_norm": 0.8644964037890213,
      "learning_rate": 7.652864321533406e-06,
      "loss": 0.3928435564041138,
      "num_input_tokens_seen": 304976336,
      "step": 9565,
      "train_runtime": 25036.4702,
      "train_tokens_per_second": 12181.283
    },
    {
      "epoch": 0.6443576622677081,
      "grad_norm": 0.7362874646222378,
      "learning_rate": 7.650622722675833e-06,
      "loss": 0.4045905590057373,
      "num_input_tokens_seen": 305138512,
      "step": 9570,
      "train_runtime": 25049.7234,
      "train_tokens_per_second": 12181.313
    },
    {
      "epoch": 0.6446943172636682,
      "grad_norm": 0.8162906089665554,
      "learning_rate": 7.648380382580067e-06,
      "loss": 0.389904260635376,
      "num_input_tokens_seen": 305299944,
      "step": 9575,
      "train_runtime": 25063.3948,
      "train_tokens_per_second": 12181.109
    },
    {
      "epoch": 0.6450309722596284,
      "grad_norm": 0.723990288129423,
      "learning_rate": 7.646137301873173e-06,
      "loss": 0.37366542816162107,
      "num_input_tokens_seen": 305458184,
      "step": 9580,
      "train_runtime": 25077.0908,
      "train_tokens_per_second": 12180.766
    },
    {
      "epoch": 0.6453676272555885,
      "grad_norm": 0.8269042502669389,
      "learning_rate": 7.643893481182421e-06,
      "loss": 0.3626148223876953,
      "num_input_tokens_seen": 305620520,
      "step": 9585,
      "train_runtime": 25090.2536,
      "train_tokens_per_second": 12180.846
    },
    {
      "epoch": 0.6457042822515486,
      "grad_norm": 0.8610511922656466,
      "learning_rate": 7.641648921135287e-06,
      "loss": 0.4106851577758789,
      "num_input_tokens_seen": 305781784,
      "step": 9590,
      "train_runtime": 25103.151,
      "train_tokens_per_second": 12181.012
    },
    {
      "epoch": 0.6460409372475088,
      "grad_norm": 0.9561088313444258,
      "learning_rate": 7.639403622359455e-06,
      "loss": 0.3923514366149902,
      "num_input_tokens_seen": 305936704,
      "step": 9595,
      "train_runtime": 25116.1382,
      "train_tokens_per_second": 12180.882
    },
    {
      "epoch": 0.6463775922434689,
      "grad_norm": 0.7872686779268968,
      "learning_rate": 7.637157585482817e-06,
      "loss": 0.3662883758544922,
      "num_input_tokens_seen": 306096480,
      "step": 9600,
      "train_runtime": 25129.1469,
      "train_tokens_per_second": 12180.934
    },
    {
      "epoch": 0.646714247239429,
      "grad_norm": 0.9076202095685655,
      "learning_rate": 7.63491081113347e-06,
      "loss": 0.4183998107910156,
      "num_input_tokens_seen": 306259616,
      "step": 9605,
      "train_runtime": 25142.1871,
      "train_tokens_per_second": 12181.105
    },
    {
      "epoch": 0.6470509022353892,
      "grad_norm": 0.774755520994566,
      "learning_rate": 7.632663299939718e-06,
      "loss": 0.404583740234375,
      "num_input_tokens_seen": 306420184,
      "step": 9610,
      "train_runtime": 25154.9485,
      "train_tokens_per_second": 12181.308
    },
    {
      "epoch": 0.6473875572313493,
      "grad_norm": 0.9220122077740786,
      "learning_rate": 7.630415052530066e-06,
      "loss": 0.39755327701568605,
      "num_input_tokens_seen": 306580520,
      "step": 9615,
      "train_runtime": 25168.2563,
      "train_tokens_per_second": 12181.238
    },
    {
      "epoch": 0.6477242122273095,
      "grad_norm": 0.8291395476380609,
      "learning_rate": 7.628166069533235e-06,
      "loss": 0.3870555400848389,
      "num_input_tokens_seen": 306739072,
      "step": 9620,
      "train_runtime": 25181.8184,
      "train_tokens_per_second": 12180.974
    },
    {
      "epoch": 0.6480608672232696,
      "grad_norm": 0.8919244139115249,
      "learning_rate": 7.625916351578141e-06,
      "loss": 0.3941284894943237,
      "num_input_tokens_seen": 306898096,
      "step": 9625,
      "train_runtime": 25194.7987,
      "train_tokens_per_second": 12181.01
    },
    {
      "epoch": 0.6483975222192297,
      "grad_norm": 0.6969023858397622,
      "learning_rate": 7.623665899293914e-06,
      "loss": 0.37345314025878906,
      "num_input_tokens_seen": 307057560,
      "step": 9630,
      "train_runtime": 25207.961,
      "train_tokens_per_second": 12180.976
    },
    {
      "epoch": 0.6487341772151899,
      "grad_norm": 0.7518861887107433,
      "learning_rate": 7.621414713309885e-06,
      "loss": 0.34434750080108645,
      "num_input_tokens_seen": 307217376,
      "step": 9635,
      "train_runtime": 25221.0785,
      "train_tokens_per_second": 12180.977
    },
    {
      "epoch": 0.64907083221115,
      "grad_norm": 0.787163708730269,
      "learning_rate": 7.61916279425559e-06,
      "loss": 0.40105438232421875,
      "num_input_tokens_seen": 307378760,
      "step": 9640,
      "train_runtime": 25234.0596,
      "train_tokens_per_second": 12181.106
    },
    {
      "epoch": 0.6494074872071102,
      "grad_norm": 0.7993018306000265,
      "learning_rate": 7.6169101427607716e-06,
      "loss": 0.3876271963119507,
      "num_input_tokens_seen": 307541640,
      "step": 9645,
      "train_runtime": 25247.5068,
      "train_tokens_per_second": 12181.07
    },
    {
      "epoch": 0.6497441422030703,
      "grad_norm": 0.9096689011807212,
      "learning_rate": 7.614656759455375e-06,
      "loss": 0.3859602212905884,
      "num_input_tokens_seen": 307703544,
      "step": 9650,
      "train_runtime": 25259.9592,
      "train_tokens_per_second": 12181.474
    },
    {
      "epoch": 0.6500807971990304,
      "grad_norm": 0.8251040937430476,
      "learning_rate": 7.612402644969555e-06,
      "loss": 0.3871248006820679,
      "num_input_tokens_seen": 307862800,
      "step": 9655,
      "train_runtime": 25273.7073,
      "train_tokens_per_second": 12181.149
    },
    {
      "epoch": 0.6504174521949906,
      "grad_norm": 0.8256474458399202,
      "learning_rate": 7.6101477999336646e-06,
      "loss": 0.3942007064819336,
      "num_input_tokens_seen": 308022760,
      "step": 9660,
      "train_runtime": 25286.9558,
      "train_tokens_per_second": 12181.093
    },
    {
      "epoch": 0.6507541071909507,
      "grad_norm": 0.8791909618349235,
      "learning_rate": 7.607892224978266e-06,
      "loss": 0.38946921825408937,
      "num_input_tokens_seen": 308178968,
      "step": 9665,
      "train_runtime": 25300.2463,
      "train_tokens_per_second": 12180.868
    },
    {
      "epoch": 0.6510907621869109,
      "grad_norm": 0.8032008443292392,
      "learning_rate": 7.605635920734122e-06,
      "loss": 0.3705273151397705,
      "num_input_tokens_seen": 308336208,
      "step": 9670,
      "train_runtime": 25312.8103,
      "train_tokens_per_second": 12181.034
    },
    {
      "epoch": 0.651427417182871,
      "grad_norm": 0.7483294169399682,
      "learning_rate": 7.603378887832202e-06,
      "loss": 0.386503267288208,
      "num_input_tokens_seen": 308492480,
      "step": 9675,
      "train_runtime": 25327.0247,
      "train_tokens_per_second": 12180.368
    },
    {
      "epoch": 0.6517640721788311,
      "grad_norm": 0.7458627092296664,
      "learning_rate": 7.601121126903676e-06,
      "loss": 0.37947180271148684,
      "num_input_tokens_seen": 308652120,
      "step": 9680,
      "train_runtime": 25339.7711,
      "train_tokens_per_second": 12180.541
    },
    {
      "epoch": 0.6521007271747913,
      "grad_norm": 0.8847560409058672,
      "learning_rate": 7.598862638579922e-06,
      "loss": 0.3812716007232666,
      "num_input_tokens_seen": 308814328,
      "step": 9685,
      "train_runtime": 25352.2157,
      "train_tokens_per_second": 12180.96
    },
    {
      "epoch": 0.6524373821707514,
      "grad_norm": 0.8906863338778178,
      "learning_rate": 7.5966034234925205e-06,
      "loss": 0.3827708959579468,
      "num_input_tokens_seen": 308977776,
      "step": 9690,
      "train_runtime": 25364.6197,
      "train_tokens_per_second": 12181.447
    },
    {
      "epoch": 0.6527740371667116,
      "grad_norm": 0.8751351809077597,
      "learning_rate": 7.59434348227325e-06,
      "loss": 0.4421831607818604,
      "num_input_tokens_seen": 309135688,
      "step": 9695,
      "train_runtime": 25377.2098,
      "train_tokens_per_second": 12181.626
    },
    {
      "epoch": 0.6531106921626717,
      "grad_norm": 0.895852937529799,
      "learning_rate": 7.592082815554098e-06,
      "loss": 0.42072525024414065,
      "num_input_tokens_seen": 309294728,
      "step": 9700,
      "train_runtime": 25389.9869,
      "train_tokens_per_second": 12181.76
    },
    {
      "epoch": 0.6534473471586318,
      "grad_norm": 0.8869418187692794,
      "learning_rate": 7.58982142396725e-06,
      "loss": 0.4356858253479004,
      "num_input_tokens_seen": 309449264,
      "step": 9705,
      "train_runtime": 25403.2826,
      "train_tokens_per_second": 12181.468
    },
    {
      "epoch": 0.653784002154592,
      "grad_norm": 0.845593712665352,
      "learning_rate": 7.5875593081451e-06,
      "loss": 0.3816426515579224,
      "num_input_tokens_seen": 309610800,
      "step": 9710,
      "train_runtime": 25416.4944,
      "train_tokens_per_second": 12181.491
    },
    {
      "epoch": 0.6541206571505521,
      "grad_norm": 0.8062485662678909,
      "learning_rate": 7.585296468720243e-06,
      "loss": 0.3842586040496826,
      "num_input_tokens_seen": 309774640,
      "step": 9715,
      "train_runtime": 25428.9282,
      "train_tokens_per_second": 12181.978
    },
    {
      "epoch": 0.6544573121465123,
      "grad_norm": 0.7522165161884529,
      "learning_rate": 7.5830329063254705e-06,
      "loss": 0.4123539924621582,
      "num_input_tokens_seen": 309931472,
      "step": 9720,
      "train_runtime": 25442.467,
      "train_tokens_per_second": 12181.66
    },
    {
      "epoch": 0.6547939671424724,
      "grad_norm": 0.7440081994074869,
      "learning_rate": 7.5807686215937835e-06,
      "loss": 0.3866130828857422,
      "num_input_tokens_seen": 310089504,
      "step": 9725,
      "train_runtime": 25455.1132,
      "train_tokens_per_second": 12181.816
    },
    {
      "epoch": 0.6551306221384325,
      "grad_norm": 0.7430494379805457,
      "learning_rate": 7.578503615158379e-06,
      "loss": 0.3380992889404297,
      "num_input_tokens_seen": 310249216,
      "step": 9730,
      "train_runtime": 25468.227,
      "train_tokens_per_second": 12181.814
    },
    {
      "epoch": 0.6554672771343927,
      "grad_norm": 0.8304087764709839,
      "learning_rate": 7.576237887652662e-06,
      "loss": 0.3985629081726074,
      "num_input_tokens_seen": 310411584,
      "step": 9735,
      "train_runtime": 25481.2626,
      "train_tokens_per_second": 12181.955
    },
    {
      "epoch": 0.6558039321303528,
      "grad_norm": 0.8693652255663747,
      "learning_rate": 7.573971439710235e-06,
      "loss": 0.3582068204879761,
      "num_input_tokens_seen": 310573600,
      "step": 9740,
      "train_runtime": 25494.8904,
      "train_tokens_per_second": 12181.798
    },
    {
      "epoch": 0.656140587126313,
      "grad_norm": 0.8290254643807483,
      "learning_rate": 7.571704271964902e-06,
      "loss": 0.38382997512817385,
      "num_input_tokens_seen": 310729920,
      "step": 9745,
      "train_runtime": 25508.4923,
      "train_tokens_per_second": 12181.43
    },
    {
      "epoch": 0.6564772421222731,
      "grad_norm": 0.8499552647007896,
      "learning_rate": 7.56943638505067e-06,
      "loss": 0.3603231906890869,
      "num_input_tokens_seen": 310892176,
      "step": 9750,
      "train_runtime": 25521.0917,
      "train_tokens_per_second": 12181.774
    },
    {
      "epoch": 0.6568138971182332,
      "grad_norm": 0.7173243765712516,
      "learning_rate": 7.567167779601746e-06,
      "loss": 0.3939405679702759,
      "num_input_tokens_seen": 311055552,
      "step": 9755,
      "train_runtime": 25534.0834,
      "train_tokens_per_second": 12181.974
    },
    {
      "epoch": 0.6571505521141934,
      "grad_norm": 0.8253879299998719,
      "learning_rate": 7.56489845625254e-06,
      "loss": 0.4330586910247803,
      "num_input_tokens_seen": 311214272,
      "step": 9760,
      "train_runtime": 25547.5932,
      "train_tokens_per_second": 12181.745
    },
    {
      "epoch": 0.6574872071101535,
      "grad_norm": 0.754252538780929,
      "learning_rate": 7.562628415637659e-06,
      "loss": 0.38188397884368896,
      "num_input_tokens_seen": 311369408,
      "step": 9765,
      "train_runtime": 25560.5101,
      "train_tokens_per_second": 12181.659
    },
    {
      "epoch": 0.6578238621061137,
      "grad_norm": 0.8506197898430271,
      "learning_rate": 7.560357658391913e-06,
      "loss": 0.4343846797943115,
      "num_input_tokens_seen": 311528856,
      "step": 9770,
      "train_runtime": 25573.2346,
      "train_tokens_per_second": 12181.832
    },
    {
      "epoch": 0.6581605171020738,
      "grad_norm": 0.8013104150536664,
      "learning_rate": 7.558086185150311e-06,
      "loss": 0.39814023971557616,
      "num_input_tokens_seen": 311685720,
      "step": 9775,
      "train_runtime": 25585.8339,
      "train_tokens_per_second": 12181.964
    },
    {
      "epoch": 0.6584971720980339,
      "grad_norm": 0.8549640317267945,
      "learning_rate": 7.555813996548065e-06,
      "loss": 0.40965728759765624,
      "num_input_tokens_seen": 311847536,
      "step": 9780,
      "train_runtime": 25598.2592,
      "train_tokens_per_second": 12182.373
    },
    {
      "epoch": 0.6588338270939941,
      "grad_norm": 0.8083135311596874,
      "learning_rate": 7.553541093220586e-06,
      "loss": 0.37792439460754396,
      "num_input_tokens_seen": 312006176,
      "step": 9785,
      "train_runtime": 25612.3599,
      "train_tokens_per_second": 12181.86
    },
    {
      "epoch": 0.6591704820899542,
      "grad_norm": 0.7756961925661809,
      "learning_rate": 7.551267475803481e-06,
      "loss": 0.3783126354217529,
      "num_input_tokens_seen": 312168360,
      "step": 9790,
      "train_runtime": 25625.4161,
      "train_tokens_per_second": 12181.982
    },
    {
      "epoch": 0.6595071370859144,
      "grad_norm": 0.8703307085760885,
      "learning_rate": 7.5489931449325635e-06,
      "loss": 0.39372930526733396,
      "num_input_tokens_seen": 312329384,
      "step": 9795,
      "train_runtime": 25639.0336,
      "train_tokens_per_second": 12181.792
    },
    {
      "epoch": 0.6598437920818745,
      "grad_norm": 1.0021899032964838,
      "learning_rate": 7.546718101243842e-06,
      "loss": 0.4002224445343018,
      "num_input_tokens_seen": 312488784,
      "step": 9800,
      "train_runtime": 25652.6326,
      "train_tokens_per_second": 12181.548
    },
    {
      "epoch": 0.6601804470778346,
      "grad_norm": 0.8820199639721414,
      "learning_rate": 7.544442345373522e-06,
      "loss": 0.3753774642944336,
      "num_input_tokens_seen": 312646848,
      "step": 9805,
      "train_runtime": 25666.2823,
      "train_tokens_per_second": 12181.228
    },
    {
      "epoch": 0.6605171020737948,
      "grad_norm": 0.6903212317132906,
      "learning_rate": 7.542165877958014e-06,
      "loss": 0.3762394428253174,
      "num_input_tokens_seen": 312804232,
      "step": 9810,
      "train_runtime": 25679.0291,
      "train_tokens_per_second": 12181.311
    },
    {
      "epoch": 0.6608537570697549,
      "grad_norm": 0.8460053276972204,
      "learning_rate": 7.539888699633926e-06,
      "loss": 0.39285638332366946,
      "num_input_tokens_seen": 312966160,
      "step": 9815,
      "train_runtime": 25691.5839,
      "train_tokens_per_second": 12181.661
    },
    {
      "epoch": 0.661190412065715,
      "grad_norm": 0.8187018480492271,
      "learning_rate": 7.5376108110380606e-06,
      "loss": 0.35872712135314944,
      "num_input_tokens_seen": 313130000,
      "step": 9820,
      "train_runtime": 25704.0018,
      "train_tokens_per_second": 12182.15
    },
    {
      "epoch": 0.6615270670616752,
      "grad_norm": 0.8070742336951411,
      "learning_rate": 7.535332212807426e-06,
      "loss": 0.35670814514160154,
      "num_input_tokens_seen": 313285744,
      "step": 9825,
      "train_runtime": 25718.1151,
      "train_tokens_per_second": 12181.52
    },
    {
      "epoch": 0.6618637220576353,
      "grad_norm": 0.8578920473261384,
      "learning_rate": 7.53305290557922e-06,
      "loss": 0.4046154975891113,
      "num_input_tokens_seen": 313449584,
      "step": 9830,
      "train_runtime": 25730.5423,
      "train_tokens_per_second": 12182.005
    },
    {
      "epoch": 0.6622003770535955,
      "grad_norm": 0.8258896744546548,
      "learning_rate": 7.530772889990848e-06,
      "loss": 0.3669726371765137,
      "num_input_tokens_seen": 313613424,
      "step": 9835,
      "train_runtime": 25742.9774,
      "train_tokens_per_second": 12182.485
    },
    {
      "epoch": 0.6625370320495556,
      "grad_norm": 0.7431375122017521,
      "learning_rate": 7.528492166679907e-06,
      "loss": 0.3685587406158447,
      "num_input_tokens_seen": 313776432,
      "step": 9840,
      "train_runtime": 25756.072,
      "train_tokens_per_second": 12182.62
    },
    {
      "epoch": 0.6628736870455157,
      "grad_norm": 0.765884184679842,
      "learning_rate": 7.526210736284194e-06,
      "loss": 0.3904863834381104,
      "num_input_tokens_seen": 313929704,
      "step": 9845,
      "train_runtime": 25769.4615,
      "train_tokens_per_second": 12182.238
    },
    {
      "epoch": 0.6632103420414759,
      "grad_norm": 0.8153739338176538,
      "learning_rate": 7.523928599441706e-06,
      "loss": 0.3999138355255127,
      "num_input_tokens_seen": 314090408,
      "step": 9850,
      "train_runtime": 25782.3192,
      "train_tokens_per_second": 12182.395
    },
    {
      "epoch": 0.663546997037436,
      "grad_norm": 0.799265914002858,
      "learning_rate": 7.5216457567906295e-06,
      "loss": 0.3770828008651733,
      "num_input_tokens_seen": 314251912,
      "step": 9855,
      "train_runtime": 25794.8459,
      "train_tokens_per_second": 12182.74
    },
    {
      "epoch": 0.6638836520333962,
      "grad_norm": 0.915491347028832,
      "learning_rate": 7.519362208969362e-06,
      "loss": 0.37642536163330076,
      "num_input_tokens_seen": 314414008,
      "step": 9860,
      "train_runtime": 25808.1348,
      "train_tokens_per_second": 12182.748
    },
    {
      "epoch": 0.6642203070293563,
      "grad_norm": 0.7841136256157107,
      "learning_rate": 7.5170779566164855e-06,
      "loss": 0.40261316299438477,
      "num_input_tokens_seen": 314576696,
      "step": 9865,
      "train_runtime": 25820.7744,
      "train_tokens_per_second": 12183.085
    },
    {
      "epoch": 0.6645569620253164,
      "grad_norm": 0.8448488716348264,
      "learning_rate": 7.514793000370783e-06,
      "loss": 0.39379613399505614,
      "num_input_tokens_seen": 314732328,
      "step": 9870,
      "train_runtime": 25833.9209,
      "train_tokens_per_second": 12182.91
    },
    {
      "epoch": 0.6648936170212766,
      "grad_norm": 0.8573553585615501,
      "learning_rate": 7.512507340871239e-06,
      "loss": 0.3863104820251465,
      "num_input_tokens_seen": 314893408,
      "step": 9875,
      "train_runtime": 25847.1427,
      "train_tokens_per_second": 12182.91
    },
    {
      "epoch": 0.6652302720172367,
      "grad_norm": 0.8665271548059639,
      "learning_rate": 7.510220978757028e-06,
      "loss": 0.4033992767333984,
      "num_input_tokens_seen": 315052296,
      "step": 9880,
      "train_runtime": 25860.1347,
      "train_tokens_per_second": 12182.933
    },
    {
      "epoch": 0.6655669270131969,
      "grad_norm": 0.846832427144157,
      "learning_rate": 7.507933914667524e-06,
      "loss": 0.4033173084259033,
      "num_input_tokens_seen": 315215136,
      "step": 9885,
      "train_runtime": 25872.8904,
      "train_tokens_per_second": 12183.221
    },
    {
      "epoch": 0.665903582009157,
      "grad_norm": 0.8821013828118095,
      "learning_rate": 7.505646149242299e-06,
      "loss": 0.38404011726379395,
      "num_input_tokens_seen": 315373928,
      "step": 9890,
      "train_runtime": 25885.4926,
      "train_tokens_per_second": 12183.424
    },
    {
      "epoch": 0.6662402370051171,
      "grad_norm": 0.6724064246825471,
      "learning_rate": 7.503357683121119e-06,
      "loss": 0.3648186683654785,
      "num_input_tokens_seen": 315537480,
      "step": 9895,
      "train_runtime": 25898.5098,
      "train_tokens_per_second": 12183.615
    },
    {
      "epoch": 0.6665768920010773,
      "grad_norm": 0.8074596985920827,
      "learning_rate": 7.501068516943944e-06,
      "loss": 0.32072086334228517,
      "num_input_tokens_seen": 315695248,
      "step": 9900,
      "train_runtime": 25911.1257,
      "train_tokens_per_second": 12183.772
    },
    {
      "epoch": 0.6669135469970374,
      "grad_norm": 0.7785943996479217,
      "learning_rate": 7.498778651350934e-06,
      "loss": 0.38613865375518797,
      "num_input_tokens_seen": 315859088,
      "step": 9905,
      "train_runtime": 25923.5549,
      "train_tokens_per_second": 12184.251
    },
    {
      "epoch": 0.6672502019929976,
      "grad_norm": 0.802215818779068,
      "learning_rate": 7.496488086982443e-06,
      "loss": 0.36365737915039065,
      "num_input_tokens_seen": 316018344,
      "step": 9910,
      "train_runtime": 25936.1738,
      "train_tokens_per_second": 12184.463
    },
    {
      "epoch": 0.6675868569889577,
      "grad_norm": 0.8375789729653921,
      "learning_rate": 7.494196824479019e-06,
      "loss": 0.37540483474731445,
      "num_input_tokens_seen": 316177272,
      "step": 9915,
      "train_runtime": 25950.4645,
      "train_tokens_per_second": 12183.877
    },
    {
      "epoch": 0.6679235119849178,
      "grad_norm": 0.8435926622684569,
      "learning_rate": 7.491904864481406e-06,
      "loss": 0.3755911111831665,
      "num_input_tokens_seen": 316331600,
      "step": 9920,
      "train_runtime": 25963.585,
      "train_tokens_per_second": 12183.664
    },
    {
      "epoch": 0.668260166980878,
      "grad_norm": 0.6913111487006595,
      "learning_rate": 7.489612207630543e-06,
      "loss": 0.3194437503814697,
      "num_input_tokens_seen": 316495392,
      "step": 9925,
      "train_runtime": 25976.6698,
      "train_tokens_per_second": 12183.832
    },
    {
      "epoch": 0.6685968219768381,
      "grad_norm": 0.8020540151106973,
      "learning_rate": 7.487318854567566e-06,
      "loss": 0.38332204818725585,
      "num_input_tokens_seen": 316651384,
      "step": 9930,
      "train_runtime": 25989.5053,
      "train_tokens_per_second": 12183.817
    },
    {
      "epoch": 0.6689334769727983,
      "grad_norm": 0.7056521943298735,
      "learning_rate": 7.485024805933803e-06,
      "loss": 0.37098064422607424,
      "num_input_tokens_seen": 316808960,
      "step": 9935,
      "train_runtime": 26003.1928,
      "train_tokens_per_second": 12183.464
    },
    {
      "epoch": 0.6692701319687584,
      "grad_norm": 0.7121499704423223,
      "learning_rate": 7.482730062370776e-06,
      "loss": 0.36148796081542967,
      "num_input_tokens_seen": 316965456,
      "step": 9940,
      "train_runtime": 26015.9718,
      "train_tokens_per_second": 12183.495
    },
    {
      "epoch": 0.6696067869647185,
      "grad_norm": 0.8781179478463884,
      "learning_rate": 7.480434624520205e-06,
      "loss": 0.355879020690918,
      "num_input_tokens_seen": 317129296,
      "step": 9945,
      "train_runtime": 26028.4002,
      "train_tokens_per_second": 12183.972
    },
    {
      "epoch": 0.6699434419606787,
      "grad_norm": 0.8044029137899269,
      "learning_rate": 7.478138493024001e-06,
      "loss": 0.418122386932373,
      "num_input_tokens_seen": 317288776,
      "step": 9950,
      "train_runtime": 26041.047,
      "train_tokens_per_second": 12184.179
    },
    {
      "epoch": 0.6702800969566388,
      "grad_norm": 0.7400060610621629,
      "learning_rate": 7.475841668524268e-06,
      "loss": 0.3775538682937622,
      "num_input_tokens_seen": 317448320,
      "step": 9955,
      "train_runtime": 26054.735,
      "train_tokens_per_second": 12183.901
    },
    {
      "epoch": 0.670616751952599,
      "grad_norm": 0.8600938989506957,
      "learning_rate": 7.473544151663309e-06,
      "loss": 0.4203483581542969,
      "num_input_tokens_seen": 317606408,
      "step": 9960,
      "train_runtime": 26067.7335,
      "train_tokens_per_second": 12183.89
    },
    {
      "epoch": 0.6709534069485591,
      "grad_norm": 0.9595844992156295,
      "learning_rate": 7.471245943083615e-06,
      "loss": 0.37787139415740967,
      "num_input_tokens_seen": 317767008,
      "step": 9965,
      "train_runtime": 26081.1934,
      "train_tokens_per_second": 12183.76
    },
    {
      "epoch": 0.6712900619445192,
      "grad_norm": 0.7062887868439995,
      "learning_rate": 7.468947043427873e-06,
      "loss": 0.3433822154998779,
      "num_input_tokens_seen": 317928392,
      "step": 9970,
      "train_runtime": 26094.5822,
      "train_tokens_per_second": 12183.693
    },
    {
      "epoch": 0.6716267169404794,
      "grad_norm": 0.7881467977054103,
      "learning_rate": 7.466647453338965e-06,
      "loss": 0.3903756380081177,
      "num_input_tokens_seen": 318090712,
      "step": 9975,
      "train_runtime": 26108.1966,
      "train_tokens_per_second": 12183.557
    },
    {
      "epoch": 0.6719633719364395,
      "grad_norm": 0.7606740723482138,
      "learning_rate": 7.464347173459958e-06,
      "loss": 0.39801740646362305,
      "num_input_tokens_seen": 318254552,
      "step": 9980,
      "train_runtime": 26120.6882,
      "train_tokens_per_second": 12184.003
    },
    {
      "epoch": 0.6723000269323997,
      "grad_norm": 0.690811652488791,
      "learning_rate": 7.462046204434126e-06,
      "loss": 0.3544230222702026,
      "num_input_tokens_seen": 318418000,
      "step": 9985,
      "train_runtime": 26133.728,
      "train_tokens_per_second": 12184.178
    },
    {
      "epoch": 0.6726366819283598,
      "grad_norm": 0.8698234595344321,
      "learning_rate": 7.459744546904922e-06,
      "loss": 0.38443453311920167,
      "num_input_tokens_seen": 318576648,
      "step": 9990,
      "train_runtime": 26146.646,
      "train_tokens_per_second": 12184.226
    },
    {
      "epoch": 0.6729733369243199,
      "grad_norm": 0.7747255494704058,
      "learning_rate": 7.457442201516001e-06,
      "loss": 0.34813766479492186,
      "num_input_tokens_seen": 318737056,
      "step": 9995,
      "train_runtime": 26159.4224,
      "train_tokens_per_second": 12184.407
    },
    {
      "epoch": 0.6733099919202801,
      "grad_norm": 0.7228010570772834,
      "learning_rate": 7.455139168911203e-06,
      "loss": 0.3796975135803223,
      "num_input_tokens_seen": 318887736,
      "step": 10000,
      "train_runtime": 26172.7351,
      "train_tokens_per_second": 12183.967
    },
    {
      "epoch": 0.6736466469162402,
      "grad_norm": 0.8105541741528355,
      "learning_rate": 7.452835449734564e-06,
      "loss": 0.3744520664215088,
      "num_input_tokens_seen": 319043976,
      "step": 10005,
      "train_runtime": 26185.7809,
      "train_tokens_per_second": 12183.863
    },
    {
      "epoch": 0.6739833019122003,
      "grad_norm": 0.7412754308474214,
      "learning_rate": 7.450531044630315e-06,
      "loss": 0.3376976490020752,
      "num_input_tokens_seen": 319202152,
      "step": 10010,
      "train_runtime": 26199.1485,
      "train_tokens_per_second": 12183.684
    },
    {
      "epoch": 0.6743199569081605,
      "grad_norm": 0.8514432321336665,
      "learning_rate": 7.448225954242874e-06,
      "loss": 0.38923444747924807,
      "num_input_tokens_seen": 319363976,
      "step": 10015,
      "train_runtime": 26211.7837,
      "train_tokens_per_second": 12183.985
    },
    {
      "epoch": 0.6746566119041206,
      "grad_norm": 0.8536921056017616,
      "learning_rate": 7.44592017921685e-06,
      "loss": 0.3782106876373291,
      "num_input_tokens_seen": 319523120,
      "step": 10020,
      "train_runtime": 26224.5069,
      "train_tokens_per_second": 12184.142
    },
    {
      "epoch": 0.6749932669000808,
      "grad_norm": 0.7833643350090261,
      "learning_rate": 7.443613720197048e-06,
      "loss": 0.39757614135742186,
      "num_input_tokens_seen": 319681568,
      "step": 10025,
      "train_runtime": 26237.4441,
      "train_tokens_per_second": 12184.173
    },
    {
      "epoch": 0.6753299218960409,
      "grad_norm": 0.800564589256561,
      "learning_rate": 7.44130657782846e-06,
      "loss": 0.3686829090118408,
      "num_input_tokens_seen": 319839112,
      "step": 10030,
      "train_runtime": 26250.2355,
      "train_tokens_per_second": 12184.238
    },
    {
      "epoch": 0.675666576892001,
      "grad_norm": 0.8081185649869477,
      "learning_rate": 7.438998752756274e-06,
      "loss": 0.38120274543762206,
      "num_input_tokens_seen": 319999856,
      "step": 10035,
      "train_runtime": 26262.7411,
      "train_tokens_per_second": 12184.557
    },
    {
      "epoch": 0.6760032318879612,
      "grad_norm": 0.7628731233372505,
      "learning_rate": 7.436690245625864e-06,
      "loss": 0.37954211235046387,
      "num_input_tokens_seen": 320159008,
      "step": 10040,
      "train_runtime": 26275.8603,
      "train_tokens_per_second": 12184.53
    },
    {
      "epoch": 0.6763398868839213,
      "grad_norm": 0.8869529404521618,
      "learning_rate": 7.434381057082797e-06,
      "loss": 0.36778442859649657,
      "num_input_tokens_seen": 320314224,
      "step": 10045,
      "train_runtime": 26289.1964,
      "train_tokens_per_second": 12184.253
    },
    {
      "epoch": 0.6766765418798815,
      "grad_norm": 0.8579576724943593,
      "learning_rate": 7.4320711877728305e-06,
      "loss": 0.4174938678741455,
      "num_input_tokens_seen": 320474872,
      "step": 10050,
      "train_runtime": 26301.9712,
      "train_tokens_per_second": 12184.443
    },
    {
      "epoch": 0.6770131968758416,
      "grad_norm": 0.7499299413591257,
      "learning_rate": 7.429760638341909e-06,
      "loss": 0.3572435140609741,
      "num_input_tokens_seen": 320631120,
      "step": 10055,
      "train_runtime": 26316.0252,
      "train_tokens_per_second": 12183.873
    },
    {
      "epoch": 0.6773498518718017,
      "grad_norm": 0.8423251053815365,
      "learning_rate": 7.427449409436176e-06,
      "loss": 0.4229987144470215,
      "num_input_tokens_seen": 320789920,
      "step": 10060,
      "train_runtime": 26328.9762,
      "train_tokens_per_second": 12183.912
    },
    {
      "epoch": 0.6776865068677619,
      "grad_norm": 0.6401417481421369,
      "learning_rate": 7.425137501701955e-06,
      "loss": 0.36402719020843505,
      "num_input_tokens_seen": 320952392,
      "step": 10065,
      "train_runtime": 26341.628,
      "train_tokens_per_second": 12184.228
    },
    {
      "epoch": 0.678023161863722,
      "grad_norm": 0.9335939898709256,
      "learning_rate": 7.422824915785765e-06,
      "loss": 0.36553995609283446,
      "num_input_tokens_seen": 321113200,
      "step": 10070,
      "train_runtime": 26354.6757,
      "train_tokens_per_second": 12184.297
    },
    {
      "epoch": 0.6783598168596822,
      "grad_norm": 0.846171357113435,
      "learning_rate": 7.420511652334313e-06,
      "loss": 0.38360445499420165,
      "num_input_tokens_seen": 321267120,
      "step": 10075,
      "train_runtime": 26367.7188,
      "train_tokens_per_second": 12184.107
    },
    {
      "epoch": 0.6786964718556423,
      "grad_norm": 0.7435239539694964,
      "learning_rate": 7.418197711994495e-06,
      "loss": 0.3540541648864746,
      "num_input_tokens_seen": 321425072,
      "step": 10080,
      "train_runtime": 26380.6528,
      "train_tokens_per_second": 12184.121
    },
    {
      "epoch": 0.6790331268516024,
      "grad_norm": 0.8459317207465824,
      "learning_rate": 7.4158830954133984e-06,
      "loss": 0.3685738563537598,
      "num_input_tokens_seen": 321588656,
      "step": 10085,
      "train_runtime": 26393.6008,
      "train_tokens_per_second": 12184.342
    },
    {
      "epoch": 0.6793697818475626,
      "grad_norm": 0.7102508995085847,
      "learning_rate": 7.413567803238298e-06,
      "loss": 0.36223621368408204,
      "num_input_tokens_seen": 321745744,
      "step": 10090,
      "train_runtime": 26406.7361,
      "train_tokens_per_second": 12184.23
    },
    {
      "epoch": 0.6797064368435227,
      "grad_norm": 0.7180570697719586,
      "learning_rate": 7.411251836116658e-06,
      "loss": 0.38073318004608153,
      "num_input_tokens_seen": 321909584,
      "step": 10095,
      "train_runtime": 26419.1613,
      "train_tokens_per_second": 12184.701
    },
    {
      "epoch": 0.6800430918394829,
      "grad_norm": 0.7381396116767438,
      "learning_rate": 7.408935194696132e-06,
      "loss": 0.3999310493469238,
      "num_input_tokens_seen": 322069320,
      "step": 10100,
      "train_runtime": 26432.2128,
      "train_tokens_per_second": 12184.728
    },
    {
      "epoch": 0.680379746835443,
      "grad_norm": 0.7846664802668101,
      "learning_rate": 7.406617879624558e-06,
      "loss": 0.39309277534484866,
      "num_input_tokens_seen": 322232920,
      "step": 10105,
      "train_runtime": 26445.2227,
      "train_tokens_per_second": 12184.92
    },
    {
      "epoch": 0.6807164018314031,
      "grad_norm": 0.724559884967463,
      "learning_rate": 7.40429989154997e-06,
      "loss": 0.36521596908569337,
      "num_input_tokens_seen": 322391680,
      "step": 10110,
      "train_runtime": 26458.8428,
      "train_tokens_per_second": 12184.648
    },
    {
      "epoch": 0.6810530568273633,
      "grad_norm": 0.8349427054484625,
      "learning_rate": 7.4019812311205826e-06,
      "loss": 0.40368123054504396,
      "num_input_tokens_seen": 322552192,
      "step": 10115,
      "train_runtime": 26471.2585,
      "train_tokens_per_second": 12184.996
    },
    {
      "epoch": 0.6813897118233234,
      "grad_norm": 0.7608025074156023,
      "learning_rate": 7.399661898984805e-06,
      "loss": 0.39532663822174074,
      "num_input_tokens_seen": 322716032,
      "step": 10120,
      "train_runtime": 26483.6791,
      "train_tokens_per_second": 12185.468
    },
    {
      "epoch": 0.6817263668192836,
      "grad_norm": 0.8229171737607052,
      "learning_rate": 7.397341895791228e-06,
      "loss": 0.39187328815460204,
      "num_input_tokens_seen": 322874208,
      "step": 10125,
      "train_runtime": 26496.6823,
      "train_tokens_per_second": 12185.458
    },
    {
      "epoch": 0.6820630218152437,
      "grad_norm": 0.957713726541619,
      "learning_rate": 7.395021222188634e-06,
      "loss": 0.36869134902954104,
      "num_input_tokens_seen": 323032128,
      "step": 10130,
      "train_runtime": 26509.6823,
      "train_tokens_per_second": 12185.439
    },
    {
      "epoch": 0.6823996768112038,
      "grad_norm": 0.7835244543232316,
      "learning_rate": 7.392699878825993e-06,
      "loss": 0.41825523376464846,
      "num_input_tokens_seen": 323189592,
      "step": 10135,
      "train_runtime": 26522.9146,
      "train_tokens_per_second": 12185.297
    },
    {
      "epoch": 0.682736331807164,
      "grad_norm": 0.7595415205703406,
      "learning_rate": 7.390377866352462e-06,
      "loss": 0.3895062685012817,
      "num_input_tokens_seen": 323343144,
      "step": 10140,
      "train_runtime": 26536.1195,
      "train_tokens_per_second": 12185.02
    },
    {
      "epoch": 0.6830729868031241,
      "grad_norm": 0.8183701998233642,
      "learning_rate": 7.388055185417382e-06,
      "loss": 0.39305891990661623,
      "num_input_tokens_seen": 323505504,
      "step": 10145,
      "train_runtime": 26549.3029,
      "train_tokens_per_second": 12185.085
    },
    {
      "epoch": 0.6834096417990843,
      "grad_norm": 0.7507889166165261,
      "learning_rate": 7.3857318366702844e-06,
      "loss": 0.3604459285736084,
      "num_input_tokens_seen": 323667624,
      "step": 10150,
      "train_runtime": 26562.5448,
      "train_tokens_per_second": 12185.114
    },
    {
      "epoch": 0.6837462967950444,
      "grad_norm": 0.8062848777006045,
      "learning_rate": 7.383407820760885e-06,
      "loss": 0.39438633918762206,
      "num_input_tokens_seen": 323828232,
      "step": 10155,
      "train_runtime": 26574.9701,
      "train_tokens_per_second": 12185.46
    },
    {
      "epoch": 0.6840829517910045,
      "grad_norm": 0.7363876304189422,
      "learning_rate": 7.381083138339089e-06,
      "loss": 0.37736239433288576,
      "num_input_tokens_seen": 323985632,
      "step": 10160,
      "train_runtime": 26588.1946,
      "train_tokens_per_second": 12185.319
    },
    {
      "epoch": 0.6844196067869647,
      "grad_norm": 0.806794215618462,
      "learning_rate": 7.378757790054988e-06,
      "loss": 0.3931479215621948,
      "num_input_tokens_seen": 324148960,
      "step": 10165,
      "train_runtime": 26600.6103,
      "train_tokens_per_second": 12185.772
    },
    {
      "epoch": 0.6847562617829248,
      "grad_norm": 0.8723766735969465,
      "learning_rate": 7.376431776558855e-06,
      "loss": 0.3919412851333618,
      "num_input_tokens_seen": 324310152,
      "step": 10170,
      "train_runtime": 26614.004,
      "train_tokens_per_second": 12185.696
    },
    {
      "epoch": 0.685092916778885,
      "grad_norm": 0.8292000512038669,
      "learning_rate": 7.3741050985011546e-06,
      "loss": 0.3860538721084595,
      "num_input_tokens_seen": 324472272,
      "step": 10175,
      "train_runtime": 26627.2282,
      "train_tokens_per_second": 12185.732
    },
    {
      "epoch": 0.6854295717748451,
      "grad_norm": 0.7307669046697011,
      "learning_rate": 7.371777756532531e-06,
      "loss": 0.3597079277038574,
      "num_input_tokens_seen": 324631096,
      "step": 10180,
      "train_runtime": 26639.7764,
      "train_tokens_per_second": 12185.954
    },
    {
      "epoch": 0.6857662267708052,
      "grad_norm": 0.8106411980276297,
      "learning_rate": 7.369449751303823e-06,
      "loss": 0.40822649002075195,
      "num_input_tokens_seen": 324787416,
      "step": 10185,
      "train_runtime": 26652.3464,
      "train_tokens_per_second": 12186.072
    },
    {
      "epoch": 0.6861028817667654,
      "grad_norm": 0.6418435864980344,
      "learning_rate": 7.367121083466048e-06,
      "loss": 0.35878982543945315,
      "num_input_tokens_seen": 324946528,
      "step": 10190,
      "train_runtime": 26664.9286,
      "train_tokens_per_second": 12186.289
    },
    {
      "epoch": 0.6864395367627255,
      "grad_norm": 0.8858295882319781,
      "learning_rate": 7.364791753670409e-06,
      "loss": 0.38283023834228513,
      "num_input_tokens_seen": 325107840,
      "step": 10195,
      "train_runtime": 26677.8385,
      "train_tokens_per_second": 12186.439
    },
    {
      "epoch": 0.6867761917586857,
      "grad_norm": 0.7859562889427392,
      "learning_rate": 7.362461762568298e-06,
      "loss": 0.3875227689743042,
      "num_input_tokens_seen": 325268576,
      "step": 10200,
      "train_runtime": 26690.5841,
      "train_tokens_per_second": 12186.641
    },
    {
      "epoch": 0.6871128467546458,
      "grad_norm": 1.0748034827954138,
      "learning_rate": 7.360131110811287e-06,
      "loss": 0.3560051918029785,
      "num_input_tokens_seen": 325428976,
      "step": 10205,
      "train_runtime": 26703.6801,
      "train_tokens_per_second": 12186.671
    },
    {
      "epoch": 0.6874495017506059,
      "grad_norm": 0.7510334458885797,
      "learning_rate": 7.357799799051138e-06,
      "loss": 0.36168246269226073,
      "num_input_tokens_seen": 325587400,
      "step": 10210,
      "train_runtime": 26716.1582,
      "train_tokens_per_second": 12186.909
    },
    {
      "epoch": 0.6877861567465661,
      "grad_norm": 0.9154613005985477,
      "learning_rate": 7.355467827939794e-06,
      "loss": 0.39054856300354,
      "num_input_tokens_seen": 325741104,
      "step": 10215,
      "train_runtime": 26728.6565,
      "train_tokens_per_second": 12186.961
    },
    {
      "epoch": 0.6881228117425262,
      "grad_norm": 0.8277638856527207,
      "learning_rate": 7.353135198129382e-06,
      "loss": 0.3998368501663208,
      "num_input_tokens_seen": 325894768,
      "step": 10220,
      "train_runtime": 26741.1182,
      "train_tokens_per_second": 12187.028
    },
    {
      "epoch": 0.6884594667384863,
      "grad_norm": 0.7956122054130275,
      "learning_rate": 7.350801910272216e-06,
      "loss": 0.3863977909088135,
      "num_input_tokens_seen": 326055928,
      "step": 10225,
      "train_runtime": 26753.7944,
      "train_tokens_per_second": 12187.278
    },
    {
      "epoch": 0.6887961217344465,
      "grad_norm": 0.8236927655364495,
      "learning_rate": 7.348467965020791e-06,
      "loss": 0.3548075437545776,
      "num_input_tokens_seen": 326214104,
      "step": 10230,
      "train_runtime": 26766.7281,
      "train_tokens_per_second": 12187.298
    },
    {
      "epoch": 0.6891327767304066,
      "grad_norm": 0.8260095036629345,
      "learning_rate": 7.3461333630277885e-06,
      "loss": 0.37176966667175293,
      "num_input_tokens_seen": 326365360,
      "step": 10235,
      "train_runtime": 26779.7742,
      "train_tokens_per_second": 12187.009
    },
    {
      "epoch": 0.6894694317263668,
      "grad_norm": 0.8776993441562492,
      "learning_rate": 7.343798104946074e-06,
      "loss": 0.375214672088623,
      "num_input_tokens_seen": 326525000,
      "step": 10240,
      "train_runtime": 26792.6634,
      "train_tokens_per_second": 12187.105
    },
    {
      "epoch": 0.6898060867223269,
      "grad_norm": 0.8444867691580653,
      "learning_rate": 7.341462191428692e-06,
      "loss": 0.3909226179122925,
      "num_input_tokens_seen": 326682568,
      "step": 10245,
      "train_runtime": 26805.9788,
      "train_tokens_per_second": 12186.929
    },
    {
      "epoch": 0.690142741718287,
      "grad_norm": 0.7456799749352725,
      "learning_rate": 7.339125623128874e-06,
      "loss": 0.39707517623901367,
      "num_input_tokens_seen": 326844200,
      "step": 10250,
      "train_runtime": 26818.9887,
      "train_tokens_per_second": 12187.044
    },
    {
      "epoch": 0.6904793967142472,
      "grad_norm": 0.7316673981380943,
      "learning_rate": 7.336788400700035e-06,
      "loss": 0.38560636043548585,
      "num_input_tokens_seen": 327001984,
      "step": 10255,
      "train_runtime": 26832.1099,
      "train_tokens_per_second": 12186.965
    },
    {
      "epoch": 0.6908160517102074,
      "grad_norm": 0.9156039289988064,
      "learning_rate": 7.334450524795771e-06,
      "loss": 0.4110163688659668,
      "num_input_tokens_seen": 327163072,
      "step": 10260,
      "train_runtime": 26844.7205,
      "train_tokens_per_second": 12187.241
    },
    {
      "epoch": 0.6911527067061676,
      "grad_norm": 0.7915660742753489,
      "learning_rate": 7.332111996069863e-06,
      "loss": 0.38647310733795165,
      "num_input_tokens_seen": 327324936,
      "step": 10265,
      "train_runtime": 26857.9086,
      "train_tokens_per_second": 12187.283
    },
    {
      "epoch": 0.6914893617021277,
      "grad_norm": 0.8704922928746365,
      "learning_rate": 7.32977281517627e-06,
      "loss": 0.4231839179992676,
      "num_input_tokens_seen": 327478368,
      "step": 10270,
      "train_runtime": 26871.2094,
      "train_tokens_per_second": 12186.961
    },
    {
      "epoch": 0.6918260166980879,
      "grad_norm": 0.805944140851274,
      "learning_rate": 7.3274329827691406e-06,
      "loss": 0.3946662902832031,
      "num_input_tokens_seen": 327632864,
      "step": 10275,
      "train_runtime": 26884.1836,
      "train_tokens_per_second": 12186.826
    },
    {
      "epoch": 0.692162671694048,
      "grad_norm": 0.7776015723872965,
      "learning_rate": 7.325092499502796e-06,
      "loss": 0.36268651485443115,
      "num_input_tokens_seen": 327794968,
      "step": 10280,
      "train_runtime": 26896.6691,
      "train_tokens_per_second": 12187.196
    },
    {
      "epoch": 0.6924993266900081,
      "grad_norm": 0.752726335448874,
      "learning_rate": 7.322751366031751e-06,
      "loss": 0.38288938999176025,
      "num_input_tokens_seen": 327956088,
      "step": 10285,
      "train_runtime": 26909.1346,
      "train_tokens_per_second": 12187.538
    },
    {
      "epoch": 0.6928359816859683,
      "grad_norm": 0.8046941449226112,
      "learning_rate": 7.320409583010691e-06,
      "loss": 0.35768985748291016,
      "num_input_tokens_seen": 328117760,
      "step": 10290,
      "train_runtime": 26922.1385,
      "train_tokens_per_second": 12187.656
    },
    {
      "epoch": 0.6931726366819284,
      "grad_norm": 0.7650369082517492,
      "learning_rate": 7.318067151094493e-06,
      "loss": 0.4216940879821777,
      "num_input_tokens_seen": 328278296,
      "step": 10295,
      "train_runtime": 26934.9865,
      "train_tokens_per_second": 12187.802
    },
    {
      "epoch": 0.6935092916778886,
      "grad_norm": 0.6536873829927895,
      "learning_rate": 7.3157240709382046e-06,
      "loss": 0.3889284372329712,
      "num_input_tokens_seen": 328440520,
      "step": 10300,
      "train_runtime": 26947.5978,
      "train_tokens_per_second": 12188.119
    },
    {
      "epoch": 0.6938459466738487,
      "grad_norm": 0.7873609957834201,
      "learning_rate": 7.313380343197066e-06,
      "loss": 0.37988121509552003,
      "num_input_tokens_seen": 328604048,
      "step": 10305,
      "train_runtime": 26960.6332,
      "train_tokens_per_second": 12188.291
    },
    {
      "epoch": 0.6941826016698088,
      "grad_norm": 0.7913397680773571,
      "learning_rate": 7.311035968526492e-06,
      "loss": 0.38131234645843504,
      "num_input_tokens_seen": 328765712,
      "step": 10310,
      "train_runtime": 26973.7833,
      "train_tokens_per_second": 12188.343
    },
    {
      "epoch": 0.694519256665769,
      "grad_norm": 0.7466596495806964,
      "learning_rate": 7.308690947582078e-06,
      "loss": 0.3827646017074585,
      "num_input_tokens_seen": 328923712,
      "step": 10315,
      "train_runtime": 26987.0572,
      "train_tokens_per_second": 12188.202
    },
    {
      "epoch": 0.6948559116617291,
      "grad_norm": 0.8085663945078247,
      "learning_rate": 7.306345281019602e-06,
      "loss": 0.3687703847885132,
      "num_input_tokens_seen": 329081360,
      "step": 10320,
      "train_runtime": 26999.979,
      "train_tokens_per_second": 12188.208
    },
    {
      "epoch": 0.6951925666576892,
      "grad_norm": 0.8430397583952443,
      "learning_rate": 7.303998969495023e-06,
      "loss": 0.40967388153076173,
      "num_input_tokens_seen": 329239984,
      "step": 10325,
      "train_runtime": 27013.1902,
      "train_tokens_per_second": 12188.119
    },
    {
      "epoch": 0.6955292216536494,
      "grad_norm": 0.8082925067739092,
      "learning_rate": 7.301652013664479e-06,
      "loss": 0.4006376266479492,
      "num_input_tokens_seen": 329401288,
      "step": 10330,
      "train_runtime": 27025.6495,
      "train_tokens_per_second": 12188.469
    },
    {
      "epoch": 0.6958658766496095,
      "grad_norm": 0.8985917537074096,
      "learning_rate": 7.299304414184288e-06,
      "loss": 0.42286376953125,
      "num_input_tokens_seen": 329564760,
      "step": 10335,
      "train_runtime": 27038.6719,
      "train_tokens_per_second": 12188.645
    },
    {
      "epoch": 0.6962025316455697,
      "grad_norm": 0.7489344308739019,
      "learning_rate": 7.296956171710951e-06,
      "loss": 0.38113889694213865,
      "num_input_tokens_seen": 329728600,
      "step": 10340,
      "train_runtime": 27051.1177,
      "train_tokens_per_second": 12189.093
    },
    {
      "epoch": 0.6965391866415298,
      "grad_norm": 0.8105629546202905,
      "learning_rate": 7.294607286901146e-06,
      "loss": 0.37472970485687257,
      "num_input_tokens_seen": 329887432,
      "step": 10345,
      "train_runtime": 27063.7834,
      "train_tokens_per_second": 12189.258
    },
    {
      "epoch": 0.69687584163749,
      "grad_norm": 0.9438656688794996,
      "learning_rate": 7.292257760411726e-06,
      "loss": 0.40358500480651854,
      "num_input_tokens_seen": 330042480,
      "step": 10350,
      "train_runtime": 27076.5213,
      "train_tokens_per_second": 12189.25
    },
    {
      "epoch": 0.6972124966334501,
      "grad_norm": 0.839061240682651,
      "learning_rate": 7.289907592899737e-06,
      "loss": 0.3847759962081909,
      "num_input_tokens_seen": 330205672,
      "step": 10355,
      "train_runtime": 27089.4133,
      "train_tokens_per_second": 12189.473
    },
    {
      "epoch": 0.6975491516294102,
      "grad_norm": 0.7292548691484909,
      "learning_rate": 7.287556785022389e-06,
      "loss": 0.3925790786743164,
      "num_input_tokens_seen": 330366824,
      "step": 10360,
      "train_runtime": 27102.3082,
      "train_tokens_per_second": 12189.619
    },
    {
      "epoch": 0.6978858066253704,
      "grad_norm": 0.7659776948021519,
      "learning_rate": 7.285205337437082e-06,
      "loss": 0.38510780334472655,
      "num_input_tokens_seen": 330519152,
      "step": 10365,
      "train_runtime": 27115.2917,
      "train_tokens_per_second": 12189.401
    },
    {
      "epoch": 0.6982224616213305,
      "grad_norm": 0.8398933019021609,
      "learning_rate": 7.2828532508013895e-06,
      "loss": 0.40337285995483396,
      "num_input_tokens_seen": 330679376,
      "step": 10370,
      "train_runtime": 27128.5198,
      "train_tokens_per_second": 12189.363
    },
    {
      "epoch": 0.6985591166172906,
      "grad_norm": 0.8750234887398435,
      "learning_rate": 7.2805005257730625e-06,
      "loss": 0.3946542263031006,
      "num_input_tokens_seen": 330835848,
      "step": 10375,
      "train_runtime": 27141.0685,
      "train_tokens_per_second": 12189.492
    },
    {
      "epoch": 0.6988957716132508,
      "grad_norm": 0.6671735235581735,
      "learning_rate": 7.278147163010037e-06,
      "loss": 0.35418856143951416,
      "num_input_tokens_seen": 330998848,
      "step": 10380,
      "train_runtime": 27153.9217,
      "train_tokens_per_second": 12189.725
    },
    {
      "epoch": 0.6992324266092109,
      "grad_norm": 0.8856550948697094,
      "learning_rate": 7.275793163170423e-06,
      "loss": 0.3769704818725586,
      "num_input_tokens_seen": 331162688,
      "step": 10385,
      "train_runtime": 27166.3553,
      "train_tokens_per_second": 12190.177
    },
    {
      "epoch": 0.6995690816051711,
      "grad_norm": 0.891365448607796,
      "learning_rate": 7.273438526912506e-06,
      "loss": 0.4074446678161621,
      "num_input_tokens_seen": 331318824,
      "step": 10390,
      "train_runtime": 27179.0287,
      "train_tokens_per_second": 12190.238
    },
    {
      "epoch": 0.6999057366011312,
      "grad_norm": 0.7815057508543167,
      "learning_rate": 7.271083254894757e-06,
      "loss": 0.3431298017501831,
      "num_input_tokens_seen": 331479824,
      "step": 10395,
      "train_runtime": 27192.6399,
      "train_tokens_per_second": 12190.057
    },
    {
      "epoch": 0.7002423915970913,
      "grad_norm": 0.8483270956901738,
      "learning_rate": 7.268727347775815e-06,
      "loss": 0.3751350402832031,
      "num_input_tokens_seen": 331641648,
      "step": 10400,
      "train_runtime": 27205.831,
      "train_tokens_per_second": 12190.094
    },
    {
      "epoch": 0.7005790465930515,
      "grad_norm": 0.7692567407492608,
      "learning_rate": 7.266370806214506e-06,
      "loss": 0.3447124004364014,
      "num_input_tokens_seen": 331802728,
      "step": 10405,
      "train_runtime": 27218.5525,
      "train_tokens_per_second": 12190.315
    },
    {
      "epoch": 0.7009157015890116,
      "grad_norm": 0.7339097738749953,
      "learning_rate": 7.264013630869827e-06,
      "loss": 0.3879251480102539,
      "num_input_tokens_seen": 331963528,
      "step": 10410,
      "train_runtime": 27231.4414,
      "train_tokens_per_second": 12190.45
    },
    {
      "epoch": 0.7012523565849718,
      "grad_norm": 0.7312168967370632,
      "learning_rate": 7.261655822400957e-06,
      "loss": 0.42895097732543946,
      "num_input_tokens_seen": 332127128,
      "step": 10415,
      "train_runtime": 27243.9665,
      "train_tokens_per_second": 12190.851
    },
    {
      "epoch": 0.7015890115809319,
      "grad_norm": 0.7284365845120744,
      "learning_rate": 7.259297381467248e-06,
      "loss": 0.3983313798904419,
      "num_input_tokens_seen": 332288656,
      "step": 10420,
      "train_runtime": 27257.7505,
      "train_tokens_per_second": 12190.612
    },
    {
      "epoch": 0.701925666576892,
      "grad_norm": 0.722320689357718,
      "learning_rate": 7.2569383087282285e-06,
      "loss": 0.35359694957733157,
      "num_input_tokens_seen": 332443968,
      "step": 10425,
      "train_runtime": 27271.0828,
      "train_tokens_per_second": 12190.347
    },
    {
      "epoch": 0.7022623215728522,
      "grad_norm": 0.7859525334844696,
      "learning_rate": 7.25457860484361e-06,
      "loss": 0.40454750061035155,
      "num_input_tokens_seen": 332601424,
      "step": 10430,
      "train_runtime": 27283.8874,
      "train_tokens_per_second": 12190.397
    },
    {
      "epoch": 0.7025989765688123,
      "grad_norm": 0.8294035043786365,
      "learning_rate": 7.252218270473274e-06,
      "loss": 0.38651447296142577,
      "num_input_tokens_seen": 332757128,
      "step": 10435,
      "train_runtime": 27297.1978,
      "train_tokens_per_second": 12190.157
    },
    {
      "epoch": 0.7029356315647725,
      "grad_norm": 0.8340251018294074,
      "learning_rate": 7.249857306277279e-06,
      "loss": 0.3877000570297241,
      "num_input_tokens_seen": 332914056,
      "step": 10440,
      "train_runtime": 27310.547,
      "train_tokens_per_second": 12189.945
    },
    {
      "epoch": 0.7032722865607326,
      "grad_norm": 0.961008413138041,
      "learning_rate": 7.247495712915864e-06,
      "loss": 0.3554750919342041,
      "num_input_tokens_seen": 333070728,
      "step": 10445,
      "train_runtime": 27323.5993,
      "train_tokens_per_second": 12189.856
    },
    {
      "epoch": 0.7036089415566927,
      "grad_norm": 0.7616112052529687,
      "learning_rate": 7.245133491049439e-06,
      "loss": 0.3589456558227539,
      "num_input_tokens_seen": 333229928,
      "step": 10450,
      "train_runtime": 27337.4177,
      "train_tokens_per_second": 12189.517
    },
    {
      "epoch": 0.7039455965526529,
      "grad_norm": 0.7626312922651117,
      "learning_rate": 7.242770641338592e-06,
      "loss": 0.41412696838378904,
      "num_input_tokens_seen": 333387496,
      "step": 10455,
      "train_runtime": 27350.139,
      "train_tokens_per_second": 12189.609
    },
    {
      "epoch": 0.704282251548613,
      "grad_norm": 0.7476257803163171,
      "learning_rate": 7.240407164444088e-06,
      "loss": 0.37882261276245116,
      "num_input_tokens_seen": 333544704,
      "step": 10460,
      "train_runtime": 27363.2394,
      "train_tokens_per_second": 12189.518
    },
    {
      "epoch": 0.7046189065445732,
      "grad_norm": 0.7807818337013487,
      "learning_rate": 7.238043061026862e-06,
      "loss": 0.34720165729522706,
      "num_input_tokens_seen": 333705176,
      "step": 10465,
      "train_runtime": 27376.9093,
      "train_tokens_per_second": 12189.293
    },
    {
      "epoch": 0.7049555615405333,
      "grad_norm": 0.7379065701950852,
      "learning_rate": 7.235678331748034e-06,
      "loss": 0.3561399459838867,
      "num_input_tokens_seen": 333859696,
      "step": 10470,
      "train_runtime": 27390.9338,
      "train_tokens_per_second": 12188.693
    },
    {
      "epoch": 0.7052922165364934,
      "grad_norm": 0.8222849846094239,
      "learning_rate": 7.233312977268887e-06,
      "loss": 0.41248311996459963,
      "num_input_tokens_seen": 334018144,
      "step": 10475,
      "train_runtime": 27403.7765,
      "train_tokens_per_second": 12188.763
    },
    {
      "epoch": 0.7056288715324536,
      "grad_norm": 0.7623273179016814,
      "learning_rate": 7.230946998250889e-06,
      "loss": 0.3755314350128174,
      "num_input_tokens_seen": 334174808,
      "step": 10480,
      "train_runtime": 27416.9027,
      "train_tokens_per_second": 12188.642
    },
    {
      "epoch": 0.7059655265284137,
      "grad_norm": 0.7248461008403937,
      "learning_rate": 7.228580395355676e-06,
      "loss": 0.35682685375213624,
      "num_input_tokens_seen": 334332776,
      "step": 10485,
      "train_runtime": 27430.0016,
      "train_tokens_per_second": 12188.58
    },
    {
      "epoch": 0.7063021815243739,
      "grad_norm": 0.8243671727037906,
      "learning_rate": 7.226213169245062e-06,
      "loss": 0.38257484436035155,
      "num_input_tokens_seen": 334494808,
      "step": 10490,
      "train_runtime": 27443.2141,
      "train_tokens_per_second": 12188.616
    },
    {
      "epoch": 0.706638836520334,
      "grad_norm": 0.7423819739459254,
      "learning_rate": 7.223845320581035e-06,
      "loss": 0.36231935024261475,
      "num_input_tokens_seen": 334654072,
      "step": 10495,
      "train_runtime": 27456.2662,
      "train_tokens_per_second": 12188.623
    },
    {
      "epoch": 0.7069754915162941,
      "grad_norm": 0.6987298915843461,
      "learning_rate": 7.221476850025754e-06,
      "loss": 0.3746235132217407,
      "num_input_tokens_seen": 334816536,
      "step": 10500,
      "train_runtime": 27468.9372,
      "train_tokens_per_second": 12188.915
    },
    {
      "epoch": 0.7073121465122543,
      "grad_norm": 0.7644208058141404,
      "learning_rate": 7.219107758241557e-06,
      "loss": 0.3961141347885132,
      "num_input_tokens_seen": 334975024,
      "step": 10505,
      "train_runtime": 27481.5614,
      "train_tokens_per_second": 12189.083
    },
    {
      "epoch": 0.7076488015082144,
      "grad_norm": 0.7491706497637073,
      "learning_rate": 7.2167380458909505e-06,
      "loss": 0.37524521350860596,
      "num_input_tokens_seen": 335138376,
      "step": 10510,
      "train_runtime": 27495.1856,
      "train_tokens_per_second": 12188.984
    },
    {
      "epoch": 0.7079854565041745,
      "grad_norm": 0.7311996786775199,
      "learning_rate": 7.214367713636619e-06,
      "loss": 0.39238715171813965,
      "num_input_tokens_seen": 335297904,
      "step": 10515,
      "train_runtime": 27507.6351,
      "train_tokens_per_second": 12189.267
    },
    {
      "epoch": 0.7083221115001347,
      "grad_norm": 0.7414040054350306,
      "learning_rate": 7.211996762141418e-06,
      "loss": 0.3972428321838379,
      "num_input_tokens_seen": 335456384,
      "step": 10520,
      "train_runtime": 27520.5384,
      "train_tokens_per_second": 12189.31
    },
    {
      "epoch": 0.7086587664960948,
      "grad_norm": 0.8165162989857522,
      "learning_rate": 7.209625192068374e-06,
      "loss": 0.40686378479003904,
      "num_input_tokens_seen": 335614784,
      "step": 10525,
      "train_runtime": 27533.0982,
      "train_tokens_per_second": 12189.503
    },
    {
      "epoch": 0.708995421492055,
      "grad_norm": 0.6662750370338081,
      "learning_rate": 7.207253004080692e-06,
      "loss": 0.3693063735961914,
      "num_input_tokens_seen": 335778624,
      "step": 10530,
      "train_runtime": 27545.5284,
      "train_tokens_per_second": 12189.95
    },
    {
      "epoch": 0.7093320764880151,
      "grad_norm": 0.8268750418106456,
      "learning_rate": 7.204880198841746e-06,
      "loss": 0.35259528160095216,
      "num_input_tokens_seen": 335932240,
      "step": 10535,
      "train_runtime": 27558.4994,
      "train_tokens_per_second": 12189.787
    },
    {
      "epoch": 0.7096687314839752,
      "grad_norm": 0.8419107529379701,
      "learning_rate": 7.2025067770150835e-06,
      "loss": 0.3927944183349609,
      "num_input_tokens_seen": 336092408,
      "step": 10540,
      "train_runtime": 27570.9648,
      "train_tokens_per_second": 12190.085
    },
    {
      "epoch": 0.7100053864799354,
      "grad_norm": 0.7818842094737565,
      "learning_rate": 7.2001327392644235e-06,
      "loss": 0.36365313529968263,
      "num_input_tokens_seen": 336254408,
      "step": 10545,
      "train_runtime": 27583.8016,
      "train_tokens_per_second": 12190.285
    },
    {
      "epoch": 0.7103420414758955,
      "grad_norm": 0.8023061722783627,
      "learning_rate": 7.197758086253659e-06,
      "loss": 0.4142625331878662,
      "num_input_tokens_seen": 336415464,
      "step": 10550,
      "train_runtime": 27597.221,
      "train_tokens_per_second": 12190.193
    },
    {
      "epoch": 0.7106786964718557,
      "grad_norm": 0.7250370208647923,
      "learning_rate": 7.195382818646854e-06,
      "loss": 0.3925436496734619,
      "num_input_tokens_seen": 336573176,
      "step": 10555,
      "train_runtime": 27611.0996,
      "train_tokens_per_second": 12189.778
    },
    {
      "epoch": 0.7110153514678158,
      "grad_norm": 0.7851252446584273,
      "learning_rate": 7.1930069371082445e-06,
      "loss": 0.3854833602905273,
      "num_input_tokens_seen": 336733304,
      "step": 10560,
      "train_runtime": 27624.7276,
      "train_tokens_per_second": 12189.561
    },
    {
      "epoch": 0.7113520064637759,
      "grad_norm": 0.7249592548132505,
      "learning_rate": 7.190630442302238e-06,
      "loss": 0.3884006977081299,
      "num_input_tokens_seen": 336893416,
      "step": 10565,
      "train_runtime": 27637.1474,
      "train_tokens_per_second": 12189.877
    },
    {
      "epoch": 0.7116886614597361,
      "grad_norm": 0.7186907426363982,
      "learning_rate": 7.188253334893414e-06,
      "loss": 0.40602989196777345,
      "num_input_tokens_seen": 337054360,
      "step": 10570,
      "train_runtime": 27650.6343,
      "train_tokens_per_second": 12189.751
    },
    {
      "epoch": 0.7120253164556962,
      "grad_norm": 0.9383929090444015,
      "learning_rate": 7.18587561554652e-06,
      "loss": 0.3929422378540039,
      "num_input_tokens_seen": 337210200,
      "step": 10575,
      "train_runtime": 27663.9117,
      "train_tokens_per_second": 12189.534
    },
    {
      "epoch": 0.7123619714516564,
      "grad_norm": 0.715860824674725,
      "learning_rate": 7.183497284926483e-06,
      "loss": 0.38482906818389895,
      "num_input_tokens_seen": 337373528,
      "step": 10580,
      "train_runtime": 27676.3714,
      "train_tokens_per_second": 12189.948
    },
    {
      "epoch": 0.7126986264476165,
      "grad_norm": 0.781166796455755,
      "learning_rate": 7.181118343698392e-06,
      "loss": 0.35902695655822753,
      "num_input_tokens_seen": 337528656,
      "step": 10585,
      "train_runtime": 27689.3314,
      "train_tokens_per_second": 12189.845
    },
    {
      "epoch": 0.7130352814435766,
      "grad_norm": 0.7794364190082421,
      "learning_rate": 7.178738792527512e-06,
      "loss": 0.4160478591918945,
      "num_input_tokens_seen": 337688792,
      "step": 10590,
      "train_runtime": 27702.5465,
      "train_tokens_per_second": 12189.81
    },
    {
      "epoch": 0.7133719364395368,
      "grad_norm": 0.9078124349340013,
      "learning_rate": 7.176358632079277e-06,
      "loss": 0.3646445035934448,
      "num_input_tokens_seen": 337847408,
      "step": 10595,
      "train_runtime": 27716.2359,
      "train_tokens_per_second": 12189.513
    },
    {
      "epoch": 0.7137085914354969,
      "grad_norm": 0.8152601091441659,
      "learning_rate": 7.173977863019288e-06,
      "loss": 0.3717693567276001,
      "num_input_tokens_seen": 338011136,
      "step": 10600,
      "train_runtime": 27728.703,
      "train_tokens_per_second": 12189.937
    },
    {
      "epoch": 0.7140452464314571,
      "grad_norm": 0.7325800560107413,
      "learning_rate": 7.171596486013323e-06,
      "loss": 0.3579261064529419,
      "num_input_tokens_seen": 338171184,
      "step": 10605,
      "train_runtime": 27741.1227,
      "train_tokens_per_second": 12190.249
    },
    {
      "epoch": 0.7143819014274172,
      "grad_norm": 0.8270114172418782,
      "learning_rate": 7.169214501727326e-06,
      "loss": 0.39139370918273925,
      "num_input_tokens_seen": 338324280,
      "step": 10610,
      "train_runtime": 27754.8468,
      "train_tokens_per_second": 12189.737
    },
    {
      "epoch": 0.7147185564233773,
      "grad_norm": 0.7587642851491243,
      "learning_rate": 7.16683191082741e-06,
      "loss": 0.369968843460083,
      "num_input_tokens_seen": 338482800,
      "step": 10615,
      "train_runtime": 27768.5105,
      "train_tokens_per_second": 12189.447
    },
    {
      "epoch": 0.7150552114193375,
      "grad_norm": 0.868044388753589,
      "learning_rate": 7.16444871397986e-06,
      "loss": 0.367620587348938,
      "num_input_tokens_seen": 338636096,
      "step": 10620,
      "train_runtime": 27781.2556,
      "train_tokens_per_second": 12189.373
    },
    {
      "epoch": 0.7153918664152976,
      "grad_norm": 0.7781232580543597,
      "learning_rate": 7.162064911851129e-06,
      "loss": 0.3659256935119629,
      "num_input_tokens_seen": 338798592,
      "step": 10625,
      "train_runtime": 27793.8587,
      "train_tokens_per_second": 12189.693
    },
    {
      "epoch": 0.7157285214112578,
      "grad_norm": 0.6606277368973356,
      "learning_rate": 7.15968050510784e-06,
      "loss": 0.395468544960022,
      "num_input_tokens_seen": 338957728,
      "step": 10630,
      "train_runtime": 27807.0685,
      "train_tokens_per_second": 12189.625
    },
    {
      "epoch": 0.7160651764072179,
      "grad_norm": 0.9534224388965269,
      "learning_rate": 7.1572954944167834e-06,
      "loss": 0.388535737991333,
      "num_input_tokens_seen": 339115264,
      "step": 10635,
      "train_runtime": 27819.9596,
      "train_tokens_per_second": 12189.639
    },
    {
      "epoch": 0.716401831403178,
      "grad_norm": 0.7964451095287164,
      "learning_rate": 7.154909880444922e-06,
      "loss": 0.3943023681640625,
      "num_input_tokens_seen": 339275584,
      "step": 10640,
      "train_runtime": 27832.5999,
      "train_tokens_per_second": 12189.863
    },
    {
      "epoch": 0.7167384863991382,
      "grad_norm": 0.7925800259984985,
      "learning_rate": 7.152523663859384e-06,
      "loss": 0.37608633041381834,
      "num_input_tokens_seen": 339436504,
      "step": 10645,
      "train_runtime": 27845.7987,
      "train_tokens_per_second": 12189.864
    },
    {
      "epoch": 0.7170751413950983,
      "grad_norm": 0.7705052861204793,
      "learning_rate": 7.150136845327466e-06,
      "loss": 0.37072129249572755,
      "num_input_tokens_seen": 339596464,
      "step": 10650,
      "train_runtime": 27858.4924,
      "train_tokens_per_second": 12190.052
    },
    {
      "epoch": 0.7174117963910585,
      "grad_norm": 0.7941401271476695,
      "learning_rate": 7.147749425516637e-06,
      "loss": 0.3894063711166382,
      "num_input_tokens_seen": 339758496,
      "step": 10655,
      "train_runtime": 27871.0792,
      "train_tokens_per_second": 12190.36
    },
    {
      "epoch": 0.7177484513870186,
      "grad_norm": 0.784326615970473,
      "learning_rate": 7.145361405094529e-06,
      "loss": 0.36672239303588866,
      "num_input_tokens_seen": 339922168,
      "step": 10660,
      "train_runtime": 27884.1143,
      "train_tokens_per_second": 12190.531
    },
    {
      "epoch": 0.7180851063829787,
      "grad_norm": 0.7198016757248019,
      "learning_rate": 7.142972784728945e-06,
      "loss": 0.3425378561019897,
      "num_input_tokens_seen": 340083264,
      "step": 10665,
      "train_runtime": 27897.5396,
      "train_tokens_per_second": 12190.439
    },
    {
      "epoch": 0.7184217613789389,
      "grad_norm": 0.8364815956422419,
      "learning_rate": 7.140583565087856e-06,
      "loss": 0.3542825698852539,
      "num_input_tokens_seen": 340240728,
      "step": 10670,
      "train_runtime": 27909.98,
      "train_tokens_per_second": 12190.648
    },
    {
      "epoch": 0.718758416374899,
      "grad_norm": 0.7674437890454157,
      "learning_rate": 7.138193746839399e-06,
      "loss": 0.4162745475769043,
      "num_input_tokens_seen": 340402880,
      "step": 10675,
      "train_runtime": 27923.2493,
      "train_tokens_per_second": 12190.662
    },
    {
      "epoch": 0.7190950713708592,
      "grad_norm": 0.7718282668105777,
      "learning_rate": 7.135803330651878e-06,
      "loss": 0.39674816131591795,
      "num_input_tokens_seen": 340565144,
      "step": 10680,
      "train_runtime": 27935.8599,
      "train_tokens_per_second": 12190.967
    },
    {
      "epoch": 0.7194317263668193,
      "grad_norm": 0.7732556536928907,
      "learning_rate": 7.133412317193766e-06,
      "loss": 0.38421599864959716,
      "num_input_tokens_seen": 340727736,
      "step": 10685,
      "train_runtime": 27948.5334,
      "train_tokens_per_second": 12191.256
    },
    {
      "epoch": 0.7197683813627794,
      "grad_norm": 0.899091902994993,
      "learning_rate": 7.131020707133703e-06,
      "loss": 0.38947148323059083,
      "num_input_tokens_seen": 340879800,
      "step": 10690,
      "train_runtime": 27961.6914,
      "train_tokens_per_second": 12190.958
    },
    {
      "epoch": 0.7201050363587396,
      "grad_norm": 0.765227904803758,
      "learning_rate": 7.128628501140495e-06,
      "loss": 0.40358428955078124,
      "num_input_tokens_seen": 341036088,
      "step": 10695,
      "train_runtime": 27974.4425,
      "train_tokens_per_second": 12190.988
    },
    {
      "epoch": 0.7204416913546997,
      "grad_norm": 0.8612854137590272,
      "learning_rate": 7.126235699883115e-06,
      "loss": 0.35994644165039064,
      "num_input_tokens_seen": 341191928,
      "step": 10700,
      "train_runtime": 27987.3153,
      "train_tokens_per_second": 12190.949
    },
    {
      "epoch": 0.7207783463506598,
      "grad_norm": 0.7557529933064636,
      "learning_rate": 7.1238423040306995e-06,
      "loss": 0.3696707010269165,
      "num_input_tokens_seen": 341354928,
      "step": 10705,
      "train_runtime": 28002.1956,
      "train_tokens_per_second": 12190.292
    },
    {
      "epoch": 0.72111500134662,
      "grad_norm": 0.7683709245482409,
      "learning_rate": 7.121448314252557e-06,
      "loss": 0.3456122636795044,
      "num_input_tokens_seen": 341518768,
      "step": 10710,
      "train_runtime": 28014.6162,
      "train_tokens_per_second": 12190.735
    },
    {
      "epoch": 0.7214516563425801,
      "grad_norm": 0.7525902569740035,
      "learning_rate": 7.1190537312181586e-06,
      "loss": 0.3708991050720215,
      "num_input_tokens_seen": 341674408,
      "step": 10715,
      "train_runtime": 28027.2339,
      "train_tokens_per_second": 12190.8
    },
    {
      "epoch": 0.7217883113385403,
      "grad_norm": 0.7682714708553864,
      "learning_rate": 7.116658555597141e-06,
      "loss": 0.36903038024902346,
      "num_input_tokens_seen": 341823976,
      "step": 10720,
      "train_runtime": 28041.0515,
      "train_tokens_per_second": 12190.127
    },
    {
      "epoch": 0.7221249663345004,
      "grad_norm": 0.7379297949200991,
      "learning_rate": 7.114262788059308e-06,
      "loss": 0.36638965606689455,
      "num_input_tokens_seen": 341981600,
      "step": 10725,
      "train_runtime": 28054.2014,
      "train_tokens_per_second": 12190.032
    },
    {
      "epoch": 0.7224616213304605,
      "grad_norm": 0.7576346068327312,
      "learning_rate": 7.111866429274627e-06,
      "loss": 0.37665524482727053,
      "num_input_tokens_seen": 342139872,
      "step": 10730,
      "train_runtime": 28068.279,
      "train_tokens_per_second": 12189.556
    },
    {
      "epoch": 0.7227982763264207,
      "grad_norm": 0.8864285703330764,
      "learning_rate": 7.1094694799132356e-06,
      "loss": 0.4015091896057129,
      "num_input_tokens_seen": 342303024,
      "step": 10735,
      "train_runtime": 28081.1618,
      "train_tokens_per_second": 12189.774
    },
    {
      "epoch": 0.7231349313223808,
      "grad_norm": 0.7276423257551156,
      "learning_rate": 7.10707194064543e-06,
      "loss": 0.3598616361618042,
      "num_input_tokens_seen": 342464544,
      "step": 10740,
      "train_runtime": 28095.3885,
      "train_tokens_per_second": 12189.351
    },
    {
      "epoch": 0.723471586318341,
      "grad_norm": 0.8087085237045454,
      "learning_rate": 7.104673812141676e-06,
      "loss": 0.4158106803894043,
      "num_input_tokens_seen": 342624888,
      "step": 10745,
      "train_runtime": 28108.0365,
      "train_tokens_per_second": 12189.57
    },
    {
      "epoch": 0.7238082413143011,
      "grad_norm": 0.872629832568239,
      "learning_rate": 7.102275095072601e-06,
      "loss": 0.39856777191162107,
      "num_input_tokens_seen": 342787792,
      "step": 10750,
      "train_runtime": 28120.7825,
      "train_tokens_per_second": 12189.838
    },
    {
      "epoch": 0.7241448963102612,
      "grad_norm": 0.7639045859682624,
      "learning_rate": 7.099875790108998e-06,
      "loss": 0.37062621116638184,
      "num_input_tokens_seen": 342949784,
      "step": 10755,
      "train_runtime": 28133.3539,
      "train_tokens_per_second": 12190.149
    },
    {
      "epoch": 0.7244815513062214,
      "grad_norm": 0.6964740220400891,
      "learning_rate": 7.097475897921827e-06,
      "loss": 0.37868902683258054,
      "num_input_tokens_seen": 343112440,
      "step": 10760,
      "train_runtime": 28146.6934,
      "train_tokens_per_second": 12190.151
    },
    {
      "epoch": 0.7248182063021815,
      "grad_norm": 0.7504051012605525,
      "learning_rate": 7.09507541918221e-06,
      "loss": 0.39956738948822024,
      "num_input_tokens_seen": 343269208,
      "step": 10765,
      "train_runtime": 28159.5197,
      "train_tokens_per_second": 12190.166
    },
    {
      "epoch": 0.7251548612981417,
      "grad_norm": 0.6549894176623092,
      "learning_rate": 7.092674354561433e-06,
      "loss": 0.39935975074768065,
      "num_input_tokens_seen": 343428440,
      "step": 10770,
      "train_runtime": 28172.2195,
      "train_tokens_per_second": 12190.322
    },
    {
      "epoch": 0.7254915162941018,
      "grad_norm": 1.254793393903918,
      "learning_rate": 7.090272704730945e-06,
      "loss": 0.35133283138275145,
      "num_input_tokens_seen": 343583144,
      "step": 10775,
      "train_runtime": 28185.3377,
      "train_tokens_per_second": 12190.138
    },
    {
      "epoch": 0.7258281712900619,
      "grad_norm": 0.7244088678840466,
      "learning_rate": 7.08787047036236e-06,
      "loss": 0.3733917951583862,
      "num_input_tokens_seen": 343746984,
      "step": 10780,
      "train_runtime": 28197.7535,
      "train_tokens_per_second": 12190.58
    },
    {
      "epoch": 0.7261648262860221,
      "grad_norm": 0.805215070944072,
      "learning_rate": 7.085467652127457e-06,
      "loss": 0.4085439682006836,
      "num_input_tokens_seen": 343908288,
      "step": 10785,
      "train_runtime": 28210.7366,
      "train_tokens_per_second": 12190.688
    },
    {
      "epoch": 0.7265014812819822,
      "grad_norm": 0.7429186764397921,
      "learning_rate": 7.083064250698175e-06,
      "loss": 0.3563601016998291,
      "num_input_tokens_seen": 344068456,
      "step": 10790,
      "train_runtime": 28223.1507,
      "train_tokens_per_second": 12191.001
    },
    {
      "epoch": 0.7268381362779424,
      "grad_norm": 0.7367035083050605,
      "learning_rate": 7.0806602667466175e-06,
      "loss": 0.386067533493042,
      "num_input_tokens_seen": 344231032,
      "step": 10795,
      "train_runtime": 28235.813,
      "train_tokens_per_second": 12191.292
    },
    {
      "epoch": 0.7271747912739025,
      "grad_norm": 0.7573296908801356,
      "learning_rate": 7.078255700945051e-06,
      "loss": 0.3892808437347412,
      "num_input_tokens_seen": 344386016,
      "step": 10800,
      "train_runtime": 28249.0749,
      "train_tokens_per_second": 12191.055
    },
    {
      "epoch": 0.7275114462698626,
      "grad_norm": 0.8938068023827014,
      "learning_rate": 7.075850553965904e-06,
      "loss": 0.3707104682922363,
      "num_input_tokens_seen": 344539736,
      "step": 10805,
      "train_runtime": 28262.1548,
      "train_tokens_per_second": 12190.852
    },
    {
      "epoch": 0.7278481012658228,
      "grad_norm": 0.9620156937207518,
      "learning_rate": 7.07344482648177e-06,
      "loss": 0.40392327308654785,
      "num_input_tokens_seen": 344697400,
      "step": 10810,
      "train_runtime": 28275.0138,
      "train_tokens_per_second": 12190.884
    },
    {
      "epoch": 0.7281847562617829,
      "grad_norm": 0.8257473123283324,
      "learning_rate": 7.071038519165402e-06,
      "loss": 0.3887991428375244,
      "num_input_tokens_seen": 344855536,
      "step": 10815,
      "train_runtime": 28287.9078,
      "train_tokens_per_second": 12190.917
    },
    {
      "epoch": 0.7285214112577431,
      "grad_norm": 0.8079518972191025,
      "learning_rate": 7.068631632689716e-06,
      "loss": 0.3810680627822876,
      "num_input_tokens_seen": 345014864,
      "step": 10820,
      "train_runtime": 28300.4853,
      "train_tokens_per_second": 12191.129
    },
    {
      "epoch": 0.7288580662537032,
      "grad_norm": 0.860753180218142,
      "learning_rate": 7.066224167727791e-06,
      "loss": 0.4030139923095703,
      "num_input_tokens_seen": 345175544,
      "step": 10825,
      "train_runtime": 28314.1332,
      "train_tokens_per_second": 12190.927
    },
    {
      "epoch": 0.7291947212496633,
      "grad_norm": 0.7666125853117247,
      "learning_rate": 7.063816124952865e-06,
      "loss": 0.36918230056762696,
      "num_input_tokens_seen": 345333976,
      "step": 10830,
      "train_runtime": 28327.0894,
      "train_tokens_per_second": 12190.945
    },
    {
      "epoch": 0.7295313762456235,
      "grad_norm": 0.7176807913897995,
      "learning_rate": 7.061407505038341e-06,
      "loss": 0.3628925561904907,
      "num_input_tokens_seen": 345495176,
      "step": 10835,
      "train_runtime": 28340.2438,
      "train_tokens_per_second": 12190.974
    },
    {
      "epoch": 0.7298680312415836,
      "grad_norm": 1.1746539208753348,
      "learning_rate": 7.058998308657782e-06,
      "loss": 0.3935364007949829,
      "num_input_tokens_seen": 345658656,
      "step": 10840,
      "train_runtime": 28353.3065,
      "train_tokens_per_second": 12191.123
    },
    {
      "epoch": 0.7302046862375438,
      "grad_norm": 0.7159174579985556,
      "learning_rate": 7.056588536484912e-06,
      "loss": 0.3441486835479736,
      "num_input_tokens_seen": 345809480,
      "step": 10845,
      "train_runtime": 28366.8538,
      "train_tokens_per_second": 12190.618
    },
    {
      "epoch": 0.7305413412335039,
      "grad_norm": 0.6628617352153386,
      "learning_rate": 7.0541781891936145e-06,
      "loss": 0.3633133888244629,
      "num_input_tokens_seen": 345968544,
      "step": 10850,
      "train_runtime": 28379.8466,
      "train_tokens_per_second": 12190.642
    },
    {
      "epoch": 0.730877996229464,
      "grad_norm": 0.8035015103488539,
      "learning_rate": 7.0517672674579364e-06,
      "loss": 0.37425904273986815,
      "num_input_tokens_seen": 346125752,
      "step": 10855,
      "train_runtime": 28392.5513,
      "train_tokens_per_second": 12190.724
    },
    {
      "epoch": 0.7312146512254242,
      "grad_norm": 0.8337969646990308,
      "learning_rate": 7.049355771952087e-06,
      "loss": 0.37378602027893065,
      "num_input_tokens_seen": 346281088,
      "step": 10860,
      "train_runtime": 28405.8732,
      "train_tokens_per_second": 12190.475
    },
    {
      "epoch": 0.7315513062213843,
      "grad_norm": 0.7659756718702523,
      "learning_rate": 7.046943703350429e-06,
      "loss": 0.381595253944397,
      "num_input_tokens_seen": 346444904,
      "step": 10865,
      "train_runtime": 28418.9565,
      "train_tokens_per_second": 12190.627
    },
    {
      "epoch": 0.7318879612173445,
      "grad_norm": 0.7463020246616815,
      "learning_rate": 7.044531062327492e-06,
      "loss": 0.37255134582519533,
      "num_input_tokens_seen": 346604232,
      "step": 10870,
      "train_runtime": 28432.0746,
      "train_tokens_per_second": 12190.606
    },
    {
      "epoch": 0.7322246162133046,
      "grad_norm": 0.7067416530205086,
      "learning_rate": 7.042117849557963e-06,
      "loss": 0.3892983436584473,
      "num_input_tokens_seen": 346767424,
      "step": 10875,
      "train_runtime": 28444.9421,
      "train_tokens_per_second": 12190.829
    },
    {
      "epoch": 0.7325612712092647,
      "grad_norm": 0.7801124589065547,
      "learning_rate": 7.0397040657166885e-06,
      "loss": 0.3601548194885254,
      "num_input_tokens_seen": 346926904,
      "step": 10880,
      "train_runtime": 28457.5529,
      "train_tokens_per_second": 12191.031
    },
    {
      "epoch": 0.7328979262052249,
      "grad_norm": 0.7649475493052427,
      "learning_rate": 7.037289711478676e-06,
      "loss": 0.3586939811706543,
      "num_input_tokens_seen": 347082872,
      "step": 10885,
      "train_runtime": 28470.9833,
      "train_tokens_per_second": 12190.758
    },
    {
      "epoch": 0.733234581201185,
      "grad_norm": 0.7907576252881421,
      "learning_rate": 7.034874787519095e-06,
      "loss": 0.3743765354156494,
      "num_input_tokens_seen": 347245600,
      "step": 10890,
      "train_runtime": 28484.0533,
      "train_tokens_per_second": 12190.877
    },
    {
      "epoch": 0.7335712361971451,
      "grad_norm": 0.7252725337230612,
      "learning_rate": 7.0324592945132655e-06,
      "loss": 0.41841630935668944,
      "num_input_tokens_seen": 347405328,
      "step": 10895,
      "train_runtime": 28496.7209,
      "train_tokens_per_second": 12191.063
    },
    {
      "epoch": 0.7339078911931053,
      "grad_norm": 0.7406333516765419,
      "learning_rate": 7.0300432331366764e-06,
      "loss": 0.38999099731445314,
      "num_input_tokens_seen": 347562472,
      "step": 10900,
      "train_runtime": 28510.1453,
      "train_tokens_per_second": 12190.835
    },
    {
      "epoch": 0.7342445461890654,
      "grad_norm": 0.7621404512742588,
      "learning_rate": 7.02762660406497e-06,
      "loss": 0.3798921346664429,
      "num_input_tokens_seen": 347723568,
      "step": 10905,
      "train_runtime": 28524.2497,
      "train_tokens_per_second": 12190.454
    },
    {
      "epoch": 0.7345812011850256,
      "grad_norm": 0.7095460004925469,
      "learning_rate": 7.02520940797395e-06,
      "loss": 0.38338747024536135,
      "num_input_tokens_seen": 347881128,
      "step": 10910,
      "train_runtime": 28537.8705,
      "train_tokens_per_second": 12190.157
    },
    {
      "epoch": 0.7349178561809857,
      "grad_norm": 0.7595574367992566,
      "learning_rate": 7.022791645539576e-06,
      "loss": 0.3737035274505615,
      "num_input_tokens_seen": 348034928,
      "step": 10915,
      "train_runtime": 28551.2091,
      "train_tokens_per_second": 12189.849
    },
    {
      "epoch": 0.7352545111769458,
      "grad_norm": 0.8053033063795544,
      "learning_rate": 7.02037331743797e-06,
      "loss": 0.4115499496459961,
      "num_input_tokens_seen": 348193504,
      "step": 10920,
      "train_runtime": 28563.9096,
      "train_tokens_per_second": 12189.981
    },
    {
      "epoch": 0.735591166172906,
      "grad_norm": 0.7859396208213398,
      "learning_rate": 7.0179544243454076e-06,
      "loss": 0.37235002517700194,
      "num_input_tokens_seen": 348356712,
      "step": 10925,
      "train_runtime": 28576.8047,
      "train_tokens_per_second": 12190.191
    },
    {
      "epoch": 0.7359278211688661,
      "grad_norm": 0.8038765297550386,
      "learning_rate": 7.0155349669383245e-06,
      "loss": 0.38625392913818357,
      "num_input_tokens_seen": 348516192,
      "step": 10930,
      "train_runtime": 28589.8397,
      "train_tokens_per_second": 12190.211
    },
    {
      "epoch": 0.7362644761648263,
      "grad_norm": 0.6995250176492618,
      "learning_rate": 7.013114945893316e-06,
      "loss": 0.35385162830352784,
      "num_input_tokens_seen": 348677232,
      "step": 10935,
      "train_runtime": 28602.2835,
      "train_tokens_per_second": 12190.538
    },
    {
      "epoch": 0.7366011311607864,
      "grad_norm": 0.6820520260462778,
      "learning_rate": 7.010694361887131e-06,
      "loss": 0.36516613960266114,
      "num_input_tokens_seen": 348841072,
      "step": 10940,
      "train_runtime": 28614.7059,
      "train_tokens_per_second": 12190.972
    },
    {
      "epoch": 0.7369377861567465,
      "grad_norm": 0.7272829036236642,
      "learning_rate": 7.008273215596681e-06,
      "loss": 0.4030038833618164,
      "num_input_tokens_seen": 349000096,
      "step": 10945,
      "train_runtime": 28627.3709,
      "train_tokens_per_second": 12191.133
    },
    {
      "epoch": 0.7372744411527067,
      "grad_norm": 0.69924006658489,
      "learning_rate": 7.005851507699029e-06,
      "loss": 0.37035179138183594,
      "num_input_tokens_seen": 349161216,
      "step": 10950,
      "train_runtime": 28640.456,
      "train_tokens_per_second": 12191.189
    },
    {
      "epoch": 0.7376110961486668,
      "grad_norm": 0.6889365743760394,
      "learning_rate": 7.003429238871399e-06,
      "loss": 0.36351757049560546,
      "num_input_tokens_seen": 349325056,
      "step": 10955,
      "train_runtime": 28652.8689,
      "train_tokens_per_second": 12191.626
    },
    {
      "epoch": 0.737947751144627,
      "grad_norm": 0.7599391803072556,
      "learning_rate": 7.001006409791171e-06,
      "loss": 0.40810480117797854,
      "num_input_tokens_seen": 349484024,
      "step": 10960,
      "train_runtime": 28665.5641,
      "train_tokens_per_second": 12191.772
    },
    {
      "epoch": 0.7382844061405871,
      "grad_norm": 0.8499585323454134,
      "learning_rate": 6.998583021135882e-06,
      "loss": 0.36732144355773927,
      "num_input_tokens_seen": 349643960,
      "step": 10965,
      "train_runtime": 28678.3827,
      "train_tokens_per_second": 12191.9
    },
    {
      "epoch": 0.7386210611365472,
      "grad_norm": 0.8167452189482237,
      "learning_rate": 6.996159073583224e-06,
      "loss": 0.38299145698547366,
      "num_input_tokens_seen": 349803848,
      "step": 10970,
      "train_runtime": 28691.5701,
      "train_tokens_per_second": 12191.868
    },
    {
      "epoch": 0.7389577161325074,
      "grad_norm": 0.7454367228465312,
      "learning_rate": 6.993734567811046e-06,
      "loss": 0.3518054008483887,
      "num_input_tokens_seen": 349964880,
      "step": 10975,
      "train_runtime": 28704.8456,
      "train_tokens_per_second": 12191.84
    },
    {
      "epoch": 0.7392943711284675,
      "grad_norm": 0.75989089905664,
      "learning_rate": 6.991309504497352e-06,
      "loss": 0.3895235061645508,
      "num_input_tokens_seen": 350127400,
      "step": 10980,
      "train_runtime": 28718.1315,
      "train_tokens_per_second": 12191.859
    },
    {
      "epoch": 0.7396310261244277,
      "grad_norm": 0.8390880777520918,
      "learning_rate": 6.988883884320307e-06,
      "loss": 0.3668651103973389,
      "num_input_tokens_seen": 350286720,
      "step": 10985,
      "train_runtime": 28732.3812,
      "train_tokens_per_second": 12191.357
    },
    {
      "epoch": 0.7399676811203878,
      "grad_norm": 0.7769868566526152,
      "learning_rate": 6.9864577079582254e-06,
      "loss": 0.3895045280456543,
      "num_input_tokens_seen": 350446680,
      "step": 10990,
      "train_runtime": 28744.8118,
      "train_tokens_per_second": 12191.65
    },
    {
      "epoch": 0.7403043361163479,
      "grad_norm": 0.7598938344773035,
      "learning_rate": 6.984030976089581e-06,
      "loss": 0.3538069486618042,
      "num_input_tokens_seen": 350606392,
      "step": 10995,
      "train_runtime": 28757.7513,
      "train_tokens_per_second": 12191.718
    },
    {
      "epoch": 0.7406409911123081,
      "grad_norm": 0.7684784808729817,
      "learning_rate": 6.981603689392999e-06,
      "loss": 0.375300669670105,
      "num_input_tokens_seen": 350765992,
      "step": 11000,
      "train_runtime": 28770.7447,
      "train_tokens_per_second": 12191.759
    },
    {
      "epoch": 0.7409776461082682,
      "grad_norm": 0.7353534368233693,
      "learning_rate": 6.979175848547263e-06,
      "loss": 0.3612549066543579,
      "num_input_tokens_seen": 350920560,
      "step": 11005,
      "train_runtime": 28783.2396,
      "train_tokens_per_second": 12191.837
    },
    {
      "epoch": 0.7413143011042284,
      "grad_norm": 0.8251660635475988,
      "learning_rate": 6.976747454231316e-06,
      "loss": 0.36198992729187013,
      "num_input_tokens_seen": 351080808,
      "step": 11010,
      "train_runtime": 28796.0067,
      "train_tokens_per_second": 12191.996
    },
    {
      "epoch": 0.7416509561001885,
      "grad_norm": 0.748045451307221,
      "learning_rate": 6.974318507124245e-06,
      "loss": 0.3802135467529297,
      "num_input_tokens_seen": 351239064,
      "step": 11015,
      "train_runtime": 28809.7118,
      "train_tokens_per_second": 12191.69
    },
    {
      "epoch": 0.7419876110961486,
      "grad_norm": 0.8269015310519211,
      "learning_rate": 6.971889007905299e-06,
      "loss": 0.3873478889465332,
      "num_input_tokens_seen": 351392016,
      "step": 11020,
      "train_runtime": 28822.5962,
      "train_tokens_per_second": 12191.546
    },
    {
      "epoch": 0.7423242660921088,
      "grad_norm": 0.7571107197374777,
      "learning_rate": 6.9694589572538795e-06,
      "loss": 0.4120494842529297,
      "num_input_tokens_seen": 351555856,
      "step": 11025,
      "train_runtime": 28835.0106,
      "train_tokens_per_second": 12191.979
    },
    {
      "epoch": 0.7426609210880689,
      "grad_norm": 0.7817612654651281,
      "learning_rate": 6.9670283558495436e-06,
      "loss": 0.38866589069366453,
      "num_input_tokens_seen": 351716488,
      "step": 11030,
      "train_runtime": 28847.8259,
      "train_tokens_per_second": 12192.132
    },
    {
      "epoch": 0.742997576084029,
      "grad_norm": 0.8313633361570423,
      "learning_rate": 6.964597204371999e-06,
      "loss": 0.4034535884857178,
      "num_input_tokens_seen": 351870424,
      "step": 11035,
      "train_runtime": 28860.5483,
      "train_tokens_per_second": 12192.091
    },
    {
      "epoch": 0.7433342310799892,
      "grad_norm": 0.746626410951858,
      "learning_rate": 6.962165503501111e-06,
      "loss": 0.3369437217712402,
      "num_input_tokens_seen": 352031240,
      "step": 11040,
      "train_runtime": 28873.3945,
      "train_tokens_per_second": 12192.236
    },
    {
      "epoch": 0.7436708860759493,
      "grad_norm": 0.7511190441872951,
      "learning_rate": 6.959733253916899e-06,
      "loss": 0.39778661727905273,
      "num_input_tokens_seen": 352195016,
      "step": 11045,
      "train_runtime": 28886.4232,
      "train_tokens_per_second": 12192.407
    },
    {
      "epoch": 0.7440075410719095,
      "grad_norm": 0.8894496072353121,
      "learning_rate": 6.95730045629953e-06,
      "loss": 0.3642216444015503,
      "num_input_tokens_seen": 352352880,
      "step": 11050,
      "train_runtime": 28900.4416,
      "train_tokens_per_second": 12191.955
    },
    {
      "epoch": 0.7443441960678696,
      "grad_norm": 0.8603292352789983,
      "learning_rate": 6.954867111329329e-06,
      "loss": 0.4088578224182129,
      "num_input_tokens_seen": 352514848,
      "step": 11055,
      "train_runtime": 28913.6684,
      "train_tokens_per_second": 12191.979
    },
    {
      "epoch": 0.7446808510638298,
      "grad_norm": 0.7913107669747537,
      "learning_rate": 6.952433219686775e-06,
      "loss": 0.38846783638000487,
      "num_input_tokens_seen": 352677112,
      "step": 11060,
      "train_runtime": 28927.1027,
      "train_tokens_per_second": 12191.927
    },
    {
      "epoch": 0.7450175060597899,
      "grad_norm": 0.7888051251464282,
      "learning_rate": 6.949998782052497e-06,
      "loss": 0.3961481094360352,
      "num_input_tokens_seen": 352837840,
      "step": 11065,
      "train_runtime": 28941.3438,
      "train_tokens_per_second": 12191.481
    },
    {
      "epoch": 0.74535416105575,
      "grad_norm": 0.729534910853266,
      "learning_rate": 6.947563799107277e-06,
      "loss": 0.35516939163208006,
      "num_input_tokens_seen": 352993664,
      "step": 11070,
      "train_runtime": 28954.3841,
      "train_tokens_per_second": 12191.372
    },
    {
      "epoch": 0.7456908160517102,
      "grad_norm": 0.8128142264821825,
      "learning_rate": 6.945128271532052e-06,
      "loss": 0.36089444160461426,
      "num_input_tokens_seen": 353157504,
      "step": 11075,
      "train_runtime": 28966.7826,
      "train_tokens_per_second": 12191.81
    },
    {
      "epoch": 0.7460274710476703,
      "grad_norm": 0.745263020066342,
      "learning_rate": 6.942692200007908e-06,
      "loss": 0.38309080600738527,
      "num_input_tokens_seen": 353315680,
      "step": 11080,
      "train_runtime": 28979.5037,
      "train_tokens_per_second": 12191.916
    },
    {
      "epoch": 0.7463641260436304,
      "grad_norm": 0.7435729424190132,
      "learning_rate": 6.9402555852160845e-06,
      "loss": 0.408322811126709,
      "num_input_tokens_seen": 353478176,
      "step": 11085,
      "train_runtime": 28992.808,
      "train_tokens_per_second": 12191.926
    },
    {
      "epoch": 0.7467007810395906,
      "grad_norm": 0.7042572138842473,
      "learning_rate": 6.937818427837975e-06,
      "loss": 0.38018217086791994,
      "num_input_tokens_seen": 353641384,
      "step": 11090,
      "train_runtime": 29005.6811,
      "train_tokens_per_second": 12192.142
    },
    {
      "epoch": 0.7470374360355507,
      "grad_norm": 0.9306564695444097,
      "learning_rate": 6.93538072855512e-06,
      "loss": 0.4043403625488281,
      "num_input_tokens_seen": 353803808,
      "step": 11095,
      "train_runtime": 29019.0423,
      "train_tokens_per_second": 12192.126
    },
    {
      "epoch": 0.7473740910315109,
      "grad_norm": 0.7416083352080415,
      "learning_rate": 6.932942488049217e-06,
      "loss": 0.33733205795288085,
      "num_input_tokens_seen": 353961024,
      "step": 11100,
      "train_runtime": 29032.4211,
      "train_tokens_per_second": 12191.922
    },
    {
      "epoch": 0.747710746027471,
      "grad_norm": 0.7473002189009285,
      "learning_rate": 6.930503707002108e-06,
      "loss": 0.40807476043701174,
      "num_input_tokens_seen": 354120896,
      "step": 11105,
      "train_runtime": 29045.8035,
      "train_tokens_per_second": 12191.809
    },
    {
      "epoch": 0.7480474010234311,
      "grad_norm": 0.6958595855081899,
      "learning_rate": 6.928064386095796e-06,
      "loss": 0.3731332540512085,
      "num_input_tokens_seen": 354284736,
      "step": 11110,
      "train_runtime": 29058.2111,
      "train_tokens_per_second": 12192.242
    },
    {
      "epoch": 0.7483840560193913,
      "grad_norm": 0.8839227583713383,
      "learning_rate": 6.9256245260124264e-06,
      "loss": 0.38963606357574465,
      "num_input_tokens_seen": 354445576,
      "step": 11115,
      "train_runtime": 29070.6483,
      "train_tokens_per_second": 12192.558
    },
    {
      "epoch": 0.7487207110153514,
      "grad_norm": 0.7435427288059749,
      "learning_rate": 6.923184127434298e-06,
      "loss": 0.3701322555541992,
      "num_input_tokens_seen": 354598736,
      "step": 11120,
      "train_runtime": 29083.7693,
      "train_tokens_per_second": 12192.324
    },
    {
      "epoch": 0.7490573660113116,
      "grad_norm": 0.9811755512839009,
      "learning_rate": 6.920743191043861e-06,
      "loss": 0.34794456958770753,
      "num_input_tokens_seen": 354754808,
      "step": 11125,
      "train_runtime": 29096.4452,
      "train_tokens_per_second": 12192.376
    },
    {
      "epoch": 0.7493940210072717,
      "grad_norm": 0.6417933775269338,
      "learning_rate": 6.918301717523714e-06,
      "loss": 0.3479334354400635,
      "num_input_tokens_seen": 354911328,
      "step": 11130,
      "train_runtime": 29109.0736,
      "train_tokens_per_second": 12192.464
    },
    {
      "epoch": 0.7497306760032318,
      "grad_norm": 0.7741077754094616,
      "learning_rate": 6.915859707556611e-06,
      "loss": 0.3851003170013428,
      "num_input_tokens_seen": 355068488,
      "step": 11135,
      "train_runtime": 29121.5598,
      "train_tokens_per_second": 12192.633
    },
    {
      "epoch": 0.750067330999192,
      "grad_norm": 0.8390171220875373,
      "learning_rate": 6.913417161825449e-06,
      "loss": 0.4130240440368652,
      "num_input_tokens_seen": 355226344,
      "step": 11140,
      "train_runtime": 29134.7528,
      "train_tokens_per_second": 12192.53
    },
    {
      "epoch": 0.7504039859951521,
      "grad_norm": 0.7487968636238541,
      "learning_rate": 6.91097408101328e-06,
      "loss": 0.3791627883911133,
      "num_input_tokens_seen": 355386896,
      "step": 11145,
      "train_runtime": 29147.8169,
      "train_tokens_per_second": 12192.573
    },
    {
      "epoch": 0.7507406409911123,
      "grad_norm": 0.9765637692779789,
      "learning_rate": 6.908530465803302e-06,
      "loss": 0.3772862434387207,
      "num_input_tokens_seen": 355540696,
      "step": 11150,
      "train_runtime": 29161.1843,
      "train_tokens_per_second": 12192.258
    },
    {
      "epoch": 0.7510772959870724,
      "grad_norm": 0.8557504378000204,
      "learning_rate": 6.906086316878865e-06,
      "loss": 0.3812490224838257,
      "num_input_tokens_seen": 355693720,
      "step": 11155,
      "train_runtime": 29174.6216,
      "train_tokens_per_second": 12191.888
    },
    {
      "epoch": 0.7514139509830325,
      "grad_norm": 0.7219876610404651,
      "learning_rate": 6.903641634923468e-06,
      "loss": 0.3538412570953369,
      "num_input_tokens_seen": 355850208,
      "step": 11160,
      "train_runtime": 29187.0605,
      "train_tokens_per_second": 12192.054
    },
    {
      "epoch": 0.7517506059789927,
      "grad_norm": 0.8396885968011758,
      "learning_rate": 6.9011964206207585e-06,
      "loss": 0.4000197410583496,
      "num_input_tokens_seen": 356003032,
      "step": 11165,
      "train_runtime": 29201.3197,
      "train_tokens_per_second": 12191.334
    },
    {
      "epoch": 0.7520872609749528,
      "grad_norm": 0.8139340499458105,
      "learning_rate": 6.898750674654531e-06,
      "loss": 0.37145266532897947,
      "num_input_tokens_seen": 356164080,
      "step": 11170,
      "train_runtime": 29215.0693,
      "train_tokens_per_second": 12191.109
    },
    {
      "epoch": 0.752423915970913,
      "grad_norm": 0.79460109141314,
      "learning_rate": 6.896304397708733e-06,
      "loss": 0.3592658281326294,
      "num_input_tokens_seen": 356321656,
      "step": 11175,
      "train_runtime": 29228.216,
      "train_tokens_per_second": 12191.016
    },
    {
      "epoch": 0.7527605709668731,
      "grad_norm": 0.8669921164210402,
      "learning_rate": 6.893857590467456e-06,
      "loss": 0.3539292573928833,
      "num_input_tokens_seen": 356476320,
      "step": 11180,
      "train_runtime": 29240.8622,
      "train_tokens_per_second": 12191.033
    },
    {
      "epoch": 0.7530972259628332,
      "grad_norm": 0.7919910070074195,
      "learning_rate": 6.891410253614944e-06,
      "loss": 0.35689308643341067,
      "num_input_tokens_seen": 356632400,
      "step": 11185,
      "train_runtime": 29254.0285,
      "train_tokens_per_second": 12190.882
    },
    {
      "epoch": 0.7534338809587934,
      "grad_norm": 0.8004556253149746,
      "learning_rate": 6.888962387835586e-06,
      "loss": 0.3711047887802124,
      "num_input_tokens_seen": 356793824,
      "step": 11190,
      "train_runtime": 29267.1883,
      "train_tokens_per_second": 12190.916
    },
    {
      "epoch": 0.7537705359547535,
      "grad_norm": 0.8134820844966203,
      "learning_rate": 6.886513993813919e-06,
      "loss": 0.40021324157714844,
      "num_input_tokens_seen": 356953920,
      "step": 11195,
      "train_runtime": 29281.1668,
      "train_tokens_per_second": 12190.563
    },
    {
      "epoch": 0.7541071909507137,
      "grad_norm": 0.7642009268562932,
      "learning_rate": 6.88406507223463e-06,
      "loss": 0.37061433792114257,
      "num_input_tokens_seen": 357117760,
      "step": 11200,
      "train_runtime": 29293.5827,
      "train_tokens_per_second": 12190.99
    },
    {
      "epoch": 0.7544438459466738,
      "grad_norm": 0.7471490246561128,
      "learning_rate": 6.881615623782551e-06,
      "loss": 0.36925468444824217,
      "num_input_tokens_seen": 357275832,
      "step": 11205,
      "train_runtime": 29307.9034,
      "train_tokens_per_second": 12190.426
    },
    {
      "epoch": 0.7547805009426339,
      "grad_norm": 0.6774004726417472,
      "learning_rate": 6.879165649142663e-06,
      "loss": 0.3666870594024658,
      "num_input_tokens_seen": 357437568,
      "step": 11210,
      "train_runtime": 29321.4085,
      "train_tokens_per_second": 12190.327
    },
    {
      "epoch": 0.7551171559385941,
      "grad_norm": 0.7872210007616245,
      "learning_rate": 6.876715149000093e-06,
      "loss": 0.37008869647979736,
      "num_input_tokens_seen": 357591616,
      "step": 11215,
      "train_runtime": 29334.3948,
      "train_tokens_per_second": 12190.182
    },
    {
      "epoch": 0.7554538109345542,
      "grad_norm": 0.757046930474841,
      "learning_rate": 6.874264124040117e-06,
      "loss": 0.37499892711639404,
      "num_input_tokens_seen": 357754320,
      "step": 11220,
      "train_runtime": 29347.3324,
      "train_tokens_per_second": 12190.352
    },
    {
      "epoch": 0.7557904659305144,
      "grad_norm": 0.8327627703764695,
      "learning_rate": 6.871812574948154e-06,
      "loss": 0.40956530570983884,
      "num_input_tokens_seen": 357913480,
      "step": 11225,
      "train_runtime": 29360.25,
      "train_tokens_per_second": 12190.41
    },
    {
      "epoch": 0.7561271209264745,
      "grad_norm": 0.8662811620632913,
      "learning_rate": 6.869360502409773e-06,
      "loss": 0.38788011074066164,
      "num_input_tokens_seen": 358074680,
      "step": 11230,
      "train_runtime": 29372.992,
      "train_tokens_per_second": 12190.61
    },
    {
      "epoch": 0.7564637759224346,
      "grad_norm": 0.8094273414786728,
      "learning_rate": 6.866907907110689e-06,
      "loss": 0.36602394580841063,
      "num_input_tokens_seen": 358230048,
      "step": 11235,
      "train_runtime": 29385.6207,
      "train_tokens_per_second": 12190.658
    },
    {
      "epoch": 0.7568004309183949,
      "grad_norm": 0.720484140196403,
      "learning_rate": 6.864454789736763e-06,
      "loss": 0.3442512512207031,
      "num_input_tokens_seen": 358393608,
      "step": 11240,
      "train_runtime": 29398.6233,
      "train_tokens_per_second": 12190.83
    },
    {
      "epoch": 0.757137085914355,
      "grad_norm": 0.6873358735140016,
      "learning_rate": 6.8620011509740005e-06,
      "loss": 0.34670538902282716,
      "num_input_tokens_seen": 358551896,
      "step": 11245,
      "train_runtime": 29411.2089,
      "train_tokens_per_second": 12190.995
    },
    {
      "epoch": 0.7574737409103152,
      "grad_norm": 0.7891565359818309,
      "learning_rate": 6.859546991508552e-06,
      "loss": 0.3740941762924194,
      "num_input_tokens_seen": 358710872,
      "step": 11250,
      "train_runtime": 29423.8088,
      "train_tokens_per_second": 12191.177
    },
    {
      "epoch": 0.7578103959062753,
      "grad_norm": 0.6993226475720027,
      "learning_rate": 6.857092312026716e-06,
      "loss": 0.34750382900238036,
      "num_input_tokens_seen": 358872592,
      "step": 11255,
      "train_runtime": 29436.3283,
      "train_tokens_per_second": 12191.486
    },
    {
      "epoch": 0.7581470509022354,
      "grad_norm": 0.7677525233859016,
      "learning_rate": 6.8546371132149395e-06,
      "loss": 0.394059944152832,
      "num_input_tokens_seen": 359030944,
      "step": 11260,
      "train_runtime": 29448.9164,
      "train_tokens_per_second": 12191.652
    },
    {
      "epoch": 0.7584837058981956,
      "grad_norm": 0.7683725348939942,
      "learning_rate": 6.852181395759807e-06,
      "loss": 0.41542778015136717,
      "num_input_tokens_seen": 359194280,
      "step": 11265,
      "train_runtime": 29461.866,
      "train_tokens_per_second": 12191.837
    },
    {
      "epoch": 0.7588203608941557,
      "grad_norm": 0.6567574430482627,
      "learning_rate": 6.849725160348054e-06,
      "loss": 0.3582466125488281,
      "num_input_tokens_seen": 359352176,
      "step": 11270,
      "train_runtime": 29474.562,
      "train_tokens_per_second": 12191.943
    },
    {
      "epoch": 0.7591570158901159,
      "grad_norm": 0.8630367003820414,
      "learning_rate": 6.847268407666557e-06,
      "loss": 0.34615659713745117,
      "num_input_tokens_seen": 359512888,
      "step": 11275,
      "train_runtime": 29487.5686,
      "train_tokens_per_second": 12192.015
    },
    {
      "epoch": 0.759493670886076,
      "grad_norm": 0.8062156136396837,
      "learning_rate": 6.844811138402341e-06,
      "loss": 0.3790543794631958,
      "num_input_tokens_seen": 359675376,
      "step": 11280,
      "train_runtime": 29500.315,
      "train_tokens_per_second": 12192.255
    },
    {
      "epoch": 0.7598303258820361,
      "grad_norm": 0.7616178646648184,
      "learning_rate": 6.842353353242574e-06,
      "loss": 0.3697885274887085,
      "num_input_tokens_seen": 359832992,
      "step": 11285,
      "train_runtime": 29513.7338,
      "train_tokens_per_second": 12192.052
    },
    {
      "epoch": 0.7601669808779963,
      "grad_norm": 0.8835486212985131,
      "learning_rate": 6.839895052874565e-06,
      "loss": 0.39465861320495604,
      "num_input_tokens_seen": 359995320,
      "step": 11290,
      "train_runtime": 29526.7553,
      "train_tokens_per_second": 12192.173
    },
    {
      "epoch": 0.7605036358739564,
      "grad_norm": 0.7299233138556133,
      "learning_rate": 6.837436237985772e-06,
      "loss": 0.35462312698364257,
      "num_input_tokens_seen": 360157272,
      "step": 11295,
      "train_runtime": 29540.0725,
      "train_tokens_per_second": 12192.159
    },
    {
      "epoch": 0.7608402908699166,
      "grad_norm": 0.8612570451494536,
      "learning_rate": 6.834976909263794e-06,
      "loss": 0.38593153953552245,
      "num_input_tokens_seen": 360317424,
      "step": 11300,
      "train_runtime": 29553.6827,
      "train_tokens_per_second": 12191.964
    },
    {
      "epoch": 0.7611769458658767,
      "grad_norm": 0.7717601359819006,
      "learning_rate": 6.832517067396374e-06,
      "loss": 0.3730529546737671,
      "num_input_tokens_seen": 360480592,
      "step": 11305,
      "train_runtime": 29566.5355,
      "train_tokens_per_second": 12192.182
    },
    {
      "epoch": 0.7615136008618368,
      "grad_norm": 0.8029276193294044,
      "learning_rate": 6.8300567130714e-06,
      "loss": 0.39319963455200196,
      "num_input_tokens_seen": 360641592,
      "step": 11310,
      "train_runtime": 29578.9829,
      "train_tokens_per_second": 12192.495
    },
    {
      "epoch": 0.761850255857797,
      "grad_norm": 0.7422145690110517,
      "learning_rate": 6.8275958469769035e-06,
      "loss": 0.39521007537841796,
      "num_input_tokens_seen": 360802776,
      "step": 11315,
      "train_runtime": 29591.4701,
      "train_tokens_per_second": 12192.797
    },
    {
      "epoch": 0.7621869108537571,
      "grad_norm": 0.8185854436182528,
      "learning_rate": 6.825134469801055e-06,
      "loss": 0.3593320369720459,
      "num_input_tokens_seen": 360965128,
      "step": 11320,
      "train_runtime": 29604.1139,
      "train_tokens_per_second": 12193.073
    },
    {
      "epoch": 0.7625235658497173,
      "grad_norm": 0.8243747513242757,
      "learning_rate": 6.8226725822321715e-06,
      "loss": 0.40591912269592284,
      "num_input_tokens_seen": 361120704,
      "step": 11325,
      "train_runtime": 29617.4144,
      "train_tokens_per_second": 12192.85
    },
    {
      "epoch": 0.7628602208456774,
      "grad_norm": 0.7739255592175426,
      "learning_rate": 6.820210184958712e-06,
      "loss": 0.39017329216003416,
      "num_input_tokens_seen": 361284512,
      "step": 11330,
      "train_runtime": 29629.8738,
      "train_tokens_per_second": 12193.252
    },
    {
      "epoch": 0.7631968758416375,
      "grad_norm": 0.7869095068819479,
      "learning_rate": 6.81774727866928e-06,
      "loss": 0.38749332427978517,
      "num_input_tokens_seen": 361444352,
      "step": 11335,
      "train_runtime": 29643.8844,
      "train_tokens_per_second": 12192.881
    },
    {
      "epoch": 0.7635335308375977,
      "grad_norm": 0.8052285572147169,
      "learning_rate": 6.815283864052617e-06,
      "loss": 0.3830084800720215,
      "num_input_tokens_seen": 361604808,
      "step": 11340,
      "train_runtime": 29657.3765,
      "train_tokens_per_second": 12192.744
    },
    {
      "epoch": 0.7638701858335578,
      "grad_norm": 0.8210171420985842,
      "learning_rate": 6.812819941797611e-06,
      "loss": 0.397029447555542,
      "num_input_tokens_seen": 361757928,
      "step": 11345,
      "train_runtime": 29669.8755,
      "train_tokens_per_second": 12192.769
    },
    {
      "epoch": 0.764206840829518,
      "grad_norm": 0.8581451721551036,
      "learning_rate": 6.810355512593288e-06,
      "loss": 0.38977923393249514,
      "num_input_tokens_seen": 361917728,
      "step": 11350,
      "train_runtime": 29682.6505,
      "train_tokens_per_second": 12192.905
    },
    {
      "epoch": 0.7645434958254781,
      "grad_norm": 0.6712921829905516,
      "learning_rate": 6.807890577128817e-06,
      "loss": 0.3445927619934082,
      "num_input_tokens_seen": 362079696,
      "step": 11355,
      "train_runtime": 29695.7704,
      "train_tokens_per_second": 12192.972
    },
    {
      "epoch": 0.7648801508214382,
      "grad_norm": 0.7863791585585402,
      "learning_rate": 6.805425136093513e-06,
      "loss": 0.37203330993652345,
      "num_input_tokens_seen": 362238440,
      "step": 11360,
      "train_runtime": 29709.2908,
      "train_tokens_per_second": 12192.766
    },
    {
      "epoch": 0.7652168058173984,
      "grad_norm": 0.8479470132816593,
      "learning_rate": 6.802959190176827e-06,
      "loss": 0.38611886501312254,
      "num_input_tokens_seen": 362396280,
      "step": 11365,
      "train_runtime": 29722.627,
      "train_tokens_per_second": 12192.606
    },
    {
      "epoch": 0.7655534608133585,
      "grad_norm": 0.8222171740359264,
      "learning_rate": 6.800492740068352e-06,
      "loss": 0.3653825283050537,
      "num_input_tokens_seen": 362549696,
      "step": 11370,
      "train_runtime": 29736.3359,
      "train_tokens_per_second": 12192.144
    },
    {
      "epoch": 0.7658901158093187,
      "grad_norm": 0.8727787911253201,
      "learning_rate": 6.798025786457825e-06,
      "loss": 0.37914419174194336,
      "num_input_tokens_seen": 362711336,
      "step": 11375,
      "train_runtime": 29748.9516,
      "train_tokens_per_second": 12192.407
    },
    {
      "epoch": 0.7662267708052788,
      "grad_norm": 0.7244506826489351,
      "learning_rate": 6.795558330035118e-06,
      "loss": 0.37247552871704104,
      "num_input_tokens_seen": 362873448,
      "step": 11380,
      "train_runtime": 29761.481,
      "train_tokens_per_second": 12192.721
    },
    {
      "epoch": 0.7665634258012389,
      "grad_norm": 0.6813233772596614,
      "learning_rate": 6.793090371490251e-06,
      "loss": 0.3876584529876709,
      "num_input_tokens_seen": 363033344,
      "step": 11385,
      "train_runtime": 29773.9183,
      "train_tokens_per_second": 12192.999
    },
    {
      "epoch": 0.7669000807971991,
      "grad_norm": 0.8147571353650083,
      "learning_rate": 6.790621911513378e-06,
      "loss": 0.3760810852050781,
      "num_input_tokens_seen": 363192144,
      "step": 11390,
      "train_runtime": 29786.911,
      "train_tokens_per_second": 12193.011
    },
    {
      "epoch": 0.7672367357931592,
      "grad_norm": 0.8211650499021289,
      "learning_rate": 6.788152950794799e-06,
      "loss": 0.39856245517730715,
      "num_input_tokens_seen": 363350632,
      "step": 11395,
      "train_runtime": 29800.0292,
      "train_tokens_per_second": 12192.962
    },
    {
      "epoch": 0.7675733907891193,
      "grad_norm": 0.8689286824664951,
      "learning_rate": 6.7856834900249495e-06,
      "loss": 0.40386271476745605,
      "num_input_tokens_seen": 363511320,
      "step": 11400,
      "train_runtime": 29812.4685,
      "train_tokens_per_second": 12193.265
    },
    {
      "epoch": 0.7679100457850795,
      "grad_norm": 0.7798105528572442,
      "learning_rate": 6.783213529894405e-06,
      "loss": 0.3570334672927856,
      "num_input_tokens_seen": 363673184,
      "step": 11405,
      "train_runtime": 29824.9984,
      "train_tokens_per_second": 12193.569
    },
    {
      "epoch": 0.7682467007810396,
      "grad_norm": 0.7147971811410965,
      "learning_rate": 6.780743071093883e-06,
      "loss": 0.34810585975646974,
      "num_input_tokens_seen": 363834200,
      "step": 11410,
      "train_runtime": 29838.4608,
      "train_tokens_per_second": 12193.464
    },
    {
      "epoch": 0.7685833557769998,
      "grad_norm": 0.8090101958649409,
      "learning_rate": 6.77827211431424e-06,
      "loss": 0.3833613395690918,
      "num_input_tokens_seen": 363993880,
      "step": 11415,
      "train_runtime": 29851.0682,
      "train_tokens_per_second": 12193.663
    },
    {
      "epoch": 0.7689200107729599,
      "grad_norm": 0.6950873673459217,
      "learning_rate": 6.775800660246471e-06,
      "loss": 0.37551772594451904,
      "num_input_tokens_seen": 364146208,
      "step": 11420,
      "train_runtime": 29864.434,
      "train_tokens_per_second": 12193.307
    },
    {
      "epoch": 0.76925666576892,
      "grad_norm": 0.8926310843991966,
      "learning_rate": 6.77332870958171e-06,
      "loss": 0.37343225479125974,
      "num_input_tokens_seen": 364307056,
      "step": 11425,
      "train_runtime": 29876.8796,
      "train_tokens_per_second": 12193.611
    },
    {
      "epoch": 0.7695933207648802,
      "grad_norm": 0.7170998110429361,
      "learning_rate": 6.770856263011229e-06,
      "loss": 0.36758739948272706,
      "num_input_tokens_seen": 364464312,
      "step": 11430,
      "train_runtime": 29890.5294,
      "train_tokens_per_second": 12193.304
    },
    {
      "epoch": 0.7699299757608403,
      "grad_norm": 0.8981508930887381,
      "learning_rate": 6.768383321226442e-06,
      "loss": 0.3995571851730347,
      "num_input_tokens_seen": 364617288,
      "step": 11435,
      "train_runtime": 29903.6194,
      "train_tokens_per_second": 12193.082
    },
    {
      "epoch": 0.7702666307568005,
      "grad_norm": 0.8147507601969611,
      "learning_rate": 6.765909884918898e-06,
      "loss": 0.38014495372772217,
      "num_input_tokens_seen": 364771728,
      "step": 11440,
      "train_runtime": 29916.8318,
      "train_tokens_per_second": 12192.86
    },
    {
      "epoch": 0.7706032857527606,
      "grad_norm": 0.7524886291121373,
      "learning_rate": 6.763435954780286e-06,
      "loss": 0.3776134729385376,
      "num_input_tokens_seen": 364931408,
      "step": 11445,
      "train_runtime": 29929.698,
      "train_tokens_per_second": 12192.953
    },
    {
      "epoch": 0.7709399407487207,
      "grad_norm": 0.8168633699889626,
      "learning_rate": 6.7609615315024325e-06,
      "loss": 0.3896333694458008,
      "num_input_tokens_seen": 365089752,
      "step": 11450,
      "train_runtime": 29942.3106,
      "train_tokens_per_second": 12193.105
    },
    {
      "epoch": 0.7712765957446809,
      "grad_norm": 0.7691934325905628,
      "learning_rate": 6.758486615777301e-06,
      "loss": 0.3641579389572144,
      "num_input_tokens_seen": 365248576,
      "step": 11455,
      "train_runtime": 29955.3533,
      "train_tokens_per_second": 12193.099
    },
    {
      "epoch": 0.771613250740641,
      "grad_norm": 0.8507056742596609,
      "learning_rate": 6.756011208296995e-06,
      "loss": 0.3841691970825195,
      "num_input_tokens_seen": 365405808,
      "step": 11460,
      "train_runtime": 29968.7497,
      "train_tokens_per_second": 12192.895
    },
    {
      "epoch": 0.7719499057366012,
      "grad_norm": 0.9103659200468566,
      "learning_rate": 6.753535309753755e-06,
      "loss": 0.37653412818908694,
      "num_input_tokens_seen": 365561856,
      "step": 11465,
      "train_runtime": 29982.0373,
      "train_tokens_per_second": 12192.696
    },
    {
      "epoch": 0.7722865607325613,
      "grad_norm": 0.8512208053762094,
      "learning_rate": 6.7510589208399565e-06,
      "loss": 0.3696767568588257,
      "num_input_tokens_seen": 365719760,
      "step": 11470,
      "train_runtime": 29994.9306,
      "train_tokens_per_second": 12192.719
    },
    {
      "epoch": 0.7726232157285214,
      "grad_norm": 0.8501120940347665,
      "learning_rate": 6.748582042248115e-06,
      "loss": 0.3655768156051636,
      "num_input_tokens_seen": 365878168,
      "step": 11475,
      "train_runtime": 30007.8885,
      "train_tokens_per_second": 12192.733
    },
    {
      "epoch": 0.7729598707244816,
      "grad_norm": 0.7818660038196905,
      "learning_rate": 6.746104674670882e-06,
      "loss": 0.4025142192840576,
      "num_input_tokens_seen": 366037760,
      "step": 11480,
      "train_runtime": 30020.4375,
      "train_tokens_per_second": 12192.952
    },
    {
      "epoch": 0.7732965257204417,
      "grad_norm": 0.8367259012984112,
      "learning_rate": 6.743626818801042e-06,
      "loss": 0.4009434700012207,
      "num_input_tokens_seen": 366198568,
      "step": 11485,
      "train_runtime": 30033.341,
      "train_tokens_per_second": 12193.068
    },
    {
      "epoch": 0.7736331807164019,
      "grad_norm": 0.785632382930976,
      "learning_rate": 6.741148475331525e-06,
      "loss": 0.3762365818023682,
      "num_input_tokens_seen": 366358576,
      "step": 11490,
      "train_runtime": 30046.6623,
      "train_tokens_per_second": 12192.987
    },
    {
      "epoch": 0.773969835712362,
      "grad_norm": 0.7088082874155206,
      "learning_rate": 6.738669644955388e-06,
      "loss": 0.36430606842041013,
      "num_input_tokens_seen": 366519664,
      "step": 11495,
      "train_runtime": 30060.0103,
      "train_tokens_per_second": 12192.932
    },
    {
      "epoch": 0.7743064907083221,
      "grad_norm": 0.8063067943270414,
      "learning_rate": 6.736190328365828e-06,
      "loss": 0.36579194068908694,
      "num_input_tokens_seen": 366679576,
      "step": 11500,
      "train_runtime": 30072.721,
      "train_tokens_per_second": 12193.096
    },
    {
      "epoch": 0.7746431457042823,
      "grad_norm": 0.8080995968205417,
      "learning_rate": 6.73371052625618e-06,
      "loss": 0.3714073896408081,
      "num_input_tokens_seen": 366841464,
      "step": 11505,
      "train_runtime": 30085.261,
      "train_tokens_per_second": 12193.395
    },
    {
      "epoch": 0.7749798007002424,
      "grad_norm": 0.7137876393206664,
      "learning_rate": 6.731230239319912e-06,
      "loss": 0.37815117835998535,
      "num_input_tokens_seen": 367003472,
      "step": 11510,
      "train_runtime": 30098.4767,
      "train_tokens_per_second": 12193.423
    },
    {
      "epoch": 0.7753164556962026,
      "grad_norm": 0.6908970692232513,
      "learning_rate": 6.728749468250627e-06,
      "loss": 0.39992058277130127,
      "num_input_tokens_seen": 367164312,
      "step": 11515,
      "train_runtime": 30110.9249,
      "train_tokens_per_second": 12193.724
    },
    {
      "epoch": 0.7756531106921627,
      "grad_norm": 0.8408036808852651,
      "learning_rate": 6.726268213742066e-06,
      "loss": 0.37851681709289553,
      "num_input_tokens_seen": 367321704,
      "step": 11520,
      "train_runtime": 30124.3634,
      "train_tokens_per_second": 12193.509
    },
    {
      "epoch": 0.7759897656881228,
      "grad_norm": 0.6754893847148205,
      "learning_rate": 6.723786476488105e-06,
      "loss": 0.3802226781845093,
      "num_input_tokens_seen": 367480496,
      "step": 11525,
      "train_runtime": 30136.9362,
      "train_tokens_per_second": 12193.691
    },
    {
      "epoch": 0.776326420684083,
      "grad_norm": 0.7437825294940577,
      "learning_rate": 6.72130425718275e-06,
      "loss": 0.3518873929977417,
      "num_input_tokens_seen": 367637624,
      "step": 11530,
      "train_runtime": 30150.8119,
      "train_tokens_per_second": 12193.291
    },
    {
      "epoch": 0.7766630756800431,
      "grad_norm": 0.8881777690824689,
      "learning_rate": 6.718821556520151e-06,
      "loss": 0.39986333847045896,
      "num_input_tokens_seen": 367794592,
      "step": 11535,
      "train_runtime": 30164.271,
      "train_tokens_per_second": 12193.054
    },
    {
      "epoch": 0.7769997306760033,
      "grad_norm": 0.7806031710739979,
      "learning_rate": 6.716338375194583e-06,
      "loss": 0.38141722679138185,
      "num_input_tokens_seen": 367955144,
      "step": 11540,
      "train_runtime": 30177.3168,
      "train_tokens_per_second": 12193.103
    },
    {
      "epoch": 0.7773363856719634,
      "grad_norm": 0.7980098500901036,
      "learning_rate": 6.713854713900463e-06,
      "loss": 0.40670599937438967,
      "num_input_tokens_seen": 368115176,
      "step": 11545,
      "train_runtime": 30190.1441,
      "train_tokens_per_second": 12193.224
    },
    {
      "epoch": 0.7776730406679235,
      "grad_norm": 0.8412667780541814,
      "learning_rate": 6.7113705733323365e-06,
      "loss": 0.3950677394866943,
      "num_input_tokens_seen": 368272216,
      "step": 11550,
      "train_runtime": 30203.2624,
      "train_tokens_per_second": 12193.127
    },
    {
      "epoch": 0.7780096956638837,
      "grad_norm": 0.9087359918885015,
      "learning_rate": 6.708885954184885e-06,
      "loss": 0.3833591461181641,
      "num_input_tokens_seen": 368432040,
      "step": 11555,
      "train_runtime": 30215.7407,
      "train_tokens_per_second": 12193.381
    },
    {
      "epoch": 0.7783463506598438,
      "grad_norm": 0.7752373268635098,
      "learning_rate": 6.7064008571529275e-06,
      "loss": 0.37513597011566163,
      "num_input_tokens_seen": 368591832,
      "step": 11560,
      "train_runtime": 30229.0695,
      "train_tokens_per_second": 12193.291
    },
    {
      "epoch": 0.778683005655804,
      "grad_norm": 0.6944239729539391,
      "learning_rate": 6.703915282931411e-06,
      "loss": 0.38096203804016116,
      "num_input_tokens_seen": 368754872,
      "step": 11565,
      "train_runtime": 30241.9184,
      "train_tokens_per_second": 12193.501
    },
    {
      "epoch": 0.7790196606517641,
      "grad_norm": 0.8103378865790789,
      "learning_rate": 6.701429232215418e-06,
      "loss": 0.4006359100341797,
      "num_input_tokens_seen": 368910552,
      "step": 11570,
      "train_runtime": 30255.017,
      "train_tokens_per_second": 12193.368
    },
    {
      "epoch": 0.7793563156477242,
      "grad_norm": 0.9051610241041872,
      "learning_rate": 6.6989427057001665e-06,
      "loss": 0.3906524181365967,
      "num_input_tokens_seen": 369072008,
      "step": 11575,
      "train_runtime": 30267.5624,
      "train_tokens_per_second": 12193.648
    },
    {
      "epoch": 0.7796929706436844,
      "grad_norm": 0.7041134301033419,
      "learning_rate": 6.696455704081006e-06,
      "loss": 0.38783533573150636,
      "num_input_tokens_seen": 369231576,
      "step": 11580,
      "train_runtime": 30280.544,
      "train_tokens_per_second": 12193.69
    },
    {
      "epoch": 0.7800296256396445,
      "grad_norm": 0.8953769843443542,
      "learning_rate": 6.693968228053416e-06,
      "loss": 0.4039203643798828,
      "num_input_tokens_seen": 369393784,
      "step": 11585,
      "train_runtime": 30294.6595,
      "train_tokens_per_second": 12193.363
    },
    {
      "epoch": 0.7803662806356046,
      "grad_norm": 0.7610614278051683,
      "learning_rate": 6.691480278313013e-06,
      "loss": 0.3878408193588257,
      "num_input_tokens_seen": 369553424,
      "step": 11590,
      "train_runtime": 30308.1726,
      "train_tokens_per_second": 12193.194
    },
    {
      "epoch": 0.7807029356315648,
      "grad_norm": 0.6798490583335733,
      "learning_rate": 6.688991855555545e-06,
      "loss": 0.33335702419281005,
      "num_input_tokens_seen": 369713328,
      "step": 11595,
      "train_runtime": 30321.2149,
      "train_tokens_per_second": 12193.223
    },
    {
      "epoch": 0.7810395906275249,
      "grad_norm": 0.7259787068866819,
      "learning_rate": 6.686502960476888e-06,
      "loss": 0.36715357303619384,
      "num_input_tokens_seen": 369877168,
      "step": 11600,
      "train_runtime": 30333.65,
      "train_tokens_per_second": 12193.625
    },
    {
      "epoch": 0.7813762456234851,
      "grad_norm": 0.7316860285582696,
      "learning_rate": 6.684013593773058e-06,
      "loss": 0.325596284866333,
      "num_input_tokens_seen": 370040232,
      "step": 11605,
      "train_runtime": 30346.5011,
      "train_tokens_per_second": 12193.835
    },
    {
      "epoch": 0.7817129006194452,
      "grad_norm": 0.8633160914589647,
      "learning_rate": 6.6815237561401955e-06,
      "loss": 0.3432020664215088,
      "num_input_tokens_seen": 370191280,
      "step": 11610,
      "train_runtime": 30359.8132,
      "train_tokens_per_second": 12193.464
    },
    {
      "epoch": 0.7820495556154053,
      "grad_norm": 0.8484681444227207,
      "learning_rate": 6.679033448274577e-06,
      "loss": 0.35393800735473635,
      "num_input_tokens_seen": 370352752,
      "step": 11615,
      "train_runtime": 30372.8037,
      "train_tokens_per_second": 12193.565
    },
    {
      "epoch": 0.7823862106113655,
      "grad_norm": 0.8029709600545277,
      "learning_rate": 6.676542670872609e-06,
      "loss": 0.38855772018432616,
      "num_input_tokens_seen": 370512544,
      "step": 11620,
      "train_runtime": 30386.2008,
      "train_tokens_per_second": 12193.447
    },
    {
      "epoch": 0.7827228656073256,
      "grad_norm": 0.7740112208651453,
      "learning_rate": 6.674051424630829e-06,
      "loss": 0.3719889402389526,
      "num_input_tokens_seen": 370676008,
      "step": 11625,
      "train_runtime": 30399.7958,
      "train_tokens_per_second": 12193.372
    },
    {
      "epoch": 0.7830595206032858,
      "grad_norm": 0.7248123417324678,
      "learning_rate": 6.671559710245905e-06,
      "loss": 0.3854742288589478,
      "num_input_tokens_seen": 370835776,
      "step": 11630,
      "train_runtime": 30412.3675,
      "train_tokens_per_second": 12193.585
    },
    {
      "epoch": 0.7833961755992459,
      "grad_norm": 0.8469304588324509,
      "learning_rate": 6.669067528414639e-06,
      "loss": 0.3675633192062378,
      "num_input_tokens_seen": 370997432,
      "step": 11635,
      "train_runtime": 30424.8914,
      "train_tokens_per_second": 12193.879
    },
    {
      "epoch": 0.783732830595206,
      "grad_norm": 0.7940908814394724,
      "learning_rate": 6.6665748798339606e-06,
      "loss": 0.4011989593505859,
      "num_input_tokens_seen": 371157608,
      "step": 11640,
      "train_runtime": 30437.9075,
      "train_tokens_per_second": 12193.927
    },
    {
      "epoch": 0.7840694855911662,
      "grad_norm": 0.826378679690705,
      "learning_rate": 6.664081765200932e-06,
      "loss": 0.3851266860961914,
      "num_input_tokens_seen": 371312352,
      "step": 11645,
      "train_runtime": 30451.5569,
      "train_tokens_per_second": 12193.542
    },
    {
      "epoch": 0.7844061405871263,
      "grad_norm": 0.7285857964794943,
      "learning_rate": 6.661588185212745e-06,
      "loss": 0.40223045349121095,
      "num_input_tokens_seen": 371475112,
      "step": 11650,
      "train_runtime": 30464.576,
      "train_tokens_per_second": 12193.674
    },
    {
      "epoch": 0.7847427955830865,
      "grad_norm": 0.8379470496059848,
      "learning_rate": 6.65909414056672e-06,
      "loss": 0.3703483581542969,
      "num_input_tokens_seen": 371634128,
      "step": 11655,
      "train_runtime": 30477.9086,
      "train_tokens_per_second": 12193.557
    },
    {
      "epoch": 0.7850794505790466,
      "grad_norm": 0.8587991790713585,
      "learning_rate": 6.6565996319603075e-06,
      "loss": 0.35973210334777833,
      "num_input_tokens_seen": 371790216,
      "step": 11660,
      "train_runtime": 30490.8776,
      "train_tokens_per_second": 12193.49
    },
    {
      "epoch": 0.7854161055750067,
      "grad_norm": 0.6897779170041637,
      "learning_rate": 6.6541046600910915e-06,
      "loss": 0.3497686624526978,
      "num_input_tokens_seen": 371947840,
      "step": 11665,
      "train_runtime": 30503.9243,
      "train_tokens_per_second": 12193.442
    },
    {
      "epoch": 0.7857527605709669,
      "grad_norm": 0.8945831990006606,
      "learning_rate": 6.6516092256567824e-06,
      "loss": 0.37085485458374023,
      "num_input_tokens_seen": 372108968,
      "step": 11670,
      "train_runtime": 30516.8604,
      "train_tokens_per_second": 12193.553
    },
    {
      "epoch": 0.786089415566927,
      "grad_norm": 0.9297348262453087,
      "learning_rate": 6.64911332935522e-06,
      "loss": 0.3924281597137451,
      "num_input_tokens_seen": 372271984,
      "step": 11675,
      "train_runtime": 30530.4881,
      "train_tokens_per_second": 12193.45
    },
    {
      "epoch": 0.7864260705628872,
      "grad_norm": 0.847973747006015,
      "learning_rate": 6.646616971884373e-06,
      "loss": 0.36455602645874025,
      "num_input_tokens_seen": 372434424,
      "step": 11680,
      "train_runtime": 30543.1516,
      "train_tokens_per_second": 12193.713
    },
    {
      "epoch": 0.7867627255588473,
      "grad_norm": 0.7118435029490061,
      "learning_rate": 6.64412015394234e-06,
      "loss": 0.3776170969009399,
      "num_input_tokens_seen": 372584992,
      "step": 11685,
      "train_runtime": 30555.9126,
      "train_tokens_per_second": 12193.548
    },
    {
      "epoch": 0.7870993805548074,
      "grad_norm": 0.7858565999471193,
      "learning_rate": 6.641622876227349e-06,
      "loss": 0.3869914531707764,
      "num_input_tokens_seen": 372742968,
      "step": 11690,
      "train_runtime": 30570.1857,
      "train_tokens_per_second": 12193.023
    },
    {
      "epoch": 0.7874360355507676,
      "grad_norm": 0.9533709005709372,
      "learning_rate": 6.639125139437756e-06,
      "loss": 0.3708347320556641,
      "num_input_tokens_seen": 372905104,
      "step": 11695,
      "train_runtime": 30582.8377,
      "train_tokens_per_second": 12193.28
    },
    {
      "epoch": 0.7877726905467277,
      "grad_norm": 0.7900532863526302,
      "learning_rate": 6.6366269442720446e-06,
      "loss": 0.37346682548522947,
      "num_input_tokens_seen": 373066232,
      "step": 11700,
      "train_runtime": 30596.3182,
      "train_tokens_per_second": 12193.174
    },
    {
      "epoch": 0.7881093455426879,
      "grad_norm": 0.7949751899978424,
      "learning_rate": 6.634128291428827e-06,
      "loss": 0.43591084480285647,
      "num_input_tokens_seen": 373225944,
      "step": 11705,
      "train_runtime": 30609.6388,
      "train_tokens_per_second": 12193.086
    },
    {
      "epoch": 0.788446000538648,
      "grad_norm": 0.7662468521530408,
      "learning_rate": 6.631629181606842e-06,
      "loss": 0.38541278839111326,
      "num_input_tokens_seen": 373387920,
      "step": 11710,
      "train_runtime": 30622.2245,
      "train_tokens_per_second": 12193.364
    },
    {
      "epoch": 0.7887826555346081,
      "grad_norm": 0.8135084460960876,
      "learning_rate": 6.629129615504961e-06,
      "loss": 0.3854079723358154,
      "num_input_tokens_seen": 373550528,
      "step": 11715,
      "train_runtime": 30635.5997,
      "train_tokens_per_second": 12193.348
    },
    {
      "epoch": 0.7891193105305683,
      "grad_norm": 0.778210766219173,
      "learning_rate": 6.626629593822178e-06,
      "loss": 0.35331389904022215,
      "num_input_tokens_seen": 373710888,
      "step": 11720,
      "train_runtime": 30648.4004,
      "train_tokens_per_second": 12193.488
    },
    {
      "epoch": 0.7894559655265284,
      "grad_norm": 0.8259052580688125,
      "learning_rate": 6.624129117257616e-06,
      "loss": 0.4102816581726074,
      "num_input_tokens_seen": 373873104,
      "step": 11725,
      "train_runtime": 30661.5996,
      "train_tokens_per_second": 12193.529
    },
    {
      "epoch": 0.7897926205224886,
      "grad_norm": 0.8767344638501338,
      "learning_rate": 6.621628186510524e-06,
      "loss": 0.3956955909729004,
      "num_input_tokens_seen": 374031224,
      "step": 11730,
      "train_runtime": 30674.9705,
      "train_tokens_per_second": 12193.369
    },
    {
      "epoch": 0.7901292755184487,
      "grad_norm": 0.7554110006400164,
      "learning_rate": 6.619126802280281e-06,
      "loss": 0.353791332244873,
      "num_input_tokens_seen": 374192080,
      "step": 11735,
      "train_runtime": 30689.1202,
      "train_tokens_per_second": 12192.988
    },
    {
      "epoch": 0.7904659305144088,
      "grad_norm": 0.7991100911297014,
      "learning_rate": 6.616624965266392e-06,
      "loss": 0.3949237823486328,
      "num_input_tokens_seen": 374351896,
      "step": 11740,
      "train_runtime": 30702.5521,
      "train_tokens_per_second": 12192.859
    },
    {
      "epoch": 0.790802585510369,
      "grad_norm": 0.8078460559476426,
      "learning_rate": 6.614122676168486e-06,
      "loss": 0.4061460018157959,
      "num_input_tokens_seen": 374509912,
      "step": 11745,
      "train_runtime": 30715.1732,
      "train_tokens_per_second": 12192.994
    },
    {
      "epoch": 0.7911392405063291,
      "grad_norm": 0.8735380125724772,
      "learning_rate": 6.611619935686321e-06,
      "loss": 0.3557483196258545,
      "num_input_tokens_seen": 374666432,
      "step": 11750,
      "train_runtime": 30728.0374,
      "train_tokens_per_second": 12192.983
    },
    {
      "epoch": 0.7914758955022893,
      "grad_norm": 0.9459188275834557,
      "learning_rate": 6.609116744519779e-06,
      "loss": 0.4006349563598633,
      "num_input_tokens_seen": 374828240,
      "step": 11755,
      "train_runtime": 30741.0749,
      "train_tokens_per_second": 12193.075
    },
    {
      "epoch": 0.7918125504982494,
      "grad_norm": 0.8179656988264756,
      "learning_rate": 6.60661310336887e-06,
      "loss": 0.3551258325576782,
      "num_input_tokens_seen": 374983584,
      "step": 11760,
      "train_runtime": 30753.8753,
      "train_tokens_per_second": 12193.051
    },
    {
      "epoch": 0.7921492054942095,
      "grad_norm": 0.7871026100210354,
      "learning_rate": 6.604109012933732e-06,
      "loss": 0.3638493061065674,
      "num_input_tokens_seen": 375134472,
      "step": 11765,
      "train_runtime": 30767.3219,
      "train_tokens_per_second": 12192.627
    },
    {
      "epoch": 0.7924858604901697,
      "grad_norm": 0.7459684768926129,
      "learning_rate": 6.601604473914622e-06,
      "loss": 0.39454286098480223,
      "num_input_tokens_seen": 375293496,
      "step": 11770,
      "train_runtime": 30780.5351,
      "train_tokens_per_second": 12192.559
    },
    {
      "epoch": 0.7928225154861298,
      "grad_norm": 0.8004297652193146,
      "learning_rate": 6.599099487011928e-06,
      "loss": 0.365021276473999,
      "num_input_tokens_seen": 375457336,
      "step": 11775,
      "train_runtime": 30792.9726,
      "train_tokens_per_second": 12192.955
    },
    {
      "epoch": 0.79315917048209,
      "grad_norm": 0.7579668293097163,
      "learning_rate": 6.596594052926163e-06,
      "loss": 0.3989898204803467,
      "num_input_tokens_seen": 375619592,
      "step": 11780,
      "train_runtime": 30805.577,
      "train_tokens_per_second": 12193.233
    },
    {
      "epoch": 0.7934958254780501,
      "grad_norm": 0.6807714942097018,
      "learning_rate": 6.5940881723579595e-06,
      "loss": 0.3649766445159912,
      "num_input_tokens_seen": 375782232,
      "step": 11785,
      "train_runtime": 30818.9442,
      "train_tokens_per_second": 12193.222
    },
    {
      "epoch": 0.7938324804740102,
      "grad_norm": 0.7349600359284761,
      "learning_rate": 6.591581846008085e-06,
      "loss": 0.34767608642578124,
      "num_input_tokens_seen": 375942792,
      "step": 11790,
      "train_runtime": 30832.0265,
      "train_tokens_per_second": 12193.256
    },
    {
      "epoch": 0.7941691354699704,
      "grad_norm": 0.7825436727327969,
      "learning_rate": 6.589075074577421e-06,
      "loss": 0.37279183864593507,
      "num_input_tokens_seen": 376100072,
      "step": 11795,
      "train_runtime": 30844.8809,
      "train_tokens_per_second": 12193.274
    },
    {
      "epoch": 0.7945057904659305,
      "grad_norm": 0.7945380666861541,
      "learning_rate": 6.5865678587669805e-06,
      "loss": 0.36253886222839354,
      "num_input_tokens_seen": 376262408,
      "step": 11800,
      "train_runtime": 30858.1836,
      "train_tokens_per_second": 12193.278
    },
    {
      "epoch": 0.7948424454618906,
      "grad_norm": 0.7691648658973432,
      "learning_rate": 6.584060199277897e-06,
      "loss": 0.3986492156982422,
      "num_input_tokens_seen": 376422000,
      "step": 11805,
      "train_runtime": 30871.6977,
      "train_tokens_per_second": 12193.11
    },
    {
      "epoch": 0.7951791004578508,
      "grad_norm": 0.7803770300153521,
      "learning_rate": 6.581552096811428e-06,
      "loss": 0.38816232681274415,
      "num_input_tokens_seen": 376580200,
      "step": 11810,
      "train_runtime": 30884.2023,
      "train_tokens_per_second": 12193.295
    },
    {
      "epoch": 0.7955157554538109,
      "grad_norm": 0.789715745960649,
      "learning_rate": 6.57904355206896e-06,
      "loss": 0.3801249980926514,
      "num_input_tokens_seen": 376739504,
      "step": 11815,
      "train_runtime": 30896.773,
      "train_tokens_per_second": 12193.49
    },
    {
      "epoch": 0.7958524104497711,
      "grad_norm": 0.7711691651459045,
      "learning_rate": 6.576534565751997e-06,
      "loss": 0.3491034507751465,
      "num_input_tokens_seen": 376898512,
      "step": 11820,
      "train_runtime": 30909.4288,
      "train_tokens_per_second": 12193.642
    },
    {
      "epoch": 0.7961890654457312,
      "grad_norm": 0.7175811661801019,
      "learning_rate": 6.574025138562172e-06,
      "loss": 0.3864483594894409,
      "num_input_tokens_seen": 377053992,
      "step": 11825,
      "train_runtime": 30922.5383,
      "train_tokens_per_second": 12193.501
    },
    {
      "epoch": 0.7965257204416913,
      "grad_norm": 0.6935684369467853,
      "learning_rate": 6.571515271201234e-06,
      "loss": 0.34622974395751954,
      "num_input_tokens_seen": 377216984,
      "step": 11830,
      "train_runtime": 30936.1385,
      "train_tokens_per_second": 12193.409
    },
    {
      "epoch": 0.7968623754376515,
      "grad_norm": 0.7168792601580581,
      "learning_rate": 6.569004964371062e-06,
      "loss": 0.3999671697616577,
      "num_input_tokens_seen": 377376648,
      "step": 11835,
      "train_runtime": 30949.1353,
      "train_tokens_per_second": 12193.447
    },
    {
      "epoch": 0.7971990304336116,
      "grad_norm": 0.7563764887229703,
      "learning_rate": 6.566494218773656e-06,
      "loss": 0.38115367889404295,
      "num_input_tokens_seen": 377538656,
      "step": 11840,
      "train_runtime": 30961.7253,
      "train_tokens_per_second": 12193.722
    },
    {
      "epoch": 0.7975356854295718,
      "grad_norm": 0.8898827525439222,
      "learning_rate": 6.563983035111136e-06,
      "loss": 0.40212154388427734,
      "num_input_tokens_seen": 377700584,
      "step": 11845,
      "train_runtime": 30974.2959,
      "train_tokens_per_second": 12194.001
    },
    {
      "epoch": 0.7978723404255319,
      "grad_norm": 0.7148248979899465,
      "learning_rate": 6.561471414085748e-06,
      "loss": 0.38744680881500243,
      "num_input_tokens_seen": 377860712,
      "step": 11850,
      "train_runtime": 30986.9252,
      "train_tokens_per_second": 12194.198
    },
    {
      "epoch": 0.798208995421492,
      "grad_norm": 0.8069153370067991,
      "learning_rate": 6.558959356399859e-06,
      "loss": 0.3977175235748291,
      "num_input_tokens_seen": 378021280,
      "step": 11855,
      "train_runtime": 31000.7958,
      "train_tokens_per_second": 12193.922
    },
    {
      "epoch": 0.7985456504174522,
      "grad_norm": 0.6951368044158354,
      "learning_rate": 6.5564468627559544e-06,
      "loss": 0.3698373079299927,
      "num_input_tokens_seen": 378179704,
      "step": 11860,
      "train_runtime": 31013.9985,
      "train_tokens_per_second": 12193.839
    },
    {
      "epoch": 0.7988823054134123,
      "grad_norm": 0.750739046403931,
      "learning_rate": 6.5539339338566495e-06,
      "loss": 0.40657758712768555,
      "num_input_tokens_seen": 378341368,
      "step": 11865,
      "train_runtime": 31027.3064,
      "train_tokens_per_second": 12193.819
    },
    {
      "epoch": 0.7992189604093725,
      "grad_norm": 0.8399156667471254,
      "learning_rate": 6.5514205704046754e-06,
      "loss": 0.36922540664672854,
      "num_input_tokens_seen": 378495416,
      "step": 11870,
      "train_runtime": 31040.4479,
      "train_tokens_per_second": 12193.62
    },
    {
      "epoch": 0.7995556154053326,
      "grad_norm": 0.715294480193292,
      "learning_rate": 6.548906773102886e-06,
      "loss": 0.3407998561859131,
      "num_input_tokens_seen": 378652224,
      "step": 11875,
      "train_runtime": 31053.6044,
      "train_tokens_per_second": 12193.503
    },
    {
      "epoch": 0.7998922704012927,
      "grad_norm": 0.8117550767915204,
      "learning_rate": 6.546392542654257e-06,
      "loss": 0.37957718372344973,
      "num_input_tokens_seen": 378809304,
      "step": 11880,
      "train_runtime": 31066.8675,
      "train_tokens_per_second": 12193.354
    },
    {
      "epoch": 0.8002289253972529,
      "grad_norm": 0.7912191799640419,
      "learning_rate": 6.543877879761882e-06,
      "loss": 0.40323448181152344,
      "num_input_tokens_seen": 378966952,
      "step": 11885,
      "train_runtime": 31079.909,
      "train_tokens_per_second": 12193.31
    },
    {
      "epoch": 0.800565580393213,
      "grad_norm": 0.7607875949533927,
      "learning_rate": 6.5413627851289825e-06,
      "loss": 0.36257200241088866,
      "num_input_tokens_seen": 379128888,
      "step": 11890,
      "train_runtime": 31092.7952,
      "train_tokens_per_second": 12193.464
    },
    {
      "epoch": 0.8009022353891732,
      "grad_norm": 0.7794719917058088,
      "learning_rate": 6.538847259458896e-06,
      "loss": 0.3527742862701416,
      "num_input_tokens_seen": 379288840,
      "step": 11895,
      "train_runtime": 31105.2593,
      "train_tokens_per_second": 12193.721
    },
    {
      "epoch": 0.8012388903851333,
      "grad_norm": 0.7891004112272468,
      "learning_rate": 6.5363313034550805e-06,
      "loss": 0.3575753688812256,
      "num_input_tokens_seen": 379447224,
      "step": 11900,
      "train_runtime": 31118.8155,
      "train_tokens_per_second": 12193.498
    },
    {
      "epoch": 0.8015755453810934,
      "grad_norm": 0.8103725616145434,
      "learning_rate": 6.533814917821114e-06,
      "loss": 0.38735027313232423,
      "num_input_tokens_seen": 379609368,
      "step": 11905,
      "train_runtime": 31132.7615,
      "train_tokens_per_second": 12193.244
    },
    {
      "epoch": 0.8019122003770536,
      "grad_norm": 0.6448241207076659,
      "learning_rate": 6.531298103260696e-06,
      "loss": 0.36227076053619384,
      "num_input_tokens_seen": 379768720,
      "step": 11910,
      "train_runtime": 31146.5083,
      "train_tokens_per_second": 12192.979
    },
    {
      "epoch": 0.8022488553730137,
      "grad_norm": 0.7736871420388424,
      "learning_rate": 6.528780860477648e-06,
      "loss": 0.3709725379943848,
      "num_input_tokens_seen": 379929752,
      "step": 11915,
      "train_runtime": 31159.2273,
      "train_tokens_per_second": 12193.17
    },
    {
      "epoch": 0.8025855103689739,
      "grad_norm": 1.1546400069916714,
      "learning_rate": 6.526263190175907e-06,
      "loss": 0.35191028118133544,
      "num_input_tokens_seen": 380092624,
      "step": 11920,
      "train_runtime": 31172.2607,
      "train_tokens_per_second": 12193.297
    },
    {
      "epoch": 0.802922165364934,
      "grad_norm": 0.7104968387271792,
      "learning_rate": 6.523745093059531e-06,
      "loss": 0.3585056304931641,
      "num_input_tokens_seen": 380252752,
      "step": 11925,
      "train_runtime": 31186.2333,
      "train_tokens_per_second": 12192.968
    },
    {
      "epoch": 0.8032588203608941,
      "grad_norm": 0.8769292534207647,
      "learning_rate": 6.521226569832699e-06,
      "loss": 0.3812487840652466,
      "num_input_tokens_seen": 380414000,
      "step": 11930,
      "train_runtime": 31199.4611,
      "train_tokens_per_second": 12192.967
    },
    {
      "epoch": 0.8035954753568543,
      "grad_norm": 0.7501362778697537,
      "learning_rate": 6.518707621199707e-06,
      "loss": 0.3717649459838867,
      "num_input_tokens_seen": 380575328,
      "step": 11935,
      "train_runtime": 31212.4775,
      "train_tokens_per_second": 12193.051
    },
    {
      "epoch": 0.8039321303528144,
      "grad_norm": 0.872381604019058,
      "learning_rate": 6.516188247864972e-06,
      "loss": 0.38443984985351565,
      "num_input_tokens_seen": 380735112,
      "step": 11940,
      "train_runtime": 31225.0093,
      "train_tokens_per_second": 12193.275
    },
    {
      "epoch": 0.8042687853487746,
      "grad_norm": 0.8545876913829898,
      "learning_rate": 6.513668450533028e-06,
      "loss": 0.3629001617431641,
      "num_input_tokens_seen": 380894424,
      "step": 11945,
      "train_runtime": 31238.0883,
      "train_tokens_per_second": 12193.269
    },
    {
      "epoch": 0.8046054403447347,
      "grad_norm": 0.827176773722661,
      "learning_rate": 6.511148229908531e-06,
      "loss": 0.38925743103027344,
      "num_input_tokens_seen": 381056832,
      "step": 11950,
      "train_runtime": 31250.9692,
      "train_tokens_per_second": 12193.44
    },
    {
      "epoch": 0.8049420953406948,
      "grad_norm": 0.9609530843404303,
      "learning_rate": 6.508627586696248e-06,
      "loss": 0.3513191699981689,
      "num_input_tokens_seen": 381213320,
      "step": 11955,
      "train_runtime": 31263.6651,
      "train_tokens_per_second": 12193.494
    },
    {
      "epoch": 0.805278750336655,
      "grad_norm": 0.783032204711069,
      "learning_rate": 6.5061065216010695e-06,
      "loss": 0.3700404167175293,
      "num_input_tokens_seen": 381373160,
      "step": 11960,
      "train_runtime": 31276.3491,
      "train_tokens_per_second": 12193.66
    },
    {
      "epoch": 0.8056154053326151,
      "grad_norm": 0.777233594202281,
      "learning_rate": 6.5035850353280075e-06,
      "loss": 0.34110169410705565,
      "num_input_tokens_seen": 381530528,
      "step": 11965,
      "train_runtime": 31289.4785,
      "train_tokens_per_second": 12193.573
    },
    {
      "epoch": 0.8059520603285752,
      "grad_norm": 0.7547207543315777,
      "learning_rate": 6.501063128582183e-06,
      "loss": 0.379962158203125,
      "num_input_tokens_seen": 381688176,
      "step": 11970,
      "train_runtime": 31303.276,
      "train_tokens_per_second": 12193.234
    },
    {
      "epoch": 0.8062887153245354,
      "grad_norm": 0.9615151114054373,
      "learning_rate": 6.498540802068843e-06,
      "loss": 0.3809072971343994,
      "num_input_tokens_seen": 381847024,
      "step": 11975,
      "train_runtime": 31316.5577,
      "train_tokens_per_second": 12193.135
    },
    {
      "epoch": 0.8066253703204955,
      "grad_norm": 0.7860950420896803,
      "learning_rate": 6.4960180564933454e-06,
      "loss": 0.36175639629364015,
      "num_input_tokens_seen": 382009984,
      "step": 11980,
      "train_runtime": 31329.3754,
      "train_tokens_per_second": 12193.348
    },
    {
      "epoch": 0.8069620253164557,
      "grad_norm": 0.8008864490800088,
      "learning_rate": 6.493494892561167e-06,
      "loss": 0.3794721126556396,
      "num_input_tokens_seen": 382166688,
      "step": 11985,
      "train_runtime": 31342.2375,
      "train_tokens_per_second": 12193.344
    },
    {
      "epoch": 0.8072986803124158,
      "grad_norm": 0.8324350584857002,
      "learning_rate": 6.490971310977907e-06,
      "loss": 0.39203081130981443,
      "num_input_tokens_seen": 382327616,
      "step": 11990,
      "train_runtime": 31354.6834,
      "train_tokens_per_second": 12193.637
    },
    {
      "epoch": 0.807635335308376,
      "grad_norm": 0.7621675364319107,
      "learning_rate": 6.488447312449273e-06,
      "loss": 0.3973099946975708,
      "num_input_tokens_seen": 382487256,
      "step": 11995,
      "train_runtime": 31367.9741,
      "train_tokens_per_second": 12193.559
    },
    {
      "epoch": 0.8079719903043361,
      "grad_norm": 0.9002205018927582,
      "learning_rate": 6.485922897681095e-06,
      "loss": 0.3939211845397949,
      "num_input_tokens_seen": 382648576,
      "step": 12000,
      "train_runtime": 31380.8063,
      "train_tokens_per_second": 12193.714
    },
    {
      "epoch": 0.8083086453002962,
      "grad_norm": 0.7995594834427153,
      "learning_rate": 6.483398067379316e-06,
      "loss": 0.417647123336792,
      "num_input_tokens_seen": 382810008,
      "step": 12005,
      "train_runtime": 31393.2985,
      "train_tokens_per_second": 12194.004
    },
    {
      "epoch": 0.8086453002962564,
      "grad_norm": 0.7264461541616981,
      "learning_rate": 6.480872822249998e-06,
      "loss": 0.3664550304412842,
      "num_input_tokens_seen": 382972040,
      "step": 12010,
      "train_runtime": 31407.1732,
      "train_tokens_per_second": 12193.776
    },
    {
      "epoch": 0.8089819552922165,
      "grad_norm": 0.7903704742653824,
      "learning_rate": 6.478347162999318e-06,
      "loss": 0.39343421459197997,
      "num_input_tokens_seen": 383131016,
      "step": 12015,
      "train_runtime": 31420.1052,
      "train_tokens_per_second": 12193.817
    },
    {
      "epoch": 0.8093186102881766,
      "grad_norm": 0.7110519624982528,
      "learning_rate": 6.475821090333568e-06,
      "loss": 0.3729506492614746,
      "num_input_tokens_seen": 383291920,
      "step": 12020,
      "train_runtime": 31432.9842,
      "train_tokens_per_second": 12193.94
    },
    {
      "epoch": 0.8096552652841368,
      "grad_norm": 0.7870705845706162,
      "learning_rate": 6.473294604959156e-06,
      "loss": 0.391314959526062,
      "num_input_tokens_seen": 383447920,
      "step": 12025,
      "train_runtime": 31446.292,
      "train_tokens_per_second": 12193.74
    },
    {
      "epoch": 0.8099919202800969,
      "grad_norm": 0.8003202166281154,
      "learning_rate": 6.470767707582605e-06,
      "loss": 0.34874165058135986,
      "num_input_tokens_seen": 383610568,
      "step": 12030,
      "train_runtime": 31459.6632,
      "train_tokens_per_second": 12193.728
    },
    {
      "epoch": 0.8103285752760571,
      "grad_norm": 1.062551099781154,
      "learning_rate": 6.468240398910555e-06,
      "loss": 0.3764477252960205,
      "num_input_tokens_seen": 383764544,
      "step": 12035,
      "train_runtime": 31473.6491,
      "train_tokens_per_second": 12193.201
    },
    {
      "epoch": 0.8106652302720172,
      "grad_norm": 0.762018094492472,
      "learning_rate": 6.4657126796497595e-06,
      "loss": 0.36851727962493896,
      "num_input_tokens_seen": 383928384,
      "step": 12040,
      "train_runtime": 31486.0724,
      "train_tokens_per_second": 12193.594
    },
    {
      "epoch": 0.8110018852679773,
      "grad_norm": 0.851367240405606,
      "learning_rate": 6.463184550507087e-06,
      "loss": 0.3719032764434814,
      "num_input_tokens_seen": 384090160,
      "step": 12045,
      "train_runtime": 31500.2109,
      "train_tokens_per_second": 12193.257
    },
    {
      "epoch": 0.8113385402639375,
      "grad_norm": 0.7869914824478589,
      "learning_rate": 6.46065601218952e-06,
      "loss": 0.3447405338287354,
      "num_input_tokens_seen": 384251224,
      "step": 12050,
      "train_runtime": 31513.6496,
      "train_tokens_per_second": 12193.168
    },
    {
      "epoch": 0.8116751952598976,
      "grad_norm": 0.7358766036384202,
      "learning_rate": 6.458127065404156e-06,
      "loss": 0.3866471290588379,
      "num_input_tokens_seen": 384414408,
      "step": 12055,
      "train_runtime": 31526.5033,
      "train_tokens_per_second": 12193.373
    },
    {
      "epoch": 0.8120118502558578,
      "grad_norm": 0.981258791483655,
      "learning_rate": 6.45559771085821e-06,
      "loss": 0.3673221588134766,
      "num_input_tokens_seen": 384575848,
      "step": 12060,
      "train_runtime": 31539.5126,
      "train_tokens_per_second": 12193.462
    },
    {
      "epoch": 0.8123485052518179,
      "grad_norm": 0.7743307413906022,
      "learning_rate": 6.453067949259005e-06,
      "loss": 0.3619658946990967,
      "num_input_tokens_seen": 384738280,
      "step": 12065,
      "train_runtime": 31552.5195,
      "train_tokens_per_second": 12193.583
    },
    {
      "epoch": 0.812685160247778,
      "grad_norm": 0.7399302549467262,
      "learning_rate": 6.450537781313982e-06,
      "loss": 0.3535545110702515,
      "num_input_tokens_seen": 384896840,
      "step": 12070,
      "train_runtime": 31565.4824,
      "train_tokens_per_second": 12193.599
    },
    {
      "epoch": 0.8130218152437382,
      "grad_norm": 0.728045628593235,
      "learning_rate": 6.448007207730693e-06,
      "loss": 0.3631807088851929,
      "num_input_tokens_seen": 385055432,
      "step": 12075,
      "train_runtime": 31578.3246,
      "train_tokens_per_second": 12193.662
    },
    {
      "epoch": 0.8133584702396983,
      "grad_norm": 0.8472448042332094,
      "learning_rate": 6.445476229216805e-06,
      "loss": 0.3557400703430176,
      "num_input_tokens_seen": 385218320,
      "step": 12080,
      "train_runtime": 31591.0798,
      "train_tokens_per_second": 12193.895
    },
    {
      "epoch": 0.8136951252356585,
      "grad_norm": 1.0485866986367949,
      "learning_rate": 6.442944846480099e-06,
      "loss": 0.40411863327026365,
      "num_input_tokens_seen": 385377384,
      "step": 12085,
      "train_runtime": 31605.191,
      "train_tokens_per_second": 12193.484
    },
    {
      "epoch": 0.8140317802316186,
      "grad_norm": 0.8510072556015127,
      "learning_rate": 6.44041306022847e-06,
      "loss": 0.38596439361572266,
      "num_input_tokens_seen": 385537192,
      "step": 12090,
      "train_runtime": 31617.608,
      "train_tokens_per_second": 12193.75
    },
    {
      "epoch": 0.8143684352275787,
      "grad_norm": 0.8422873278675023,
      "learning_rate": 6.43788087116992e-06,
      "loss": 0.3989336252212524,
      "num_input_tokens_seen": 385699808,
      "step": 12095,
      "train_runtime": 31630.055,
      "train_tokens_per_second": 12194.092
    },
    {
      "epoch": 0.8147050902235389,
      "grad_norm": 0.85877522938351,
      "learning_rate": 6.43534828001257e-06,
      "loss": 0.3854092597961426,
      "num_input_tokens_seen": 385858016,
      "step": 12100,
      "train_runtime": 31642.5939,
      "train_tokens_per_second": 12194.26
    },
    {
      "epoch": 0.815041745219499,
      "grad_norm": 0.857029302516094,
      "learning_rate": 6.43281528746465e-06,
      "loss": 0.37953619956970214,
      "num_input_tokens_seen": 386011872,
      "step": 12105,
      "train_runtime": 31655.8764,
      "train_tokens_per_second": 12194.004
    },
    {
      "epoch": 0.8153784002154592,
      "grad_norm": 0.6433767713001484,
      "learning_rate": 6.430281894234504e-06,
      "loss": 0.37154712677001955,
      "num_input_tokens_seen": 386171288,
      "step": 12110,
      "train_runtime": 31668.6913,
      "train_tokens_per_second": 12194.103
    },
    {
      "epoch": 0.8157150552114193,
      "grad_norm": 0.7461633622032543,
      "learning_rate": 6.427748101030587e-06,
      "loss": 0.38327181339263916,
      "num_input_tokens_seen": 386326664,
      "step": 12115,
      "train_runtime": 31681.278,
      "train_tokens_per_second": 12194.163
    },
    {
      "epoch": 0.8160517102073794,
      "grad_norm": 0.7772147899433034,
      "learning_rate": 6.425213908561468e-06,
      "loss": 0.37313592433929443,
      "num_input_tokens_seen": 386488264,
      "step": 12120,
      "train_runtime": 31693.9949,
      "train_tokens_per_second": 12194.369
    },
    {
      "epoch": 0.8163883652033396,
      "grad_norm": 0.7206897367997068,
      "learning_rate": 6.422679317535823e-06,
      "loss": 0.3711486577987671,
      "num_input_tokens_seen": 386647320,
      "step": 12125,
      "train_runtime": 31708.009,
      "train_tokens_per_second": 12193.996
    },
    {
      "epoch": 0.8167250201992997,
      "grad_norm": 0.8057360201793661,
      "learning_rate": 6.420144328662441e-06,
      "loss": 0.43384904861450196,
      "num_input_tokens_seen": 386809216,
      "step": 12130,
      "train_runtime": 31721.0471,
      "train_tokens_per_second": 12194.087
    },
    {
      "epoch": 0.8170616751952599,
      "grad_norm": 0.7717114124652743,
      "learning_rate": 6.417608942650229e-06,
      "loss": 0.38415067195892333,
      "num_input_tokens_seen": 386963784,
      "step": 12135,
      "train_runtime": 31733.5763,
      "train_tokens_per_second": 12194.144
    },
    {
      "epoch": 0.81739833019122,
      "grad_norm": 0.7522938088955058,
      "learning_rate": 6.415073160208196e-06,
      "loss": 0.37050905227661135,
      "num_input_tokens_seen": 387125464,
      "step": 12140,
      "train_runtime": 31746.7675,
      "train_tokens_per_second": 12194.17
    },
    {
      "epoch": 0.8177349851871801,
      "grad_norm": 0.7993705575590916,
      "learning_rate": 6.412536982045465e-06,
      "loss": 0.3859358310699463,
      "num_input_tokens_seen": 387278152,
      "step": 12145,
      "train_runtime": 31760.7122,
      "train_tokens_per_second": 12193.623
    },
    {
      "epoch": 0.8180716401831403,
      "grad_norm": 0.894561034480977,
      "learning_rate": 6.410000408871273e-06,
      "loss": 0.38177971839904784,
      "num_input_tokens_seen": 387436304,
      "step": 12150,
      "train_runtime": 31774.1816,
      "train_tokens_per_second": 12193.431
    },
    {
      "epoch": 0.8184082951791004,
      "grad_norm": 0.9046929829045078,
      "learning_rate": 6.407463441394961e-06,
      "loss": 0.3825327157974243,
      "num_input_tokens_seen": 387599384,
      "step": 12155,
      "train_runtime": 31787.0279,
      "train_tokens_per_second": 12193.634
    },
    {
      "epoch": 0.8187449501750605,
      "grad_norm": 0.8758116040527244,
      "learning_rate": 6.4049260803259874e-06,
      "loss": 0.39542713165283205,
      "num_input_tokens_seen": 387761568,
      "step": 12160,
      "train_runtime": 31799.6084,
      "train_tokens_per_second": 12193.91
    },
    {
      "epoch": 0.8190816051710207,
      "grad_norm": 0.8766352701967576,
      "learning_rate": 6.402388326373914e-06,
      "loss": 0.350589919090271,
      "num_input_tokens_seen": 387920744,
      "step": 12165,
      "train_runtime": 31812.7441,
      "train_tokens_per_second": 12193.879
    },
    {
      "epoch": 0.8194182601669808,
      "grad_norm": 0.7660274946843856,
      "learning_rate": 6.399850180248418e-06,
      "loss": 0.4039796829223633,
      "num_input_tokens_seen": 388084584,
      "step": 12170,
      "train_runtime": 31825.1633,
      "train_tokens_per_second": 12194.268
    },
    {
      "epoch": 0.819754915162941,
      "grad_norm": 0.8158015564138612,
      "learning_rate": 6.397311642659283e-06,
      "loss": 0.39557342529296874,
      "num_input_tokens_seen": 388242832,
      "step": 12175,
      "train_runtime": 31838.0977,
      "train_tokens_per_second": 12194.285
    },
    {
      "epoch": 0.8200915701589011,
      "grad_norm": 0.8021180553137337,
      "learning_rate": 6.394772714316402e-06,
      "loss": 0.37081055641174315,
      "num_input_tokens_seen": 388404784,
      "step": 12180,
      "train_runtime": 31851.239,
      "train_tokens_per_second": 12194.338
    },
    {
      "epoch": 0.8204282251548612,
      "grad_norm": 0.7901376160060586,
      "learning_rate": 6.392233395929779e-06,
      "loss": 0.3661466598510742,
      "num_input_tokens_seen": 388560464,
      "step": 12185,
      "train_runtime": 31864.4365,
      "train_tokens_per_second": 12194.173
    },
    {
      "epoch": 0.8207648801508214,
      "grad_norm": 0.7891281122473084,
      "learning_rate": 6.389693688209526e-06,
      "loss": 0.35634541511535645,
      "num_input_tokens_seen": 388720120,
      "step": 12190,
      "train_runtime": 31877.8118,
      "train_tokens_per_second": 12194.065
    },
    {
      "epoch": 0.8211015351467815,
      "grad_norm": 0.8809576462144182,
      "learning_rate": 6.387153591865867e-06,
      "loss": 0.3973454236984253,
      "num_input_tokens_seen": 388882304,
      "step": 12195,
      "train_runtime": 31890.3993,
      "train_tokens_per_second": 12194.338
    },
    {
      "epoch": 0.8214381901427417,
      "grad_norm": 0.7341726379590188,
      "learning_rate": 6.384613107609127e-06,
      "loss": 0.3915351152420044,
      "num_input_tokens_seen": 389041048,
      "step": 12200,
      "train_runtime": 31903.9012,
      "train_tokens_per_second": 12194.153
    },
    {
      "epoch": 0.8217748451387018,
      "grad_norm": 0.7874314855638644,
      "learning_rate": 6.382072236149747e-06,
      "loss": 0.366193962097168,
      "num_input_tokens_seen": 389202064,
      "step": 12205,
      "train_runtime": 31916.8361,
      "train_tokens_per_second": 12194.256
    },
    {
      "epoch": 0.8221115001346619,
      "grad_norm": 0.6933149435196174,
      "learning_rate": 6.379530978198273e-06,
      "loss": 0.332134485244751,
      "num_input_tokens_seen": 389362872,
      "step": 12210,
      "train_runtime": 31930.3204,
      "train_tokens_per_second": 12194.142
    },
    {
      "epoch": 0.8224481551306222,
      "grad_norm": 0.6940979609333797,
      "learning_rate": 6.376989334465361e-06,
      "loss": 0.35556254386901853,
      "num_input_tokens_seen": 389525024,
      "step": 12215,
      "train_runtime": 31943.5439,
      "train_tokens_per_second": 12194.171
    },
    {
      "epoch": 0.8227848101265823,
      "grad_norm": 0.7227755774407384,
      "learning_rate": 6.374447305661772e-06,
      "loss": 0.3727825164794922,
      "num_input_tokens_seen": 389688864,
      "step": 12220,
      "train_runtime": 31955.9606,
      "train_tokens_per_second": 12194.56
    },
    {
      "epoch": 0.8231214651225425,
      "grad_norm": 0.6988930605292222,
      "learning_rate": 6.3719048924983776e-06,
      "loss": 0.3540207386016846,
      "num_input_tokens_seen": 389845792,
      "step": 12225,
      "train_runtime": 31969.0842,
      "train_tokens_per_second": 12194.462
    },
    {
      "epoch": 0.8234581201185026,
      "grad_norm": 0.8310840987190118,
      "learning_rate": 6.369362095686152e-06,
      "loss": 0.36890084743499757,
      "num_input_tokens_seen": 390001440,
      "step": 12230,
      "train_runtime": 31982.3101,
      "train_tokens_per_second": 12194.286
    },
    {
      "epoch": 0.8237947751144628,
      "grad_norm": 0.8124630924945688,
      "learning_rate": 6.366818915936185e-06,
      "loss": 0.35982298851013184,
      "num_input_tokens_seen": 390159296,
      "step": 12235,
      "train_runtime": 31995.118,
      "train_tokens_per_second": 12194.338
    },
    {
      "epoch": 0.8241314301104229,
      "grad_norm": 0.8372260135657977,
      "learning_rate": 6.364275353959667e-06,
      "loss": 0.37121498584747314,
      "num_input_tokens_seen": 390319312,
      "step": 12240,
      "train_runtime": 32008.7376,
      "train_tokens_per_second": 12194.149
    },
    {
      "epoch": 0.824468085106383,
      "grad_norm": 0.8313294107945519,
      "learning_rate": 6.3617314104678966e-06,
      "loss": 0.35704631805419923,
      "num_input_tokens_seen": 390482904,
      "step": 12245,
      "train_runtime": 32021.15,
      "train_tokens_per_second": 12194.531
    },
    {
      "epoch": 0.8248047401023432,
      "grad_norm": 0.7843128998750705,
      "learning_rate": 6.359187086172278e-06,
      "loss": 0.39774384498596194,
      "num_input_tokens_seen": 390637808,
      "step": 12250,
      "train_runtime": 32034.861,
      "train_tokens_per_second": 12194.147
    },
    {
      "epoch": 0.8251413950983033,
      "grad_norm": 0.7876615477955945,
      "learning_rate": 6.356642381784326e-06,
      "loss": 0.3775974750518799,
      "num_input_tokens_seen": 390795024,
      "step": 12255,
      "train_runtime": 32048.4447,
      "train_tokens_per_second": 12193.884
    },
    {
      "epoch": 0.8254780500942634,
      "grad_norm": 0.9216690376822768,
      "learning_rate": 6.354097298015658e-06,
      "loss": 0.362050199508667,
      "num_input_tokens_seen": 390944648,
      "step": 12260,
      "train_runtime": 32061.1133,
      "train_tokens_per_second": 12193.733
    },
    {
      "epoch": 0.8258147050902236,
      "grad_norm": 0.7988495958835761,
      "learning_rate": 6.3515518355779985e-06,
      "loss": 0.3888446807861328,
      "num_input_tokens_seen": 391099056,
      "step": 12265,
      "train_runtime": 32074.6449,
      "train_tokens_per_second": 12193.403
    },
    {
      "epoch": 0.8261513600861837,
      "grad_norm": 0.7138841083910371,
      "learning_rate": 6.349005995183177e-06,
      "loss": 0.3540192127227783,
      "num_input_tokens_seen": 391260800,
      "step": 12270,
      "train_runtime": 32087.7866,
      "train_tokens_per_second": 12193.449
    },
    {
      "epoch": 0.8264880150821439,
      "grad_norm": 0.6480792425134566,
      "learning_rate": 6.346459777543131e-06,
      "loss": 0.3683655023574829,
      "num_input_tokens_seen": 391419832,
      "step": 12275,
      "train_runtime": 32101.2186,
      "train_tokens_per_second": 12193.301
    },
    {
      "epoch": 0.826824670078104,
      "grad_norm": 0.8044098314399165,
      "learning_rate": 6.343913183369902e-06,
      "loss": 0.36744327545166017,
      "num_input_tokens_seen": 391577560,
      "step": 12280,
      "train_runtime": 32114.3778,
      "train_tokens_per_second": 12193.216
    },
    {
      "epoch": 0.8271613250740641,
      "grad_norm": 0.779058959186197,
      "learning_rate": 6.341366213375637e-06,
      "loss": 0.3613635778427124,
      "num_input_tokens_seen": 391739864,
      "step": 12285,
      "train_runtime": 32126.9024,
      "train_tokens_per_second": 12193.515
    },
    {
      "epoch": 0.8274979800700243,
      "grad_norm": 0.6948633336185721,
      "learning_rate": 6.338818868272587e-06,
      "loss": 0.371414852142334,
      "num_input_tokens_seen": 391899528,
      "step": 12290,
      "train_runtime": 32140.458,
      "train_tokens_per_second": 12193.34
    },
    {
      "epoch": 0.8278346350659844,
      "grad_norm": 0.7438132262080402,
      "learning_rate": 6.336271148773111e-06,
      "loss": 0.38228573799133303,
      "num_input_tokens_seen": 392060072,
      "step": 12295,
      "train_runtime": 32152.891,
      "train_tokens_per_second": 12193.618
    },
    {
      "epoch": 0.8281712900619446,
      "grad_norm": 0.7453550718352379,
      "learning_rate": 6.333723055589669e-06,
      "loss": 0.4205354690551758,
      "num_input_tokens_seen": 392221320,
      "step": 12300,
      "train_runtime": 32165.6354,
      "train_tokens_per_second": 12193.8
    },
    {
      "epoch": 0.8285079450579047,
      "grad_norm": 0.7820962932589253,
      "learning_rate": 6.331174589434826e-06,
      "loss": 0.3833889961242676,
      "num_input_tokens_seen": 392380968,
      "step": 12305,
      "train_runtime": 32178.9551,
      "train_tokens_per_second": 12193.714
    },
    {
      "epoch": 0.8288446000538648,
      "grad_norm": 0.8149128419904814,
      "learning_rate": 6.328625751021254e-06,
      "loss": 0.34661312103271485,
      "num_input_tokens_seen": 392535416,
      "step": 12310,
      "train_runtime": 32192.133,
      "train_tokens_per_second": 12193.52
    },
    {
      "epoch": 0.829181255049825,
      "grad_norm": 0.7736543261604117,
      "learning_rate": 6.326076541061729e-06,
      "loss": 0.3403275489807129,
      "num_input_tokens_seen": 392695256,
      "step": 12315,
      "train_runtime": 32205.282,
      "train_tokens_per_second": 12193.505
    },
    {
      "epoch": 0.8295179100457851,
      "grad_norm": 0.7660002945299594,
      "learning_rate": 6.323526960269127e-06,
      "loss": 0.40069074630737306,
      "num_input_tokens_seen": 392850600,
      "step": 12320,
      "train_runtime": 32218.3002,
      "train_tokens_per_second": 12193.399
    },
    {
      "epoch": 0.8298545650417453,
      "grad_norm": 0.7413057333512925,
      "learning_rate": 6.3209770093564315e-06,
      "loss": 0.3497325897216797,
      "num_input_tokens_seen": 393012200,
      "step": 12325,
      "train_runtime": 32231.3359,
      "train_tokens_per_second": 12193.482
    },
    {
      "epoch": 0.8301912200377054,
      "grad_norm": 0.866586278590947,
      "learning_rate": 6.318426689036727e-06,
      "loss": 0.37470273971557616,
      "num_input_tokens_seen": 393172032,
      "step": 12330,
      "train_runtime": 32244.4995,
      "train_tokens_per_second": 12193.461
    },
    {
      "epoch": 0.8305278750336655,
      "grad_norm": 0.8055220009336936,
      "learning_rate": 6.315876000023202e-06,
      "loss": 0.3900545597076416,
      "num_input_tokens_seen": 393328824,
      "step": 12335,
      "train_runtime": 32257.6693,
      "train_tokens_per_second": 12193.343
    },
    {
      "epoch": 0.8308645300296257,
      "grad_norm": 0.8293665669039914,
      "learning_rate": 6.313324943029151e-06,
      "loss": 0.39550180435180665,
      "num_input_tokens_seen": 393489888,
      "step": 12340,
      "train_runtime": 32270.5043,
      "train_tokens_per_second": 12193.484
    },
    {
      "epoch": 0.8312011850255858,
      "grad_norm": 0.7568454728583817,
      "learning_rate": 6.310773518767967e-06,
      "loss": 0.3788604736328125,
      "num_input_tokens_seen": 393647456,
      "step": 12345,
      "train_runtime": 32283.9325,
      "train_tokens_per_second": 12193.293
    },
    {
      "epoch": 0.831537840021546,
      "grad_norm": 0.7713231753128954,
      "learning_rate": 6.30822172795315e-06,
      "loss": 0.385570764541626,
      "num_input_tokens_seen": 393799648,
      "step": 12350,
      "train_runtime": 32296.8967,
      "train_tokens_per_second": 12193.111
    },
    {
      "epoch": 0.8318744950175061,
      "grad_norm": 0.9146456078996831,
      "learning_rate": 6.3056695712982965e-06,
      "loss": 0.42081537246704104,
      "num_input_tokens_seen": 393956744,
      "step": 12355,
      "train_runtime": 32310.0641,
      "train_tokens_per_second": 12193.004
    },
    {
      "epoch": 0.8322111500134662,
      "grad_norm": 0.7474238076935905,
      "learning_rate": 6.303117049517111e-06,
      "loss": 0.3621697425842285,
      "num_input_tokens_seen": 394116760,
      "step": 12360,
      "train_runtime": 32323.663,
      "train_tokens_per_second": 12192.825
    },
    {
      "epoch": 0.8325478050094264,
      "grad_norm": 0.795037804719156,
      "learning_rate": 6.300564163323398e-06,
      "loss": 0.35672545433044434,
      "num_input_tokens_seen": 394275192,
      "step": 12365,
      "train_runtime": 32336.9922,
      "train_tokens_per_second": 12192.698
    },
    {
      "epoch": 0.8328844600053865,
      "grad_norm": 0.8854781920548779,
      "learning_rate": 6.298010913431065e-06,
      "loss": 0.38801937103271483,
      "num_input_tokens_seen": 394436384,
      "step": 12370,
      "train_runtime": 32350.6413,
      "train_tokens_per_second": 12192.537
    },
    {
      "epoch": 0.8332211150013467,
      "grad_norm": 0.7980405079907703,
      "learning_rate": 6.295457300554119e-06,
      "loss": 0.35552163124084474,
      "num_input_tokens_seen": 394595952,
      "step": 12375,
      "train_runtime": 32363.214,
      "train_tokens_per_second": 12192.731
    },
    {
      "epoch": 0.8335577699973068,
      "grad_norm": 0.8074996115185465,
      "learning_rate": 6.292903325406668e-06,
      "loss": 0.39913334846496584,
      "num_input_tokens_seen": 394753408,
      "step": 12380,
      "train_runtime": 32376.9682,
      "train_tokens_per_second": 12192.414
    },
    {
      "epoch": 0.8338944249932669,
      "grad_norm": 0.7837634454065847,
      "learning_rate": 6.2903489887029255e-06,
      "loss": 0.35669057369232177,
      "num_input_tokens_seen": 394915088,
      "step": 12385,
      "train_runtime": 32390.7895,
      "train_tokens_per_second": 12192.203
    },
    {
      "epoch": 0.8342310799892271,
      "grad_norm": 0.7084060362320729,
      "learning_rate": 6.287794291157204e-06,
      "loss": 0.3691649198532104,
      "num_input_tokens_seen": 395072224,
      "step": 12390,
      "train_runtime": 32403.4217,
      "train_tokens_per_second": 12192.3
    },
    {
      "epoch": 0.8345677349851872,
      "grad_norm": 0.7529181885364955,
      "learning_rate": 6.285239233483915e-06,
      "loss": 0.3749589204788208,
      "num_input_tokens_seen": 395234944,
      "step": 12395,
      "train_runtime": 32416.1394,
      "train_tokens_per_second": 12192.536
    },
    {
      "epoch": 0.8349043899811474,
      "grad_norm": 0.769698710449742,
      "learning_rate": 6.2826838163975724e-06,
      "loss": 0.386875581741333,
      "num_input_tokens_seen": 395391440,
      "step": 12400,
      "train_runtime": 32429.5844,
      "train_tokens_per_second": 12192.307
    },
    {
      "epoch": 0.8352410449771075,
      "grad_norm": 0.7488662560308984,
      "learning_rate": 6.2801280406127905e-06,
      "loss": 0.37825257778167726,
      "num_input_tokens_seen": 395553904,
      "step": 12405,
      "train_runtime": 32442.8698,
      "train_tokens_per_second": 12192.322
    },
    {
      "epoch": 0.8355776999730676,
      "grad_norm": 0.7774460685339145,
      "learning_rate": 6.2775719068442845e-06,
      "loss": 0.34813885688781737,
      "num_input_tokens_seen": 395712112,
      "step": 12410,
      "train_runtime": 32455.2933,
      "train_tokens_per_second": 12192.529
    },
    {
      "epoch": 0.8359143549690278,
      "grad_norm": 0.8239099626289091,
      "learning_rate": 6.275015415806869e-06,
      "loss": 0.350118088722229,
      "num_input_tokens_seen": 395867608,
      "step": 12415,
      "train_runtime": 32468.3786,
      "train_tokens_per_second": 12192.405
    },
    {
      "epoch": 0.8362510099649879,
      "grad_norm": 0.7219717097214713,
      "learning_rate": 6.272458568215458e-06,
      "loss": 0.3892244338989258,
      "num_input_tokens_seen": 396026312,
      "step": 12420,
      "train_runtime": 32481.0107,
      "train_tokens_per_second": 12192.549
    },
    {
      "epoch": 0.836587664960948,
      "grad_norm": 0.8301911040076562,
      "learning_rate": 6.269901364785066e-06,
      "loss": 0.37661147117614746,
      "num_input_tokens_seen": 396187992,
      "step": 12425,
      "train_runtime": 32494.4183,
      "train_tokens_per_second": 12192.494
    },
    {
      "epoch": 0.8369243199569082,
      "grad_norm": 0.7788347916884113,
      "learning_rate": 6.2673438062308055e-06,
      "loss": 0.3893986701965332,
      "num_input_tokens_seen": 396351832,
      "step": 12430,
      "train_runtime": 32506.8253,
      "train_tokens_per_second": 12192.88
    },
    {
      "epoch": 0.8372609749528683,
      "grad_norm": 0.8573069353423344,
      "learning_rate": 6.264785893267892e-06,
      "loss": 0.3590924501419067,
      "num_input_tokens_seen": 396502888,
      "step": 12435,
      "train_runtime": 32520.0825,
      "train_tokens_per_second": 12192.555
    },
    {
      "epoch": 0.8375976299488285,
      "grad_norm": 0.6943543037801961,
      "learning_rate": 6.2622276266116366e-06,
      "loss": 0.3629936218261719,
      "num_input_tokens_seen": 396661248,
      "step": 12440,
      "train_runtime": 32532.5715,
      "train_tokens_per_second": 12192.742
    },
    {
      "epoch": 0.8379342849447886,
      "grad_norm": 0.6725735055738319,
      "learning_rate": 6.25966900697745e-06,
      "loss": 0.3948936462402344,
      "num_input_tokens_seen": 396821928,
      "step": 12445,
      "train_runtime": 32545.3617,
      "train_tokens_per_second": 12192.887
    },
    {
      "epoch": 0.8382709399407487,
      "grad_norm": 0.7204533884937845,
      "learning_rate": 6.257110035080843e-06,
      "loss": 0.349798583984375,
      "num_input_tokens_seen": 396981288,
      "step": 12450,
      "train_runtime": 32557.9665,
      "train_tokens_per_second": 12193.062
    },
    {
      "epoch": 0.8386075949367089,
      "grad_norm": 0.7786696931575168,
      "learning_rate": 6.254550711637422e-06,
      "loss": 0.40909461975097655,
      "num_input_tokens_seen": 397136128,
      "step": 12455,
      "train_runtime": 32570.8675,
      "train_tokens_per_second": 12192.986
    },
    {
      "epoch": 0.838944249932669,
      "grad_norm": 0.7836857074035071,
      "learning_rate": 6.251991037362897e-06,
      "loss": 0.3861907958984375,
      "num_input_tokens_seen": 397289120,
      "step": 12460,
      "train_runtime": 32584.0985,
      "train_tokens_per_second": 12192.73
    },
    {
      "epoch": 0.8392809049286292,
      "grad_norm": 0.7477115764264664,
      "learning_rate": 6.249431012973068e-06,
      "loss": 0.38586227893829345,
      "num_input_tokens_seen": 397450760,
      "step": 12465,
      "train_runtime": 32597.3657,
      "train_tokens_per_second": 12192.726
    },
    {
      "epoch": 0.8396175599245893,
      "grad_norm": 0.724986745577849,
      "learning_rate": 6.246870639183843e-06,
      "loss": 0.3684225082397461,
      "num_input_tokens_seen": 397608824,
      "step": 12470,
      "train_runtime": 32610.9582,
      "train_tokens_per_second": 12192.491
    },
    {
      "epoch": 0.8399542149205494,
      "grad_norm": 0.8242561774267139,
      "learning_rate": 6.2443099167112185e-06,
      "loss": 0.4017937660217285,
      "num_input_tokens_seen": 397769304,
      "step": 12475,
      "train_runtime": 32623.9407,
      "train_tokens_per_second": 12192.558
    },
    {
      "epoch": 0.8402908699165096,
      "grad_norm": 0.7507190913705805,
      "learning_rate": 6.241748846271293e-06,
      "loss": 0.3807955503463745,
      "num_input_tokens_seen": 397932720,
      "step": 12480,
      "train_runtime": 32637.5938,
      "train_tokens_per_second": 12192.465
    },
    {
      "epoch": 0.8406275249124697,
      "grad_norm": 0.8215140678733114,
      "learning_rate": 6.239187428580263e-06,
      "loss": 0.3828019142150879,
      "num_input_tokens_seen": 398083808,
      "step": 12485,
      "train_runtime": 32650.0526,
      "train_tokens_per_second": 12192.44
    },
    {
      "epoch": 0.8409641799084299,
      "grad_norm": 0.7815103193784516,
      "learning_rate": 6.236625664354421e-06,
      "loss": 0.3829792499542236,
      "num_input_tokens_seen": 398246040,
      "step": 12490,
      "train_runtime": 32663.3273,
      "train_tokens_per_second": 12192.452
    },
    {
      "epoch": 0.84130083490439,
      "grad_norm": 0.7680839527275634,
      "learning_rate": 6.2340635543101545e-06,
      "loss": 0.35940754413604736,
      "num_input_tokens_seen": 398402216,
      "step": 12495,
      "train_runtime": 32676.2112,
      "train_tokens_per_second": 12192.424
    },
    {
      "epoch": 0.8416374899003501,
      "grad_norm": 0.7600370119133143,
      "learning_rate": 6.23150109916395e-06,
      "loss": 0.3576354026794434,
      "num_input_tokens_seen": 398562400,
      "step": 12500,
      "train_runtime": 32688.8958,
      "train_tokens_per_second": 12192.593
    },
    {
      "epoch": 0.8419741448963103,
      "grad_norm": 0.967728150217961,
      "learning_rate": 6.22893829963239e-06,
      "loss": 0.38012650012969973,
      "num_input_tokens_seen": 398720680,
      "step": 12505,
      "train_runtime": 32702.268,
      "train_tokens_per_second": 12192.447
    },
    {
      "epoch": 0.8423107998922704,
      "grad_norm": 0.8573841807457057,
      "learning_rate": 6.226375156432153e-06,
      "loss": 0.3830904245376587,
      "num_input_tokens_seen": 398883144,
      "step": 12510,
      "train_runtime": 32715.5969,
      "train_tokens_per_second": 12192.446
    },
    {
      "epoch": 0.8426474548882306,
      "grad_norm": 0.779343348897009,
      "learning_rate": 6.2238116702800145e-06,
      "loss": 0.37163257598876953,
      "num_input_tokens_seen": 399040224,
      "step": 12515,
      "train_runtime": 32729.1218,
      "train_tokens_per_second": 12192.207
    },
    {
      "epoch": 0.8429841098841907,
      "grad_norm": 0.8388240735122002,
      "learning_rate": 6.221247841892846e-06,
      "loss": 0.3885157346725464,
      "num_input_tokens_seen": 399195792,
      "step": 12520,
      "train_runtime": 32741.6564,
      "train_tokens_per_second": 12192.291
    },
    {
      "epoch": 0.8433207648801508,
      "grad_norm": 0.7763817090582618,
      "learning_rate": 6.218683671987611e-06,
      "loss": 0.3673069477081299,
      "num_input_tokens_seen": 399358328,
      "step": 12525,
      "train_runtime": 32754.3198,
      "train_tokens_per_second": 12192.539
    },
    {
      "epoch": 0.843657419876111,
      "grad_norm": 0.8125506760711246,
      "learning_rate": 6.216119161281373e-06,
      "loss": 0.37899010181427,
      "num_input_tokens_seen": 399518240,
      "step": 12530,
      "train_runtime": 32766.7832,
      "train_tokens_per_second": 12192.782
    },
    {
      "epoch": 0.8439940748720711,
      "grad_norm": 0.7264369828709771,
      "learning_rate": 6.213554310491291e-06,
      "loss": 0.3721353530883789,
      "num_input_tokens_seen": 399673560,
      "step": 12535,
      "train_runtime": 32779.7526,
      "train_tokens_per_second": 12192.696
    },
    {
      "epoch": 0.8443307298680313,
      "grad_norm": 0.6895907366753489,
      "learning_rate": 6.210989120334616e-06,
      "loss": 0.37443482875823975,
      "num_input_tokens_seen": 399831168,
      "step": 12540,
      "train_runtime": 32792.7824,
      "train_tokens_per_second": 12192.658
    },
    {
      "epoch": 0.8446673848639914,
      "grad_norm": 0.8937676714699526,
      "learning_rate": 6.208423591528694e-06,
      "loss": 0.39349913597106934,
      "num_input_tokens_seen": 399988176,
      "step": 12545,
      "train_runtime": 32805.4958,
      "train_tokens_per_second": 12192.719
    },
    {
      "epoch": 0.8450040398599515,
      "grad_norm": 0.7316902300296461,
      "learning_rate": 6.205857724790969e-06,
      "loss": 0.3690343379974365,
      "num_input_tokens_seen": 400144368,
      "step": 12550,
      "train_runtime": 32818.4211,
      "train_tokens_per_second": 12192.676
    },
    {
      "epoch": 0.8453406948559117,
      "grad_norm": 0.851807690881045,
      "learning_rate": 6.203291520838974e-06,
      "loss": 0.40696015357971194,
      "num_input_tokens_seen": 400307104,
      "step": 12555,
      "train_runtime": 32831.8307,
      "train_tokens_per_second": 12192.653
    },
    {
      "epoch": 0.8456773498518718,
      "grad_norm": 0.7130685225382905,
      "learning_rate": 6.200724980390344e-06,
      "loss": 0.3358847856521606,
      "num_input_tokens_seen": 400460520,
      "step": 12560,
      "train_runtime": 32845.8782,
      "train_tokens_per_second": 12192.109
    },
    {
      "epoch": 0.846014004847832,
      "grad_norm": 0.8268186235538687,
      "learning_rate": 6.198158104162804e-06,
      "loss": 0.32937169075012207,
      "num_input_tokens_seen": 400617512,
      "step": 12565,
      "train_runtime": 32859.2701,
      "train_tokens_per_second": 12191.918
    },
    {
      "epoch": 0.8463506598437921,
      "grad_norm": 0.7976638888606921,
      "learning_rate": 6.1955908928741685e-06,
      "loss": 0.3984895944595337,
      "num_input_tokens_seen": 400781352,
      "step": 12570,
      "train_runtime": 32871.6765,
      "train_tokens_per_second": 12192.3
    },
    {
      "epoch": 0.8466873148397522,
      "grad_norm": 0.7462138033530691,
      "learning_rate": 6.193023347242353e-06,
      "loss": 0.3751025676727295,
      "num_input_tokens_seen": 400943568,
      "step": 12575,
      "train_runtime": 32884.254,
      "train_tokens_per_second": 12192.57
    },
    {
      "epoch": 0.8470239698357124,
      "grad_norm": 0.8073717142308791,
      "learning_rate": 6.190455467985361e-06,
      "loss": 0.3694050073623657,
      "num_input_tokens_seen": 401103848,
      "step": 12580,
      "train_runtime": 32896.7843,
      "train_tokens_per_second": 12192.798
    },
    {
      "epoch": 0.8473606248316725,
      "grad_norm": 0.8302570849938541,
      "learning_rate": 6.187887255821295e-06,
      "loss": 0.40051612854003904,
      "num_input_tokens_seen": 401264768,
      "step": 12585,
      "train_runtime": 32909.7114,
      "train_tokens_per_second": 12192.898
    },
    {
      "epoch": 0.8476972798276327,
      "grad_norm": 0.9316038846522084,
      "learning_rate": 6.185318711468345e-06,
      "loss": 0.3508431434631348,
      "num_input_tokens_seen": 401417800,
      "step": 12590,
      "train_runtime": 32922.2671,
      "train_tokens_per_second": 12192.897
    },
    {
      "epoch": 0.8480339348235928,
      "grad_norm": 0.781890086152173,
      "learning_rate": 6.182749835644799e-06,
      "loss": 0.3576326370239258,
      "num_input_tokens_seen": 401572640,
      "step": 12595,
      "train_runtime": 32935.0293,
      "train_tokens_per_second": 12192.873
    },
    {
      "epoch": 0.8483705898195529,
      "grad_norm": 0.750843738984787,
      "learning_rate": 6.180180629069031e-06,
      "loss": 0.36127438545227053,
      "num_input_tokens_seen": 401736480,
      "step": 12600,
      "train_runtime": 32947.456,
      "train_tokens_per_second": 12193.247
    },
    {
      "epoch": 0.8487072448155131,
      "grad_norm": 0.7819741535922999,
      "learning_rate": 6.177611092459514e-06,
      "loss": 0.3495652675628662,
      "num_input_tokens_seen": 401893736,
      "step": 12605,
      "train_runtime": 32960.1075,
      "train_tokens_per_second": 12193.338
    },
    {
      "epoch": 0.8490438998114732,
      "grad_norm": 0.722893346123921,
      "learning_rate": 6.175041226534809e-06,
      "loss": 0.3683613300323486,
      "num_input_tokens_seen": 402056200,
      "step": 12610,
      "train_runtime": 32972.7637,
      "train_tokens_per_second": 12193.585
    },
    {
      "epoch": 0.8493805548074334,
      "grad_norm": 0.7742308438988654,
      "learning_rate": 6.172471032013574e-06,
      "loss": 0.3373131036758423,
      "num_input_tokens_seen": 402219256,
      "step": 12615,
      "train_runtime": 32985.6121,
      "train_tokens_per_second": 12193.779
    },
    {
      "epoch": 0.8497172098033935,
      "grad_norm": 0.6794174128467396,
      "learning_rate": 6.169900509614553e-06,
      "loss": 0.367539119720459,
      "num_input_tokens_seen": 402382456,
      "step": 12620,
      "train_runtime": 32999.2118,
      "train_tokens_per_second": 12193.699
    },
    {
      "epoch": 0.8500538647993536,
      "grad_norm": 0.8002691506605063,
      "learning_rate": 6.167329660056585e-06,
      "loss": 0.37078495025634767,
      "num_input_tokens_seen": 402541424,
      "step": 12625,
      "train_runtime": 33012.879,
      "train_tokens_per_second": 12193.466
    },
    {
      "epoch": 0.8503905197953138,
      "grad_norm": 0.6905025863939127,
      "learning_rate": 6.1647584840586e-06,
      "loss": 0.37767093181610106,
      "num_input_tokens_seen": 402703848,
      "step": 12630,
      "train_runtime": 33026.1611,
      "train_tokens_per_second": 12193.48
    },
    {
      "epoch": 0.8507271747912739,
      "grad_norm": 0.779807320390761,
      "learning_rate": 6.1621869823396175e-06,
      "loss": 0.36885714530944824,
      "num_input_tokens_seen": 402865168,
      "step": 12635,
      "train_runtime": 33038.6353,
      "train_tokens_per_second": 12193.759
    },
    {
      "epoch": 0.851063829787234,
      "grad_norm": 0.7601342008663453,
      "learning_rate": 6.159615155618752e-06,
      "loss": 0.3726487636566162,
      "num_input_tokens_seen": 403026192,
      "step": 12640,
      "train_runtime": 33051.1543,
      "train_tokens_per_second": 12194.013
    },
    {
      "epoch": 0.8514004847831942,
      "grad_norm": 0.8509473851267122,
      "learning_rate": 6.157043004615207e-06,
      "loss": 0.4008166313171387,
      "num_input_tokens_seen": 403190032,
      "step": 12645,
      "train_runtime": 33063.5762,
      "train_tokens_per_second": 12194.387
    },
    {
      "epoch": 0.8517371397791543,
      "grad_norm": 0.839165803676292,
      "learning_rate": 6.1544705300482755e-06,
      "loss": 0.39062521457672117,
      "num_input_tokens_seen": 403350784,
      "step": 12650,
      "train_runtime": 33076.6802,
      "train_tokens_per_second": 12194.416
    },
    {
      "epoch": 0.8520737947751145,
      "grad_norm": 0.7195594961981636,
      "learning_rate": 6.151897732637341e-06,
      "loss": 0.3450798988342285,
      "num_input_tokens_seen": 403508536,
      "step": 12655,
      "train_runtime": 33090.6945,
      "train_tokens_per_second": 12194.018
    },
    {
      "epoch": 0.8524104497710746,
      "grad_norm": 1.1397000959712602,
      "learning_rate": 6.149324613101877e-06,
      "loss": 0.37447295188903806,
      "num_input_tokens_seen": 403668736,
      "step": 12660,
      "train_runtime": 33103.5006,
      "train_tokens_per_second": 12194.14
    },
    {
      "epoch": 0.8527471047670347,
      "grad_norm": 0.7181606459255789,
      "learning_rate": 6.146751172161453e-06,
      "loss": 0.34419114589691163,
      "num_input_tokens_seen": 403828664,
      "step": 12665,
      "train_runtime": 33115.9592,
      "train_tokens_per_second": 12194.382
    },
    {
      "epoch": 0.8530837597629949,
      "grad_norm": 0.9039847988524667,
      "learning_rate": 6.144177410535719e-06,
      "loss": 0.3686807632446289,
      "num_input_tokens_seen": 403987816,
      "step": 12670,
      "train_runtime": 33128.9954,
      "train_tokens_per_second": 12194.388
    },
    {
      "epoch": 0.853420414758955,
      "grad_norm": 0.7687917755016143,
      "learning_rate": 6.14160332894442e-06,
      "loss": 0.3606849670410156,
      "num_input_tokens_seen": 404140960,
      "step": 12675,
      "train_runtime": 33142.3875,
      "train_tokens_per_second": 12194.081
    },
    {
      "epoch": 0.8537570697549152,
      "grad_norm": 0.8066960521776163,
      "learning_rate": 6.139028928107391e-06,
      "loss": 0.37674221992492674,
      "num_input_tokens_seen": 404302360,
      "step": 12680,
      "train_runtime": 33156.028,
      "train_tokens_per_second": 12193.932
    },
    {
      "epoch": 0.8540937247508753,
      "grad_norm": 0.7204606916110843,
      "learning_rate": 6.136454208744553e-06,
      "loss": 0.3510474920272827,
      "num_input_tokens_seen": 404462192,
      "step": 12685,
      "train_runtime": 33169.4945,
      "train_tokens_per_second": 12193.8
    },
    {
      "epoch": 0.8544303797468354,
      "grad_norm": 1.0080917081615282,
      "learning_rate": 6.133879171575921e-06,
      "loss": 0.37795443534851075,
      "num_input_tokens_seen": 404621512,
      "step": 12690,
      "train_runtime": 33182.3362,
      "train_tokens_per_second": 12193.883
    },
    {
      "epoch": 0.8547670347427956,
      "grad_norm": 0.7643254989413517,
      "learning_rate": 6.131303817321595e-06,
      "loss": 0.40749382972717285,
      "num_input_tokens_seen": 404785352,
      "step": 12695,
      "train_runtime": 33194.7633,
      "train_tokens_per_second": 12194.253
    },
    {
      "epoch": 0.8551036897387557,
      "grad_norm": 0.6664348025714469,
      "learning_rate": 6.128728146701763e-06,
      "loss": 0.3764413118362427,
      "num_input_tokens_seen": 404947808,
      "step": 12700,
      "train_runtime": 33208.2587,
      "train_tokens_per_second": 12194.19
    },
    {
      "epoch": 0.8554403447347159,
      "grad_norm": 0.6983900921830037,
      "learning_rate": 6.126152160436705e-06,
      "loss": 0.3326622247695923,
      "num_input_tokens_seen": 405110144,
      "step": 12705,
      "train_runtime": 33220.9893,
      "train_tokens_per_second": 12194.403
    },
    {
      "epoch": 0.855776999730676,
      "grad_norm": 0.9281329338517793,
      "learning_rate": 6.123575859246784e-06,
      "loss": 0.382525372505188,
      "num_input_tokens_seen": 405268424,
      "step": 12710,
      "train_runtime": 33233.7579,
      "train_tokens_per_second": 12194.481
    },
    {
      "epoch": 0.8561136547266361,
      "grad_norm": 0.8498839317386597,
      "learning_rate": 6.120999243852459e-06,
      "loss": 0.37018160820007323,
      "num_input_tokens_seen": 405426872,
      "step": 12715,
      "train_runtime": 33246.9018,
      "train_tokens_per_second": 12194.426
    },
    {
      "epoch": 0.8564503097225963,
      "grad_norm": 0.8928689419493613,
      "learning_rate": 6.118422314974269e-06,
      "loss": 0.39122605323791504,
      "num_input_tokens_seen": 405585688,
      "step": 12720,
      "train_runtime": 33259.834,
      "train_tokens_per_second": 12194.459
    },
    {
      "epoch": 0.8567869647185564,
      "grad_norm": 0.8352537355686774,
      "learning_rate": 6.115845073332845e-06,
      "loss": 0.3892014503479004,
      "num_input_tokens_seen": 405747808,
      "step": 12725,
      "train_runtime": 33273.0813,
      "train_tokens_per_second": 12194.477
    },
    {
      "epoch": 0.8571236197145166,
      "grad_norm": 0.8985030821020111,
      "learning_rate": 6.113267519648905e-06,
      "loss": 0.4044198036193848,
      "num_input_tokens_seen": 405911648,
      "step": 12730,
      "train_runtime": 33285.5058,
      "train_tokens_per_second": 12194.847
    },
    {
      "epoch": 0.8574602747104767,
      "grad_norm": 0.658944427605796,
      "learning_rate": 6.110689654643253e-06,
      "loss": 0.37032299041748046,
      "num_input_tokens_seen": 406069808,
      "step": 12735,
      "train_runtime": 33299.0048,
      "train_tokens_per_second": 12194.653
    },
    {
      "epoch": 0.8577969297064368,
      "grad_norm": 0.7314294668266916,
      "learning_rate": 6.1081114790367805e-06,
      "loss": 0.35985755920410156,
      "num_input_tokens_seen": 406225784,
      "step": 12740,
      "train_runtime": 33312.1797,
      "train_tokens_per_second": 12194.512
    },
    {
      "epoch": 0.858133584702397,
      "grad_norm": 0.7398811878050553,
      "learning_rate": 6.105532993550467e-06,
      "loss": 0.3731865882873535,
      "num_input_tokens_seen": 406387064,
      "step": 12745,
      "train_runtime": 33325.0991,
      "train_tokens_per_second": 12194.624
    },
    {
      "epoch": 0.8584702396983571,
      "grad_norm": 0.7378257625909477,
      "learning_rate": 6.1029541989053765e-06,
      "loss": 0.3603785276412964,
      "num_input_tokens_seen": 406549584,
      "step": 12750,
      "train_runtime": 33337.9649,
      "train_tokens_per_second": 12194.793
    },
    {
      "epoch": 0.8588068946943173,
      "grad_norm": 0.7450638904454566,
      "learning_rate": 6.100375095822661e-06,
      "loss": 0.39239749908447263,
      "num_input_tokens_seen": 406706312,
      "step": 12755,
      "train_runtime": 33350.6992,
      "train_tokens_per_second": 12194.836
    },
    {
      "epoch": 0.8591435496902774,
      "grad_norm": 0.7179052780595722,
      "learning_rate": 6.097795685023558e-06,
      "loss": 0.34895436763763427,
      "num_input_tokens_seen": 406865072,
      "step": 12760,
      "train_runtime": 33363.855,
      "train_tokens_per_second": 12194.786
    },
    {
      "epoch": 0.8594802046862375,
      "grad_norm": 0.7301341564039867,
      "learning_rate": 6.095215967229396e-06,
      "loss": 0.37278900146484373,
      "num_input_tokens_seen": 407020208,
      "step": 12765,
      "train_runtime": 33377.6106,
      "train_tokens_per_second": 12194.408
    },
    {
      "epoch": 0.8598168596821977,
      "grad_norm": 0.7692591032657112,
      "learning_rate": 6.092635943161578e-06,
      "loss": 0.3720006227493286,
      "num_input_tokens_seen": 407179696,
      "step": 12770,
      "train_runtime": 33390.5361,
      "train_tokens_per_second": 12194.464
    },
    {
      "epoch": 0.8601535146781578,
      "grad_norm": 0.7962386613929501,
      "learning_rate": 6.090055613541603e-06,
      "loss": 0.3699465751647949,
      "num_input_tokens_seen": 407342288,
      "step": 12775,
      "train_runtime": 33403.2121,
      "train_tokens_per_second": 12194.704
    },
    {
      "epoch": 0.860490169674118,
      "grad_norm": 0.8485093072788619,
      "learning_rate": 6.087474979091052e-06,
      "loss": 0.3725653886795044,
      "num_input_tokens_seen": 407498232,
      "step": 12780,
      "train_runtime": 33416.1338,
      "train_tokens_per_second": 12194.655
    },
    {
      "epoch": 0.8608268246700781,
      "grad_norm": 0.7860612783111367,
      "learning_rate": 6.084894040531591e-06,
      "loss": 0.3481574773788452,
      "num_input_tokens_seen": 407659744,
      "step": 12785,
      "train_runtime": 33428.6235,
      "train_tokens_per_second": 12194.931
    },
    {
      "epoch": 0.8611634796660382,
      "grad_norm": 0.7658199357483233,
      "learning_rate": 6.0823127985849705e-06,
      "loss": 0.3575266361236572,
      "num_input_tokens_seen": 407815360,
      "step": 12790,
      "train_runtime": 33441.9637,
      "train_tokens_per_second": 12194.719
    },
    {
      "epoch": 0.8615001346619984,
      "grad_norm": 0.77805466995071,
      "learning_rate": 6.079731253973028e-06,
      "loss": 0.3454300880432129,
      "num_input_tokens_seen": 407976112,
      "step": 12795,
      "train_runtime": 33455.3744,
      "train_tokens_per_second": 12194.636
    },
    {
      "epoch": 0.8618367896579585,
      "grad_norm": 0.7881419636748516,
      "learning_rate": 6.077149407417683e-06,
      "loss": 0.35955722332000734,
      "num_input_tokens_seen": 408134624,
      "step": 12800,
      "train_runtime": 33467.9949,
      "train_tokens_per_second": 12194.774
    },
    {
      "epoch": 0.8621734446539187,
      "grad_norm": 0.7557302069572392,
      "learning_rate": 6.07456725964094e-06,
      "loss": 0.377480673789978,
      "num_input_tokens_seen": 408293632,
      "step": 12805,
      "train_runtime": 33480.9384,
      "train_tokens_per_second": 12194.808
    },
    {
      "epoch": 0.8625100996498788,
      "grad_norm": 0.6942248951997614,
      "learning_rate": 6.07198481136489e-06,
      "loss": 0.34640040397644045,
      "num_input_tokens_seen": 408454688,
      "step": 12810,
      "train_runtime": 33495.47,
      "train_tokens_per_second": 12194.326
    },
    {
      "epoch": 0.8628467546458389,
      "grad_norm": 0.8195548991443865,
      "learning_rate": 6.069402063311706e-06,
      "loss": 0.36253116130828855,
      "num_input_tokens_seen": 408610896,
      "step": 12815,
      "train_runtime": 33508.3651,
      "train_tokens_per_second": 12194.295
    },
    {
      "epoch": 0.8631834096417991,
      "grad_norm": 0.8913361911505505,
      "learning_rate": 6.066819016203646e-06,
      "loss": 0.37452623844146726,
      "num_input_tokens_seen": 408768952,
      "step": 12820,
      "train_runtime": 33522.1167,
      "train_tokens_per_second": 12194.008
    },
    {
      "epoch": 0.8635200646377592,
      "grad_norm": 0.782206360214692,
      "learning_rate": 6.064235670763048e-06,
      "loss": 0.3726938724517822,
      "num_input_tokens_seen": 408930208,
      "step": 12825,
      "train_runtime": 33535.2514,
      "train_tokens_per_second": 12194.04
    },
    {
      "epoch": 0.8638567196337193,
      "grad_norm": 0.7699473020225349,
      "learning_rate": 6.06165202771234e-06,
      "loss": 0.40253872871398927,
      "num_input_tokens_seen": 409091504,
      "step": 12830,
      "train_runtime": 33548.3445,
      "train_tokens_per_second": 12194.089
    },
    {
      "epoch": 0.8641933746296795,
      "grad_norm": 0.6868305867696708,
      "learning_rate": 6.059068087774026e-06,
      "loss": 0.34599416255950927,
      "num_input_tokens_seen": 409249648,
      "step": 12835,
      "train_runtime": 33561.6571,
      "train_tokens_per_second": 12193.964
    },
    {
      "epoch": 0.8645300296256396,
      "grad_norm": 0.7689136943985069,
      "learning_rate": 6.0564838516707005e-06,
      "loss": 0.4018962860107422,
      "num_input_tokens_seen": 409412632,
      "step": 12840,
      "train_runtime": 33574.5644,
      "train_tokens_per_second": 12194.131
    },
    {
      "epoch": 0.8648666846215998,
      "grad_norm": 0.7139669760122431,
      "learning_rate": 6.053899320125033e-06,
      "loss": 0.3490764141082764,
      "num_input_tokens_seen": 409573872,
      "step": 12845,
      "train_runtime": 33588.1341,
      "train_tokens_per_second": 12194.005
    },
    {
      "epoch": 0.8652033396175599,
      "grad_norm": 0.8518848234602384,
      "learning_rate": 6.051314493859782e-06,
      "loss": 0.38599472045898436,
      "num_input_tokens_seen": 409735800,
      "step": 12850,
      "train_runtime": 33601.184,
      "train_tokens_per_second": 12194.088
    },
    {
      "epoch": 0.86553999461352,
      "grad_norm": 0.6950213087304614,
      "learning_rate": 6.048729373597786e-06,
      "loss": 0.36157453060150146,
      "num_input_tokens_seen": 409897232,
      "step": 12855,
      "train_runtime": 33615.2059,
      "train_tokens_per_second": 12193.804
    },
    {
      "epoch": 0.8658766496094802,
      "grad_norm": 0.8002476802050338,
      "learning_rate": 6.046143960061963e-06,
      "loss": 0.3891181468963623,
      "num_input_tokens_seen": 410056144,
      "step": 12860,
      "train_runtime": 33628.1962,
      "train_tokens_per_second": 12193.819
    },
    {
      "epoch": 0.8662133046054403,
      "grad_norm": 0.8918912706460341,
      "learning_rate": 6.043558253975319e-06,
      "loss": 0.3897066354751587,
      "num_input_tokens_seen": 410218256,
      "step": 12865,
      "train_runtime": 33641.3634,
      "train_tokens_per_second": 12193.865
    },
    {
      "epoch": 0.8665499596014005,
      "grad_norm": 0.7729623391622175,
      "learning_rate": 6.040972256060936e-06,
      "loss": 0.3626770734786987,
      "num_input_tokens_seen": 410376472,
      "step": 12870,
      "train_runtime": 33654.9437,
      "train_tokens_per_second": 12193.646
    },
    {
      "epoch": 0.8668866145973606,
      "grad_norm": 0.6766534276158708,
      "learning_rate": 6.038385967041982e-06,
      "loss": 0.36469197273254395,
      "num_input_tokens_seen": 410534264,
      "step": 12875,
      "train_runtime": 33668.4176,
      "train_tokens_per_second": 12193.453
    },
    {
      "epoch": 0.8672232695933207,
      "grad_norm": 0.7032803539437972,
      "learning_rate": 6.035799387641703e-06,
      "loss": 0.38130509853363037,
      "num_input_tokens_seen": 410694576,
      "step": 12880,
      "train_runtime": 33680.8215,
      "train_tokens_per_second": 12193.722
    },
    {
      "epoch": 0.8675599245892809,
      "grad_norm": 0.7762694515631788,
      "learning_rate": 6.033212518583427e-06,
      "loss": 0.38940274715423584,
      "num_input_tokens_seen": 410853008,
      "step": 12885,
      "train_runtime": 33693.3584,
      "train_tokens_per_second": 12193.887
    },
    {
      "epoch": 0.867896579585241,
      "grad_norm": 0.8474303988806787,
      "learning_rate": 6.030625360590567e-06,
      "loss": 0.37598414421081544,
      "num_input_tokens_seen": 411011920,
      "step": 12890,
      "train_runtime": 33705.8183,
      "train_tokens_per_second": 12194.094
    },
    {
      "epoch": 0.8682332345812012,
      "grad_norm": 0.7087498583601097,
      "learning_rate": 6.028037914386609e-06,
      "loss": 0.3651896953582764,
      "num_input_tokens_seen": 411172416,
      "step": 12895,
      "train_runtime": 33719.5992,
      "train_tokens_per_second": 12193.87
    },
    {
      "epoch": 0.8685698895771613,
      "grad_norm": 0.8373781193441389,
      "learning_rate": 6.025450180695128e-06,
      "loss": 0.39384140968322756,
      "num_input_tokens_seen": 411335176,
      "step": 12900,
      "train_runtime": 33733.031,
      "train_tokens_per_second": 12193.84
    },
    {
      "epoch": 0.8689065445731214,
      "grad_norm": 0.8052785411699693,
      "learning_rate": 6.022862160239774e-06,
      "loss": 0.3569489002227783,
      "num_input_tokens_seen": 411495936,
      "step": 12905,
      "train_runtime": 33745.9405,
      "train_tokens_per_second": 12193.939
    },
    {
      "epoch": 0.8692431995690816,
      "grad_norm": 0.6918394011509618,
      "learning_rate": 6.020273853744277e-06,
      "loss": 0.35686182975769043,
      "num_input_tokens_seen": 411655640,
      "step": 12910,
      "train_runtime": 33759.0314,
      "train_tokens_per_second": 12193.941
    },
    {
      "epoch": 0.8695798545650417,
      "grad_norm": 0.7458713683528733,
      "learning_rate": 6.017685261932452e-06,
      "loss": 0.37008347511291506,
      "num_input_tokens_seen": 411803848,
      "step": 12915,
      "train_runtime": 33771.6074,
      "train_tokens_per_second": 12193.789
    },
    {
      "epoch": 0.8699165095610019,
      "grad_norm": 0.7805079347128676,
      "learning_rate": 6.015096385528189e-06,
      "loss": 0.37687828540802004,
      "num_input_tokens_seen": 411962496,
      "step": 12920,
      "train_runtime": 33785.3897,
      "train_tokens_per_second": 12193.51
    },
    {
      "epoch": 0.870253164556962,
      "grad_norm": 0.7288989433804928,
      "learning_rate": 6.012507225255457e-06,
      "loss": 0.38899846076965333,
      "num_input_tokens_seen": 412114792,
      "step": 12925,
      "train_runtime": 33798.9921,
      "train_tokens_per_second": 12193.109
    },
    {
      "epoch": 0.8705898195529221,
      "grad_norm": 0.7233373226940525,
      "learning_rate": 6.009917781838309e-06,
      "loss": 0.3851949691772461,
      "num_input_tokens_seen": 412273624,
      "step": 12930,
      "train_runtime": 33812.6646,
      "train_tokens_per_second": 12192.876
    },
    {
      "epoch": 0.8709264745488823,
      "grad_norm": 0.7173827686166462,
      "learning_rate": 6.007328056000873e-06,
      "loss": 0.3766153573989868,
      "num_input_tokens_seen": 412433736,
      "step": 12935,
      "train_runtime": 33825.6316,
      "train_tokens_per_second": 12192.935
    },
    {
      "epoch": 0.8712631295448424,
      "grad_norm": 0.8833430194303291,
      "learning_rate": 6.004738048467359e-06,
      "loss": 0.3949904203414917,
      "num_input_tokens_seen": 412596208,
      "step": 12940,
      "train_runtime": 33838.9673,
      "train_tokens_per_second": 12192.931
    },
    {
      "epoch": 0.8715997845408026,
      "grad_norm": 0.7593578933751526,
      "learning_rate": 6.002147759962054e-06,
      "loss": 0.3708235263824463,
      "num_input_tokens_seen": 412757856,
      "step": 12945,
      "train_runtime": 33852.1298,
      "train_tokens_per_second": 12192.966
    },
    {
      "epoch": 0.8719364395367627,
      "grad_norm": 0.977158902305095,
      "learning_rate": 5.999557191209323e-06,
      "loss": 0.39374637603759766,
      "num_input_tokens_seen": 412921696,
      "step": 12950,
      "train_runtime": 33864.5388,
      "train_tokens_per_second": 12193.336
    },
    {
      "epoch": 0.8722730945327228,
      "grad_norm": 0.792808958727639,
      "learning_rate": 5.996966342933611e-06,
      "loss": 0.3341085433959961,
      "num_input_tokens_seen": 413083192,
      "step": 12955,
      "train_runtime": 33877.59,
      "train_tokens_per_second": 12193.405
    },
    {
      "epoch": 0.872609749528683,
      "grad_norm": 0.7928487673183624,
      "learning_rate": 5.99437521585944e-06,
      "loss": 0.37221875190734866,
      "num_input_tokens_seen": 413240568,
      "step": 12960,
      "train_runtime": 33890.7697,
      "train_tokens_per_second": 12193.307
    },
    {
      "epoch": 0.8729464045246431,
      "grad_norm": 0.7959131200819058,
      "learning_rate": 5.99178381071141e-06,
      "loss": 0.3860820531845093,
      "num_input_tokens_seen": 413398952,
      "step": 12965,
      "train_runtime": 33903.3226,
      "train_tokens_per_second": 12193.464
    },
    {
      "epoch": 0.8732830595206033,
      "grad_norm": 0.8411813990718096,
      "learning_rate": 5.989192128214201e-06,
      "loss": 0.3725898265838623,
      "num_input_tokens_seen": 413555456,
      "step": 12970,
      "train_runtime": 33916.2722,
      "train_tokens_per_second": 12193.423
    },
    {
      "epoch": 0.8736197145165634,
      "grad_norm": 0.8261820749612145,
      "learning_rate": 5.986600169092568e-06,
      "loss": 0.36458826065063477,
      "num_input_tokens_seen": 413715760,
      "step": 12975,
      "train_runtime": 33929.4407,
      "train_tokens_per_second": 12193.415
    },
    {
      "epoch": 0.8739563695125235,
      "grad_norm": 0.808581678931111,
      "learning_rate": 5.984007934071343e-06,
      "loss": 0.39312138557434084,
      "num_input_tokens_seen": 413879600,
      "step": 12980,
      "train_runtime": 33941.8605,
      "train_tokens_per_second": 12193.781
    },
    {
      "epoch": 0.8742930245084837,
      "grad_norm": 0.7669903965127057,
      "learning_rate": 5.981415423875436e-06,
      "loss": 0.3765831708908081,
      "num_input_tokens_seen": 414040488,
      "step": 12985,
      "train_runtime": 33954.3008,
      "train_tokens_per_second": 12194.051
    },
    {
      "epoch": 0.8746296795044438,
      "grad_norm": 0.874873818693881,
      "learning_rate": 5.978822639229836e-06,
      "loss": 0.36043601036071776,
      "num_input_tokens_seen": 414192888,
      "step": 12990,
      "train_runtime": 33967.2873,
      "train_tokens_per_second": 12193.876
    },
    {
      "epoch": 0.874966334500404,
      "grad_norm": 0.7834550704306165,
      "learning_rate": 5.976229580859607e-06,
      "loss": 0.3817555904388428,
      "num_input_tokens_seen": 414354168,
      "step": 12995,
      "train_runtime": 33980.3661,
      "train_tokens_per_second": 12193.929
    },
    {
      "epoch": 0.8753029894963641,
      "grad_norm": 0.7948978778888021,
      "learning_rate": 5.9736362494898895e-06,
      "loss": 0.3524010181427002,
      "num_input_tokens_seen": 414512088,
      "step": 13000,
      "train_runtime": 33993.6209,
      "train_tokens_per_second": 12193.82
    },
    {
      "epoch": 0.8756396444923242,
      "grad_norm": 0.7408346475228127,
      "learning_rate": 5.971042645845897e-06,
      "loss": 0.37934682369232176,
      "num_input_tokens_seen": 414670616,
      "step": 13005,
      "train_runtime": 34006.3425,
      "train_tokens_per_second": 12193.92
    },
    {
      "epoch": 0.8759762994882844,
      "grad_norm": 0.7493180498429644,
      "learning_rate": 5.968448770652926e-06,
      "loss": 0.36423530578613283,
      "num_input_tokens_seen": 414832728,
      "step": 13010,
      "train_runtime": 34019.3361,
      "train_tokens_per_second": 12194.028
    },
    {
      "epoch": 0.8763129544842445,
      "grad_norm": 0.8020445034511668,
      "learning_rate": 5.965854624636345e-06,
      "loss": 0.40179805755615233,
      "num_input_tokens_seen": 414987776,
      "step": 13015,
      "train_runtime": 34032.7217,
      "train_tokens_per_second": 12193.787
    },
    {
      "epoch": 0.8766496094802047,
      "grad_norm": 0.8162241297318947,
      "learning_rate": 5.963260208521598e-06,
      "loss": 0.37693307399749754,
      "num_input_tokens_seen": 415145824,
      "step": 13020,
      "train_runtime": 34045.743,
      "train_tokens_per_second": 12193.766
    },
    {
      "epoch": 0.8769862644761648,
      "grad_norm": 0.7712413820992793,
      "learning_rate": 5.960665523034205e-06,
      "loss": 0.3809064388275146,
      "num_input_tokens_seen": 415301600,
      "step": 13025,
      "train_runtime": 34058.7228,
      "train_tokens_per_second": 12193.693
    },
    {
      "epoch": 0.8773229194721249,
      "grad_norm": 1.0377289687679634,
      "learning_rate": 5.958070568899762e-06,
      "loss": 0.3460919618606567,
      "num_input_tokens_seen": 415465440,
      "step": 13030,
      "train_runtime": 34071.3476,
      "train_tokens_per_second": 12193.983
    },
    {
      "epoch": 0.8776595744680851,
      "grad_norm": 0.7689126908297821,
      "learning_rate": 5.955475346843939e-06,
      "loss": 0.3675867557525635,
      "num_input_tokens_seen": 415618288,
      "step": 13035,
      "train_runtime": 34084.0033,
      "train_tokens_per_second": 12193.94
    },
    {
      "epoch": 0.8779962294640452,
      "grad_norm": 0.7433218130193447,
      "learning_rate": 5.952879857592482e-06,
      "loss": 0.3404216766357422,
      "num_input_tokens_seen": 415776832,
      "step": 13040,
      "train_runtime": 34097.5085,
      "train_tokens_per_second": 12193.76
    },
    {
      "epoch": 0.8783328844600053,
      "grad_norm": 0.7903239259849757,
      "learning_rate": 5.950284101871212e-06,
      "loss": 0.39055964946746824,
      "num_input_tokens_seen": 415932720,
      "step": 13045,
      "train_runtime": 34110.8904,
      "train_tokens_per_second": 12193.546
    },
    {
      "epoch": 0.8786695394559655,
      "grad_norm": 0.7705301724281072,
      "learning_rate": 5.947688080406023e-06,
      "loss": 0.3599937915802002,
      "num_input_tokens_seen": 416088168,
      "step": 13050,
      "train_runtime": 34124.608,
      "train_tokens_per_second": 12193.2
    },
    {
      "epoch": 0.8790061944519256,
      "grad_norm": 0.7624602425386903,
      "learning_rate": 5.945091793922885e-06,
      "loss": 0.3516396999359131,
      "num_input_tokens_seen": 416246624,
      "step": 13055,
      "train_runtime": 34138.3357,
      "train_tokens_per_second": 12192.938
    },
    {
      "epoch": 0.8793428494478858,
      "grad_norm": 0.7815443147904304,
      "learning_rate": 5.942495243147839e-06,
      "loss": 0.4112947940826416,
      "num_input_tokens_seen": 416405152,
      "step": 13060,
      "train_runtime": 34151.175,
      "train_tokens_per_second": 12192.996
    },
    {
      "epoch": 0.8796795044438459,
      "grad_norm": 0.7263386803421994,
      "learning_rate": 5.939898428807004e-06,
      "loss": 0.3809444189071655,
      "num_input_tokens_seen": 416564968,
      "step": 13065,
      "train_runtime": 34163.9252,
      "train_tokens_per_second": 12193.124
    },
    {
      "epoch": 0.880016159439806,
      "grad_norm": 0.8454893587056186,
      "learning_rate": 5.937301351626571e-06,
      "loss": 0.3815094709396362,
      "num_input_tokens_seen": 416723320,
      "step": 13070,
      "train_runtime": 34176.8969,
      "train_tokens_per_second": 12193.129
    },
    {
      "epoch": 0.8803528144357662,
      "grad_norm": 0.776536709943707,
      "learning_rate": 5.934704012332803e-06,
      "loss": 0.3818216323852539,
      "num_input_tokens_seen": 416885952,
      "step": 13075,
      "train_runtime": 34190.2846,
      "train_tokens_per_second": 12193.112
    },
    {
      "epoch": 0.8806894694317263,
      "grad_norm": 0.8626439121563402,
      "learning_rate": 5.93210641165204e-06,
      "loss": 0.41301589012145995,
      "num_input_tokens_seen": 417048096,
      "step": 13080,
      "train_runtime": 34203.47,
      "train_tokens_per_second": 12193.152
    },
    {
      "epoch": 0.8810261244276865,
      "grad_norm": 0.9674089686262991,
      "learning_rate": 5.92950855031069e-06,
      "loss": 0.38540401458740237,
      "num_input_tokens_seen": 417208088,
      "step": 13085,
      "train_runtime": 34216.7544,
      "train_tokens_per_second": 12193.094
    },
    {
      "epoch": 0.8813627794236466,
      "grad_norm": 0.7668980878724503,
      "learning_rate": 5.926910429035239e-06,
      "loss": 0.3853085994720459,
      "num_input_tokens_seen": 417364864,
      "step": 13090,
      "train_runtime": 34229.4106,
      "train_tokens_per_second": 12193.165
    },
    {
      "epoch": 0.8816994344196067,
      "grad_norm": 0.7280443921582831,
      "learning_rate": 5.924312048552241e-06,
      "loss": 0.36177754402160645,
      "num_input_tokens_seen": 417525448,
      "step": 13095,
      "train_runtime": 34242.4949,
      "train_tokens_per_second": 12193.196
    },
    {
      "epoch": 0.8820360894155669,
      "grad_norm": 2.261440141786363,
      "learning_rate": 5.9217134095883265e-06,
      "loss": 0.3674927234649658,
      "num_input_tokens_seen": 417689288,
      "step": 13100,
      "train_runtime": 34254.9196,
      "train_tokens_per_second": 12193.556
    },
    {
      "epoch": 0.882372744411527,
      "grad_norm": 0.8097265428499648,
      "learning_rate": 5.919114512870197e-06,
      "loss": 0.3579556941986084,
      "num_input_tokens_seen": 417852832,
      "step": 13105,
      "train_runtime": 34267.9309,
      "train_tokens_per_second": 12193.699
    },
    {
      "epoch": 0.8827093994074872,
      "grad_norm": 0.7486064152000329,
      "learning_rate": 5.916515359124623e-06,
      "loss": 0.35618577003479,
      "num_input_tokens_seen": 418007328,
      "step": 13110,
      "train_runtime": 34280.9474,
      "train_tokens_per_second": 12193.576
    },
    {
      "epoch": 0.8830460544034473,
      "grad_norm": 0.8119623364837927,
      "learning_rate": 5.913915949078453e-06,
      "loss": 0.37612075805664064,
      "num_input_tokens_seen": 418168656,
      "step": 13115,
      "train_runtime": 34293.8792,
      "train_tokens_per_second": 12193.682
    },
    {
      "epoch": 0.8833827093994074,
      "grad_norm": 0.8147606284926963,
      "learning_rate": 5.911316283458601e-06,
      "loss": 0.36575419902801515,
      "num_input_tokens_seen": 418326656,
      "step": 13120,
      "train_runtime": 34308.0516,
      "train_tokens_per_second": 12193.25
    },
    {
      "epoch": 0.8837193643953676,
      "grad_norm": 0.7533740176987012,
      "learning_rate": 5.908716362992057e-06,
      "loss": 0.3725563049316406,
      "num_input_tokens_seen": 418482984,
      "step": 13125,
      "train_runtime": 34320.946,
      "train_tokens_per_second": 12193.224
    },
    {
      "epoch": 0.8840560193913277,
      "grad_norm": 0.7424864211809662,
      "learning_rate": 5.90611618840588e-06,
      "loss": 0.38197789192199705,
      "num_input_tokens_seen": 418642768,
      "step": 13130,
      "train_runtime": 34333.3443,
      "train_tokens_per_second": 12193.475
    },
    {
      "epoch": 0.8843926743872879,
      "grad_norm": 0.8235444028639038,
      "learning_rate": 5.903515760427198e-06,
      "loss": 0.3972452163696289,
      "num_input_tokens_seen": 418800520,
      "step": 13135,
      "train_runtime": 34346.3685,
      "train_tokens_per_second": 12193.444
    },
    {
      "epoch": 0.884729329383248,
      "grad_norm": 0.8248324207654019,
      "learning_rate": 5.900915079783218e-06,
      "loss": 0.36539907455444337,
      "num_input_tokens_seen": 418963816,
      "step": 13140,
      "train_runtime": 34359.2617,
      "train_tokens_per_second": 12193.621
    },
    {
      "epoch": 0.8850659843792081,
      "grad_norm": 0.8342316226053779,
      "learning_rate": 5.898314147201207e-06,
      "loss": 0.3543950319290161,
      "num_input_tokens_seen": 419119120,
      "step": 13145,
      "train_runtime": 34372.7761,
      "train_tokens_per_second": 12193.345
    },
    {
      "epoch": 0.8854026393751683,
      "grad_norm": 0.9497082491364792,
      "learning_rate": 5.895712963408511e-06,
      "loss": 0.4082606315612793,
      "num_input_tokens_seen": 419278560,
      "step": 13150,
      "train_runtime": 34386.7536,
      "train_tokens_per_second": 12193.025
    },
    {
      "epoch": 0.8857392943711284,
      "grad_norm": 0.8149686642704281,
      "learning_rate": 5.893111529132539e-06,
      "loss": 0.4085671901702881,
      "num_input_tokens_seen": 419433680,
      "step": 13155,
      "train_runtime": 34399.8527,
      "train_tokens_per_second": 12192.892
    },
    {
      "epoch": 0.8860759493670886,
      "grad_norm": 0.7998442546583407,
      "learning_rate": 5.8905098451007755e-06,
      "loss": 0.3963483810424805,
      "num_input_tokens_seen": 419591312,
      "step": 13160,
      "train_runtime": 34413.3568,
      "train_tokens_per_second": 12192.688
    },
    {
      "epoch": 0.8864126043630487,
      "grad_norm": 0.761298875678993,
      "learning_rate": 5.887907912040774e-06,
      "loss": 0.3726170539855957,
      "num_input_tokens_seen": 419747280,
      "step": 13165,
      "train_runtime": 34425.8225,
      "train_tokens_per_second": 12192.803
    },
    {
      "epoch": 0.8867492593590088,
      "grad_norm": 0.8493023219785234,
      "learning_rate": 5.885305730680158e-06,
      "loss": 0.3757298469543457,
      "num_input_tokens_seen": 419909128,
      "step": 13170,
      "train_runtime": 34438.3494,
      "train_tokens_per_second": 12193.068
    },
    {
      "epoch": 0.887085914354969,
      "grad_norm": 0.733066963399798,
      "learning_rate": 5.882703301746616e-06,
      "loss": 0.3611739635467529,
      "num_input_tokens_seen": 420064800,
      "step": 13175,
      "train_runtime": 34450.8689,
      "train_tokens_per_second": 12193.155
    },
    {
      "epoch": 0.8874225693509291,
      "grad_norm": 0.7543688289110454,
      "learning_rate": 5.88010062596791e-06,
      "loss": 0.35973658561706545,
      "num_input_tokens_seen": 420223928,
      "step": 13180,
      "train_runtime": 34463.9725,
      "train_tokens_per_second": 12193.137
    },
    {
      "epoch": 0.8877592243468893,
      "grad_norm": 0.7161801494945618,
      "learning_rate": 5.877497704071871e-06,
      "loss": 0.345694899559021,
      "num_input_tokens_seen": 420385552,
      "step": 13185,
      "train_runtime": 34477.1725,
      "train_tokens_per_second": 12193.156
    },
    {
      "epoch": 0.8880958793428494,
      "grad_norm": 0.7665959576898872,
      "learning_rate": 5.874894536786396e-06,
      "loss": 0.33997349739074706,
      "num_input_tokens_seen": 420543264,
      "step": 13190,
      "train_runtime": 34490.3552,
      "train_tokens_per_second": 12193.069
    },
    {
      "epoch": 0.8884325343388096,
      "grad_norm": 0.7291767394288539,
      "learning_rate": 5.8722911248394555e-06,
      "loss": 0.35655348300933837,
      "num_input_tokens_seen": 420691600,
      "step": 13195,
      "train_runtime": 34503.9133,
      "train_tokens_per_second": 12192.576
    },
    {
      "epoch": 0.8887691893347698,
      "grad_norm": 0.7936495104352435,
      "learning_rate": 5.8696874689590824e-06,
      "loss": 0.34440979957580564,
      "num_input_tokens_seen": 420847856,
      "step": 13200,
      "train_runtime": 34516.8749,
      "train_tokens_per_second": 12192.525
    },
    {
      "epoch": 0.8891058443307299,
      "grad_norm": 0.8796854926201408,
      "learning_rate": 5.867083569873383e-06,
      "loss": 0.3730029106140137,
      "num_input_tokens_seen": 421008608,
      "step": 13205,
      "train_runtime": 34529.909,
      "train_tokens_per_second": 12192.578
    },
    {
      "epoch": 0.8894424993266901,
      "grad_norm": 0.8482665946134398,
      "learning_rate": 5.8644794283105255e-06,
      "loss": 0.36316943168640137,
      "num_input_tokens_seen": 421168040,
      "step": 13210,
      "train_runtime": 34543.413,
      "train_tokens_per_second": 12192.427
    },
    {
      "epoch": 0.8897791543226502,
      "grad_norm": 0.7923660818371361,
      "learning_rate": 5.861875044998755e-06,
      "loss": 0.4084757328033447,
      "num_input_tokens_seen": 421327608,
      "step": 13215,
      "train_runtime": 34556.4132,
      "train_tokens_per_second": 12192.458
    },
    {
      "epoch": 0.8901158093186103,
      "grad_norm": 0.8396418544847153,
      "learning_rate": 5.859270420666377e-06,
      "loss": 0.37616701126098634,
      "num_input_tokens_seen": 421486288,
      "step": 13220,
      "train_runtime": 34568.9584,
      "train_tokens_per_second": 12192.623
    },
    {
      "epoch": 0.8904524643145705,
      "grad_norm": 0.758085318159808,
      "learning_rate": 5.856665556041764e-06,
      "loss": 0.3437247037887573,
      "num_input_tokens_seen": 421640568,
      "step": 13225,
      "train_runtime": 34582.0221,
      "train_tokens_per_second": 12192.479
    },
    {
      "epoch": 0.8907891193105306,
      "grad_norm": 0.7865276082622846,
      "learning_rate": 5.8540604518533605e-06,
      "loss": 0.40551133155822755,
      "num_input_tokens_seen": 421803496,
      "step": 13230,
      "train_runtime": 34595.496,
      "train_tokens_per_second": 12192.44
    },
    {
      "epoch": 0.8911257743064908,
      "grad_norm": 0.8254416895174947,
      "learning_rate": 5.851455108829675e-06,
      "loss": 0.36055431365966795,
      "num_input_tokens_seen": 421959376,
      "step": 13235,
      "train_runtime": 34608.3605,
      "train_tokens_per_second": 12192.412
    },
    {
      "epoch": 0.8914624293024509,
      "grad_norm": 0.7907894831125426,
      "learning_rate": 5.848849527699283e-06,
      "loss": 0.3840185642242432,
      "num_input_tokens_seen": 422116888,
      "step": 13240,
      "train_runtime": 34621.7384,
      "train_tokens_per_second": 12192.25
    },
    {
      "epoch": 0.891799084298411,
      "grad_norm": 0.8520052004501267,
      "learning_rate": 5.846243709190827e-06,
      "loss": 0.3593330383300781,
      "num_input_tokens_seen": 422277392,
      "step": 13245,
      "train_runtime": 34635.1367,
      "train_tokens_per_second": 12192.168
    },
    {
      "epoch": 0.8921357392943712,
      "grad_norm": 0.8784061305526504,
      "learning_rate": 5.843637654033019e-06,
      "loss": 0.38776819705963134,
      "num_input_tokens_seen": 422432704,
      "step": 13250,
      "train_runtime": 34649.1972,
      "train_tokens_per_second": 12191.702
    },
    {
      "epoch": 0.8924723942903313,
      "grad_norm": 0.7548385463443169,
      "learning_rate": 5.841031362954629e-06,
      "loss": 0.39724297523498536,
      "num_input_tokens_seen": 422593264,
      "step": 13255,
      "train_runtime": 34662.2209,
      "train_tokens_per_second": 12191.754
    },
    {
      "epoch": 0.8928090492862915,
      "grad_norm": 0.741688836475473,
      "learning_rate": 5.8384248366845e-06,
      "loss": 0.385623836517334,
      "num_input_tokens_seen": 422753056,
      "step": 13260,
      "train_runtime": 34675.3123,
      "train_tokens_per_second": 12191.759
    },
    {
      "epoch": 0.8931457042822516,
      "grad_norm": 0.737776550227351,
      "learning_rate": 5.8358180759515396e-06,
      "loss": 0.3823493480682373,
      "num_input_tokens_seen": 422911352,
      "step": 13265,
      "train_runtime": 34688.5118,
      "train_tokens_per_second": 12191.683
    },
    {
      "epoch": 0.8934823592782117,
      "grad_norm": 0.8057927040258587,
      "learning_rate": 5.833211081484719e-06,
      "loss": 0.3861107587814331,
      "num_input_tokens_seen": 423067592,
      "step": 13270,
      "train_runtime": 34701.1427,
      "train_tokens_per_second": 12191.748
    },
    {
      "epoch": 0.8938190142741719,
      "grad_norm": 1.0800560287480332,
      "learning_rate": 5.830603854013077e-06,
      "loss": 0.3815847635269165,
      "num_input_tokens_seen": 423226848,
      "step": 13275,
      "train_runtime": 34714.4814,
      "train_tokens_per_second": 12191.651
    },
    {
      "epoch": 0.894155669270132,
      "grad_norm": 0.8122470031767568,
      "learning_rate": 5.827996394265716e-06,
      "loss": 0.3494251251220703,
      "num_input_tokens_seen": 423383168,
      "step": 13280,
      "train_runtime": 34727.742,
      "train_tokens_per_second": 12191.497
    },
    {
      "epoch": 0.8944923242660922,
      "grad_norm": 0.7649105753596509,
      "learning_rate": 5.825388702971802e-06,
      "loss": 0.3518451452255249,
      "num_input_tokens_seen": 423543040,
      "step": 13285,
      "train_runtime": 34741.4488,
      "train_tokens_per_second": 12191.289
    },
    {
      "epoch": 0.8948289792620523,
      "grad_norm": 2.3009187415826093,
      "learning_rate": 5.822780780860571e-06,
      "loss": 0.38631153106689453,
      "num_input_tokens_seen": 423701712,
      "step": 13290,
      "train_runtime": 34754.9864,
      "train_tokens_per_second": 12191.106
    },
    {
      "epoch": 0.8951656342580124,
      "grad_norm": 0.7778464229800114,
      "learning_rate": 5.820172628661317e-06,
      "loss": 0.3693638563156128,
      "num_input_tokens_seen": 423864912,
      "step": 13295,
      "train_runtime": 34767.792,
      "train_tokens_per_second": 12191.309
    },
    {
      "epoch": 0.8955022892539726,
      "grad_norm": 0.7634265641248088,
      "learning_rate": 5.8175642471034045e-06,
      "loss": 0.3683186054229736,
      "num_input_tokens_seen": 424025016,
      "step": 13300,
      "train_runtime": 34780.4805,
      "train_tokens_per_second": 12191.465
    },
    {
      "epoch": 0.8958389442499327,
      "grad_norm": 0.8892845495152538,
      "learning_rate": 5.8149556369162565e-06,
      "loss": 0.3742434740066528,
      "num_input_tokens_seen": 424178288,
      "step": 13305,
      "train_runtime": 34793.1428,
      "train_tokens_per_second": 12191.434
    },
    {
      "epoch": 0.8961755992458929,
      "grad_norm": 0.8154867999024771,
      "learning_rate": 5.812346798829361e-06,
      "loss": 0.39190590381622314,
      "num_input_tokens_seen": 424338760,
      "step": 13310,
      "train_runtime": 34806.1804,
      "train_tokens_per_second": 12191.477
    },
    {
      "epoch": 0.896512254241853,
      "grad_norm": 0.719550354234362,
      "learning_rate": 5.809737733572276e-06,
      "loss": 0.34295594692230225,
      "num_input_tokens_seen": 424496088,
      "step": 13315,
      "train_runtime": 34819.6975,
      "train_tokens_per_second": 12191.263
    },
    {
      "epoch": 0.8968489092378131,
      "grad_norm": 0.8544400335941631,
      "learning_rate": 5.8071284418746166e-06,
      "loss": 0.3649880886077881,
      "num_input_tokens_seen": 424655168,
      "step": 13320,
      "train_runtime": 34832.2818,
      "train_tokens_per_second": 12191.425
    },
    {
      "epoch": 0.8971855642337733,
      "grad_norm": 0.756172677011678,
      "learning_rate": 5.80451892446606e-06,
      "loss": 0.3966256856918335,
      "num_input_tokens_seen": 424818296,
      "step": 13325,
      "train_runtime": 34845.145,
      "train_tokens_per_second": 12191.607
    },
    {
      "epoch": 0.8975222192297334,
      "grad_norm": 0.7021946840319796,
      "learning_rate": 5.801909182076354e-06,
      "loss": 0.3582056999206543,
      "num_input_tokens_seen": 424977920,
      "step": 13330,
      "train_runtime": 34857.789,
      "train_tokens_per_second": 12191.763
    },
    {
      "epoch": 0.8978588742256935,
      "grad_norm": 0.8159392595911944,
      "learning_rate": 5.7992992154353e-06,
      "loss": 0.39885869026184084,
      "num_input_tokens_seen": 425135944,
      "step": 13335,
      "train_runtime": 34870.4793,
      "train_tokens_per_second": 12191.858
    },
    {
      "epoch": 0.8981955292216537,
      "grad_norm": 0.7767614328805698,
      "learning_rate": 5.79668902527277e-06,
      "loss": 0.3844503879547119,
      "num_input_tokens_seen": 425291464,
      "step": 13340,
      "train_runtime": 34883.5049,
      "train_tokens_per_second": 12191.764
    },
    {
      "epoch": 0.8985321842176138,
      "grad_norm": 6.050973785331642,
      "learning_rate": 5.794078612318695e-06,
      "loss": 0.36563012599945066,
      "num_input_tokens_seen": 425451848,
      "step": 13345,
      "train_runtime": 34896.2743,
      "train_tokens_per_second": 12191.899
    },
    {
      "epoch": 0.898868839213574,
      "grad_norm": 0.7993592315509648,
      "learning_rate": 5.791467977303068e-06,
      "loss": 0.36136345863342284,
      "num_input_tokens_seen": 425603376,
      "step": 13350,
      "train_runtime": 34908.7813,
      "train_tokens_per_second": 12191.872
    },
    {
      "epoch": 0.8992054942095341,
      "grad_norm": 0.9150884954642485,
      "learning_rate": 5.788857120955945e-06,
      "loss": 0.38617358207702634,
      "num_input_tokens_seen": 425760352,
      "step": 13355,
      "train_runtime": 34922.0552,
      "train_tokens_per_second": 12191.732
    },
    {
      "epoch": 0.8995421492054942,
      "grad_norm": 0.7518432041045526,
      "learning_rate": 5.786246044007443e-06,
      "loss": 0.34881646633148194,
      "num_input_tokens_seen": 425918112,
      "step": 13360,
      "train_runtime": 34935.6834,
      "train_tokens_per_second": 12191.492
    },
    {
      "epoch": 0.8998788042014544,
      "grad_norm": 0.7444423272321159,
      "learning_rate": 5.783634747187743e-06,
      "loss": 0.3730934143066406,
      "num_input_tokens_seen": 426078528,
      "step": 13365,
      "train_runtime": 34948.2517,
      "train_tokens_per_second": 12191.698
    },
    {
      "epoch": 0.9002154591974145,
      "grad_norm": 0.79883218900397,
      "learning_rate": 5.781023231227084e-06,
      "loss": 0.38190927505493166,
      "num_input_tokens_seen": 426239664,
      "step": 13370,
      "train_runtime": 34962.1191,
      "train_tokens_per_second": 12191.471
    },
    {
      "epoch": 0.9005521141933747,
      "grad_norm": 0.73227697232891,
      "learning_rate": 5.778411496855769e-06,
      "loss": 0.3742210865020752,
      "num_input_tokens_seen": 426399000,
      "step": 13375,
      "train_runtime": 34974.7821,
      "train_tokens_per_second": 12191.613
    },
    {
      "epoch": 0.9008887691893348,
      "grad_norm": 0.8508150230525733,
      "learning_rate": 5.775799544804161e-06,
      "loss": 0.3930072784423828,
      "num_input_tokens_seen": 426559528,
      "step": 13380,
      "train_runtime": 34987.6981,
      "train_tokens_per_second": 12191.7
    },
    {
      "epoch": 0.9012254241852949,
      "grad_norm": 0.7986874064589201,
      "learning_rate": 5.7731873758026845e-06,
      "loss": 0.35907678604125975,
      "num_input_tokens_seen": 426717488,
      "step": 13385,
      "train_runtime": 35000.3757,
      "train_tokens_per_second": 12191.797
    },
    {
      "epoch": 0.9015620791812551,
      "grad_norm": 0.833203601985892,
      "learning_rate": 5.770574990581823e-06,
      "loss": 0.3693462133407593,
      "num_input_tokens_seen": 426876280,
      "step": 13390,
      "train_runtime": 35013.3937,
      "train_tokens_per_second": 12191.8
    },
    {
      "epoch": 0.9018987341772152,
      "grad_norm": 0.8392835446698765,
      "learning_rate": 5.767962389872124e-06,
      "loss": 0.33575925827026365,
      "num_input_tokens_seen": 427034720,
      "step": 13395,
      "train_runtime": 35027.4122,
      "train_tokens_per_second": 12191.444
    },
    {
      "epoch": 0.9022353891731754,
      "grad_norm": 0.7612415598428929,
      "learning_rate": 5.765349574404191e-06,
      "loss": 0.3605826377868652,
      "num_input_tokens_seen": 427189792,
      "step": 13400,
      "train_runtime": 35040.7719,
      "train_tokens_per_second": 12191.221
    },
    {
      "epoch": 0.9025720441691355,
      "grad_norm": 0.7294017964348306,
      "learning_rate": 5.7627365449086905e-06,
      "loss": 0.3711873769760132,
      "num_input_tokens_seen": 427340544,
      "step": 13405,
      "train_runtime": 35054.3118,
      "train_tokens_per_second": 12190.813
    },
    {
      "epoch": 0.9029086991650956,
      "grad_norm": 0.7903338814651439,
      "learning_rate": 5.760123302116345e-06,
      "loss": 0.3938957691192627,
      "num_input_tokens_seen": 427500888,
      "step": 13410,
      "train_runtime": 35067.2002,
      "train_tokens_per_second": 12190.904
    },
    {
      "epoch": 0.9032453541610558,
      "grad_norm": 0.8750741220457778,
      "learning_rate": 5.757509846757945e-06,
      "loss": 0.38195626735687255,
      "num_input_tokens_seen": 427660128,
      "step": 13415,
      "train_runtime": 35079.6346,
      "train_tokens_per_second": 12191.123
    },
    {
      "epoch": 0.9035820091570159,
      "grad_norm": 0.8513761284518359,
      "learning_rate": 5.75489617956433e-06,
      "loss": 0.395311975479126,
      "num_input_tokens_seen": 427823896,
      "step": 13420,
      "train_runtime": 35092.7454,
      "train_tokens_per_second": 12191.235
    },
    {
      "epoch": 0.9039186641529761,
      "grad_norm": 0.732615221170027,
      "learning_rate": 5.752282301266406e-06,
      "loss": 0.3921032905578613,
      "num_input_tokens_seen": 427983136,
      "step": 13425,
      "train_runtime": 35105.3297,
      "train_tokens_per_second": 12191.401
    },
    {
      "epoch": 0.9042553191489362,
      "grad_norm": 0.7325929552729605,
      "learning_rate": 5.749668212595136e-06,
      "loss": 0.3708685636520386,
      "num_input_tokens_seen": 428140912,
      "step": 13430,
      "train_runtime": 35118.702,
      "train_tokens_per_second": 12191.251
    },
    {
      "epoch": 0.9045919741448963,
      "grad_norm": 0.7955016985373715,
      "learning_rate": 5.747053914281539e-06,
      "loss": 0.3694139957427979,
      "num_input_tokens_seen": 428295448,
      "step": 13435,
      "train_runtime": 35131.9078,
      "train_tokens_per_second": 12191.067
    },
    {
      "epoch": 0.9049286291408565,
      "grad_norm": 0.7558650972837816,
      "learning_rate": 5.744439407056699e-06,
      "loss": 0.34850897789001467,
      "num_input_tokens_seen": 428453056,
      "step": 13440,
      "train_runtime": 35145.0095,
      "train_tokens_per_second": 12191.007
    },
    {
      "epoch": 0.9052652841368166,
      "grad_norm": 0.661957279232183,
      "learning_rate": 5.741824691651752e-06,
      "loss": 0.34030466079711913,
      "num_input_tokens_seen": 428611560,
      "step": 13445,
      "train_runtime": 35157.4822,
      "train_tokens_per_second": 12191.19
    },
    {
      "epoch": 0.9056019391327768,
      "grad_norm": 0.8470233964267815,
      "learning_rate": 5.739209768797896e-06,
      "loss": 0.3740764379501343,
      "num_input_tokens_seen": 428760712,
      "step": 13450,
      "train_runtime": 35170.5838,
      "train_tokens_per_second": 12190.89
    },
    {
      "epoch": 0.9059385941287369,
      "grad_norm": 0.9046747430698882,
      "learning_rate": 5.736594639226385e-06,
      "loss": 0.3245997428894043,
      "num_input_tokens_seen": 428918448,
      "step": 13455,
      "train_runtime": 35183.9583,
      "train_tokens_per_second": 12190.739
    },
    {
      "epoch": 0.906275249124697,
      "grad_norm": 0.786718501154718,
      "learning_rate": 5.7339793036685306e-06,
      "loss": 0.37534098625183104,
      "num_input_tokens_seen": 429081160,
      "step": 13460,
      "train_runtime": 35197.0567,
      "train_tokens_per_second": 12190.825
    },
    {
      "epoch": 0.9066119041206572,
      "grad_norm": 0.7355374187550461,
      "learning_rate": 5.731363762855706e-06,
      "loss": 0.3430805683135986,
      "num_input_tokens_seen": 429237800,
      "step": 13465,
      "train_runtime": 35209.989,
      "train_tokens_per_second": 12190.796
    },
    {
      "epoch": 0.9069485591166173,
      "grad_norm": 0.8179229468010564,
      "learning_rate": 5.728748017519337e-06,
      "loss": 0.4214427947998047,
      "num_input_tokens_seen": 429394720,
      "step": 13470,
      "train_runtime": 35223.2317,
      "train_tokens_per_second": 12190.668
    },
    {
      "epoch": 0.9072852141125775,
      "grad_norm": 0.717317464187995,
      "learning_rate": 5.726132068390908e-06,
      "loss": 0.35298852920532225,
      "num_input_tokens_seen": 429550408,
      "step": 13475,
      "train_runtime": 35237.5715,
      "train_tokens_per_second": 12190.125
    },
    {
      "epoch": 0.9076218691085376,
      "grad_norm": 0.8994482695267444,
      "learning_rate": 5.723515916201962e-06,
      "loss": 0.408160924911499,
      "num_input_tokens_seen": 429710680,
      "step": 13480,
      "train_runtime": 35251.0465,
      "train_tokens_per_second": 12190.012
    },
    {
      "epoch": 0.9079585241044977,
      "grad_norm": 0.7128442602046751,
      "learning_rate": 5.720899561684098e-06,
      "loss": 0.3687018871307373,
      "num_input_tokens_seen": 429868840,
      "step": 13485,
      "train_runtime": 35264.3494,
      "train_tokens_per_second": 12189.899
    },
    {
      "epoch": 0.9082951791004579,
      "grad_norm": 0.771041703269352,
      "learning_rate": 5.71828300556897e-06,
      "loss": 0.368972110748291,
      "num_input_tokens_seen": 430031736,
      "step": 13490,
      "train_runtime": 35277.8187,
      "train_tokens_per_second": 12189.862
    },
    {
      "epoch": 0.908631834096418,
      "grad_norm": 0.6915849834536898,
      "learning_rate": 5.715666248588289e-06,
      "loss": 0.35815114974975587,
      "num_input_tokens_seen": 430191368,
      "step": 13495,
      "train_runtime": 35291.0069,
      "train_tokens_per_second": 12189.83
    },
    {
      "epoch": 0.9089684890923782,
      "grad_norm": 0.8219703743548636,
      "learning_rate": 5.713049291473825e-06,
      "loss": 0.410277271270752,
      "num_input_tokens_seen": 430353792,
      "step": 13500,
      "train_runtime": 35304.3287,
      "train_tokens_per_second": 12189.831
    },
    {
      "epoch": 0.9093051440883383,
      "grad_norm": 0.7533280344436573,
      "learning_rate": 5.710432134957399e-06,
      "loss": 0.3380936861038208,
      "num_input_tokens_seen": 430506472,
      "step": 13505,
      "train_runtime": 35317.5565,
      "train_tokens_per_second": 12189.588
    },
    {
      "epoch": 0.9096417990842984,
      "grad_norm": 0.7271423360840092,
      "learning_rate": 5.707814779770892e-06,
      "loss": 0.3381925106048584,
      "num_input_tokens_seen": 430663480,
      "step": 13510,
      "train_runtime": 35331.3323,
      "train_tokens_per_second": 12189.279
    },
    {
      "epoch": 0.9099784540802586,
      "grad_norm": 0.7494573171412836,
      "learning_rate": 5.7051972266462395e-06,
      "loss": 0.35069563388824465,
      "num_input_tokens_seen": 430820424,
      "step": 13515,
      "train_runtime": 35344.1052,
      "train_tokens_per_second": 12189.315
    },
    {
      "epoch": 0.9103151090762187,
      "grad_norm": 0.8839719491103175,
      "learning_rate": 5.702579476315431e-06,
      "loss": 0.3460576057434082,
      "num_input_tokens_seen": 430980072,
      "step": 13520,
      "train_runtime": 35357.5549,
      "train_tokens_per_second": 12189.193
    },
    {
      "epoch": 0.9106517640721788,
      "grad_norm": 0.9271110943988584,
      "learning_rate": 5.699961529510513e-06,
      "loss": 0.35433759689331057,
      "num_input_tokens_seen": 431136096,
      "step": 13525,
      "train_runtime": 35371.3455,
      "train_tokens_per_second": 12188.852
    },
    {
      "epoch": 0.910988419068139,
      "grad_norm": 0.8580491885781263,
      "learning_rate": 5.697343386963586e-06,
      "loss": 0.3892681121826172,
      "num_input_tokens_seen": 431294640,
      "step": 13530,
      "train_runtime": 35383.8313,
      "train_tokens_per_second": 12189.032
    },
    {
      "epoch": 0.9113250740640991,
      "grad_norm": 0.7241399273882059,
      "learning_rate": 5.694725049406801e-06,
      "loss": 0.36517624855041503,
      "num_input_tokens_seen": 431455288,
      "step": 13535,
      "train_runtime": 35397.474,
      "train_tokens_per_second": 12188.872
    },
    {
      "epoch": 0.9116617290600593,
      "grad_norm": 0.7289588534116486,
      "learning_rate": 5.692106517572373e-06,
      "loss": 0.40781345367431643,
      "num_input_tokens_seen": 431611864,
      "step": 13540,
      "train_runtime": 35411.1184,
      "train_tokens_per_second": 12188.597
    },
    {
      "epoch": 0.9119983840560194,
      "grad_norm": 0.8110702410343946,
      "learning_rate": 5.689487792192566e-06,
      "loss": 0.39562430381774905,
      "num_input_tokens_seen": 431774536,
      "step": 13545,
      "train_runtime": 35424.1801,
      "train_tokens_per_second": 12188.695
    },
    {
      "epoch": 0.9123350390519795,
      "grad_norm": 0.774027019373964,
      "learning_rate": 5.686868873999696e-06,
      "loss": 0.3897235870361328,
      "num_input_tokens_seen": 431936448,
      "step": 13550,
      "train_runtime": 35437.3122,
      "train_tokens_per_second": 12188.747
    },
    {
      "epoch": 0.9126716940479397,
      "grad_norm": 0.740651979537237,
      "learning_rate": 5.684249763726135e-06,
      "loss": 0.39835100173950194,
      "num_input_tokens_seen": 432094168,
      "step": 13555,
      "train_runtime": 35450.1126,
      "train_tokens_per_second": 12188.795
    },
    {
      "epoch": 0.9130083490438998,
      "grad_norm": 0.7282103232114707,
      "learning_rate": 5.681630462104308e-06,
      "loss": 0.36507291793823243,
      "num_input_tokens_seen": 432249800,
      "step": 13560,
      "train_runtime": 35463.4428,
      "train_tokens_per_second": 12188.602
    },
    {
      "epoch": 0.91334500403986,
      "grad_norm": 0.7686449436564229,
      "learning_rate": 5.679010969866697e-06,
      "loss": 0.3869205951690674,
      "num_input_tokens_seen": 432404208,
      "step": 13565,
      "train_runtime": 35476.4032,
      "train_tokens_per_second": 12188.502
    },
    {
      "epoch": 0.9136816590358201,
      "grad_norm": 0.8027550391239413,
      "learning_rate": 5.676391287745835e-06,
      "loss": 0.3621238708496094,
      "num_input_tokens_seen": 432567472,
      "step": 13570,
      "train_runtime": 35490.0567,
      "train_tokens_per_second": 12188.413
    },
    {
      "epoch": 0.9140183140317802,
      "grad_norm": 0.7611785956967995,
      "learning_rate": 5.673771416474305e-06,
      "loss": 0.3609843969345093,
      "num_input_tokens_seen": 432728400,
      "step": 13575,
      "train_runtime": 35502.9729,
      "train_tokens_per_second": 12188.512
    },
    {
      "epoch": 0.9143549690277404,
      "grad_norm": 0.8677035639172358,
      "learning_rate": 5.671151356784747e-06,
      "loss": 0.3543481588363647,
      "num_input_tokens_seen": 432885576,
      "step": 13580,
      "train_runtime": 35516.8724,
      "train_tokens_per_second": 12188.167
    },
    {
      "epoch": 0.9146916240237005,
      "grad_norm": 0.7125369019855851,
      "learning_rate": 5.668531109409851e-06,
      "loss": 0.3914100408554077,
      "num_input_tokens_seen": 433044352,
      "step": 13585,
      "train_runtime": 35530.0053,
      "train_tokens_per_second": 12188.131
    },
    {
      "epoch": 0.9150282790196607,
      "grad_norm": 0.8082639911621977,
      "learning_rate": 5.665910675082362e-06,
      "loss": 0.36147136688232423,
      "num_input_tokens_seen": 433206304,
      "step": 13590,
      "train_runtime": 35543.0493,
      "train_tokens_per_second": 12188.214
    },
    {
      "epoch": 0.9153649340156208,
      "grad_norm": 0.7332669084158943,
      "learning_rate": 5.663290054535078e-06,
      "loss": 0.34022817611694334,
      "num_input_tokens_seen": 433362288,
      "step": 13595,
      "train_runtime": 35556.4007,
      "train_tokens_per_second": 12188.025
    },
    {
      "epoch": 0.9157015890115809,
      "grad_norm": 1.189938791704453,
      "learning_rate": 5.660669248500846e-06,
      "loss": 0.42548360824584963,
      "num_input_tokens_seen": 433520616,
      "step": 13600,
      "train_runtime": 35569.0173,
      "train_tokens_per_second": 12188.153
    },
    {
      "epoch": 0.9160382440075411,
      "grad_norm": 0.7993602089800331,
      "learning_rate": 5.658048257712563e-06,
      "loss": 0.360113263130188,
      "num_input_tokens_seen": 433682056,
      "step": 13605,
      "train_runtime": 35581.7367,
      "train_tokens_per_second": 12188.333
    },
    {
      "epoch": 0.9163748990035012,
      "grad_norm": 0.7627463130965576,
      "learning_rate": 5.655427082903184e-06,
      "loss": 0.3676086664199829,
      "num_input_tokens_seen": 433842048,
      "step": 13610,
      "train_runtime": 35595.2357,
      "train_tokens_per_second": 12188.206
    },
    {
      "epoch": 0.9167115539994614,
      "grad_norm": 0.8487433121029654,
      "learning_rate": 5.652805724805711e-06,
      "loss": 0.35324907302856445,
      "num_input_tokens_seen": 434001544,
      "step": 13615,
      "train_runtime": 35608.115,
      "train_tokens_per_second": 12188.276
    },
    {
      "epoch": 0.9170482089954215,
      "grad_norm": 0.7823794681501385,
      "learning_rate": 5.650184184153199e-06,
      "loss": 0.3836455106735229,
      "num_input_tokens_seen": 434164840,
      "step": 13620,
      "train_runtime": 35621.0438,
      "train_tokens_per_second": 12188.437
    },
    {
      "epoch": 0.9173848639913816,
      "grad_norm": 0.7481177520800324,
      "learning_rate": 5.647562461678754e-06,
      "loss": 0.4001375675201416,
      "num_input_tokens_seen": 434326616,
      "step": 13625,
      "train_runtime": 35633.7007,
      "train_tokens_per_second": 12188.647
    },
    {
      "epoch": 0.9177215189873418,
      "grad_norm": 0.8028197451382855,
      "learning_rate": 5.64494055811553e-06,
      "loss": 0.3532702922821045,
      "num_input_tokens_seen": 434487072,
      "step": 13630,
      "train_runtime": 35646.1422,
      "train_tokens_per_second": 12188.895
    },
    {
      "epoch": 0.9180581739833019,
      "grad_norm": 0.7786459826230611,
      "learning_rate": 5.642318474196737e-06,
      "loss": 0.353279709815979,
      "num_input_tokens_seen": 434647968,
      "step": 13635,
      "train_runtime": 35659.2346,
      "train_tokens_per_second": 12188.931
    },
    {
      "epoch": 0.9183948289792621,
      "grad_norm": 0.7501623544904534,
      "learning_rate": 5.63969621065563e-06,
      "loss": 0.3520198345184326,
      "num_input_tokens_seen": 434805472,
      "step": 13640,
      "train_runtime": 35672.5319,
      "train_tokens_per_second": 12188.803
    },
    {
      "epoch": 0.9187314839752222,
      "grad_norm": 0.6997699725511924,
      "learning_rate": 5.637073768225517e-06,
      "loss": 0.3382099151611328,
      "num_input_tokens_seen": 434964592,
      "step": 13645,
      "train_runtime": 35685.812,
      "train_tokens_per_second": 12188.726
    },
    {
      "epoch": 0.9190681389711823,
      "grad_norm": 0.8305126878904916,
      "learning_rate": 5.634451147639758e-06,
      "loss": 0.3398799657821655,
      "num_input_tokens_seen": 435119368,
      "step": 13650,
      "train_runtime": 35698.9502,
      "train_tokens_per_second": 12188.576
    },
    {
      "epoch": 0.9194047939671425,
      "grad_norm": 0.7218866910068225,
      "learning_rate": 5.631828349631757e-06,
      "loss": 0.33959915637969973,
      "num_input_tokens_seen": 435277080,
      "step": 13655,
      "train_runtime": 35711.6912,
      "train_tokens_per_second": 12188.644
    },
    {
      "epoch": 0.9197414489631026,
      "grad_norm": 0.9045107922970621,
      "learning_rate": 5.629205374934974e-06,
      "loss": 0.3553279161453247,
      "num_input_tokens_seen": 435437208,
      "step": 13660,
      "train_runtime": 35726.7389,
      "train_tokens_per_second": 12187.992
    },
    {
      "epoch": 0.9200781039590628,
      "grad_norm": 0.7403388418217152,
      "learning_rate": 5.626582224282913e-06,
      "loss": 0.3548685789108276,
      "num_input_tokens_seen": 435599272,
      "step": 13665,
      "train_runtime": 35740.0237,
      "train_tokens_per_second": 12187.996
    },
    {
      "epoch": 0.9204147589550229,
      "grad_norm": 0.7582255009938885,
      "learning_rate": 5.623958898409132e-06,
      "loss": 0.3900468826293945,
      "num_input_tokens_seen": 435759224,
      "step": 13670,
      "train_runtime": 35753.5795,
      "train_tokens_per_second": 12187.849
    },
    {
      "epoch": 0.920751413950983,
      "grad_norm": 0.7343918306482689,
      "learning_rate": 5.621335398047234e-06,
      "loss": 0.3853520154953003,
      "num_input_tokens_seen": 435919368,
      "step": 13675,
      "train_runtime": 35766.9338,
      "train_tokens_per_second": 12187.776
    },
    {
      "epoch": 0.9210880689469432,
      "grad_norm": 0.8284130013815835,
      "learning_rate": 5.618711723930874e-06,
      "loss": 0.3782789707183838,
      "num_input_tokens_seen": 436075592,
      "step": 13680,
      "train_runtime": 35780.1438,
      "train_tokens_per_second": 12187.642
    },
    {
      "epoch": 0.9214247239429033,
      "grad_norm": 0.7510678690702973,
      "learning_rate": 5.616087876793753e-06,
      "loss": 0.37306177616119385,
      "num_input_tokens_seen": 436236768,
      "step": 13685,
      "train_runtime": 35793.2417,
      "train_tokens_per_second": 12187.685
    },
    {
      "epoch": 0.9217613789388635,
      "grad_norm": 0.7455238631733695,
      "learning_rate": 5.613463857369621e-06,
      "loss": 0.38380353450775145,
      "num_input_tokens_seen": 436400208,
      "step": 13690,
      "train_runtime": 35806.3473,
      "train_tokens_per_second": 12187.789
    },
    {
      "epoch": 0.9220980339348236,
      "grad_norm": 0.7926070300483002,
      "learning_rate": 5.610839666392278e-06,
      "loss": 0.360155987739563,
      "num_input_tokens_seen": 436562384,
      "step": 13695,
      "train_runtime": 35819.5642,
      "train_tokens_per_second": 12187.82
    },
    {
      "epoch": 0.9224346889307837,
      "grad_norm": 0.7945456937110481,
      "learning_rate": 5.608215304595571e-06,
      "loss": 0.3858468532562256,
      "num_input_tokens_seen": 436724472,
      "step": 13700,
      "train_runtime": 35832.7907,
      "train_tokens_per_second": 12187.844
    },
    {
      "epoch": 0.9227713439267439,
      "grad_norm": 0.7245062158291434,
      "learning_rate": 5.605590772713393e-06,
      "loss": 0.35793142318725585,
      "num_input_tokens_seen": 436881528,
      "step": 13705,
      "train_runtime": 35846.7912,
      "train_tokens_per_second": 12187.465
    },
    {
      "epoch": 0.923107998922704,
      "grad_norm": 0.7954866085668164,
      "learning_rate": 5.602966071479687e-06,
      "loss": 0.37648849487304686,
      "num_input_tokens_seen": 437037280,
      "step": 13710,
      "train_runtime": 35860.1152,
      "train_tokens_per_second": 12187.28
    },
    {
      "epoch": 0.9234446539186641,
      "grad_norm": 0.6873798437968969,
      "learning_rate": 5.600341201628439e-06,
      "loss": 0.3618522882461548,
      "num_input_tokens_seen": 437199080,
      "step": 13715,
      "train_runtime": 35873.596,
      "train_tokens_per_second": 12187.211
    },
    {
      "epoch": 0.9237813089146243,
      "grad_norm": 0.7155045224900212,
      "learning_rate": 5.597716163893692e-06,
      "loss": 0.3160524368286133,
      "num_input_tokens_seen": 437361216,
      "step": 13720,
      "train_runtime": 35886.2012,
      "train_tokens_per_second": 12187.448
    },
    {
      "epoch": 0.9241179639105844,
      "grad_norm": 0.8178363046279644,
      "learning_rate": 5.595090959009525e-06,
      "loss": 0.3420442581176758,
      "num_input_tokens_seen": 437521968,
      "step": 13725,
      "train_runtime": 35898.9889,
      "train_tokens_per_second": 12187.585
    },
    {
      "epoch": 0.9244546189065446,
      "grad_norm": 0.7183387031247779,
      "learning_rate": 5.592465587710068e-06,
      "loss": 0.33625240325927735,
      "num_input_tokens_seen": 437683384,
      "step": 13730,
      "train_runtime": 35911.468,
      "train_tokens_per_second": 12187.844
    },
    {
      "epoch": 0.9247912739025047,
      "grad_norm": 0.7219246101906679,
      "learning_rate": 5.589840050729498e-06,
      "loss": 0.3400043725967407,
      "num_input_tokens_seen": 437843760,
      "step": 13735,
      "train_runtime": 35924.6045,
      "train_tokens_per_second": 12187.852
    },
    {
      "epoch": 0.9251279288984648,
      "grad_norm": 0.8497876871994456,
      "learning_rate": 5.587214348802039e-06,
      "loss": 0.4082938194274902,
      "num_input_tokens_seen": 438002784,
      "step": 13740,
      "train_runtime": 35937.5558,
      "train_tokens_per_second": 12187.885
    },
    {
      "epoch": 0.925464583894425,
      "grad_norm": 0.745585340686398,
      "learning_rate": 5.5845884826619615e-06,
      "loss": 0.3598623752593994,
      "num_input_tokens_seen": 438163264,
      "step": 13745,
      "train_runtime": 35951.213,
      "train_tokens_per_second": 12187.719
    },
    {
      "epoch": 0.9258012388903851,
      "grad_norm": 1.2043469270837208,
      "learning_rate": 5.5819624530435775e-06,
      "loss": 0.3688164234161377,
      "num_input_tokens_seen": 438323856,
      "step": 13750,
      "train_runtime": 35964.1505,
      "train_tokens_per_second": 12187.8
    },
    {
      "epoch": 0.9261378938863453,
      "grad_norm": 0.7585430791296551,
      "learning_rate": 5.57933626068125e-06,
      "loss": 0.40395293235778806,
      "num_input_tokens_seen": 438485320,
      "step": 13755,
      "train_runtime": 35977.0602,
      "train_tokens_per_second": 12187.914
    },
    {
      "epoch": 0.9264745488823054,
      "grad_norm": 0.7982525057780485,
      "learning_rate": 5.5767099063093835e-06,
      "loss": 0.37363805770874026,
      "num_input_tokens_seen": 438644152,
      "step": 13760,
      "train_runtime": 35989.6404,
      "train_tokens_per_second": 12188.067
    },
    {
      "epoch": 0.9268112038782655,
      "grad_norm": 1.0198431554371141,
      "learning_rate": 5.574083390662431e-06,
      "loss": 0.38454232215881345,
      "num_input_tokens_seen": 438805168,
      "step": 13765,
      "train_runtime": 36002.5293,
      "train_tokens_per_second": 12188.176
    },
    {
      "epoch": 0.9271478588742257,
      "grad_norm": 0.8058498846239168,
      "learning_rate": 5.57145671447489e-06,
      "loss": 0.3844372510910034,
      "num_input_tokens_seen": 438958120,
      "step": 13770,
      "train_runtime": 36015.496,
      "train_tokens_per_second": 12188.035
    },
    {
      "epoch": 0.9274845138701858,
      "grad_norm": 0.8573202254881467,
      "learning_rate": 5.5688298784813e-06,
      "loss": 0.3866144895553589,
      "num_input_tokens_seen": 439117688,
      "step": 13775,
      "train_runtime": 36029.6355,
      "train_tokens_per_second": 12187.681
    },
    {
      "epoch": 0.927821168866146,
      "grad_norm": 0.7556113281045054,
      "learning_rate": 5.566202883416249e-06,
      "loss": 0.4027858734130859,
      "num_input_tokens_seen": 439279152,
      "step": 13780,
      "train_runtime": 36042.0966,
      "train_tokens_per_second": 12187.947
    },
    {
      "epoch": 0.9281578238621061,
      "grad_norm": 0.8969969329826353,
      "learning_rate": 5.563575730014366e-06,
      "loss": 0.40751848220825193,
      "num_input_tokens_seen": 439434992,
      "step": 13785,
      "train_runtime": 36055.4626,
      "train_tokens_per_second": 12187.751
    },
    {
      "epoch": 0.9284944788580662,
      "grad_norm": 0.7953107672749814,
      "learning_rate": 5.560948419010328e-06,
      "loss": 0.3921651363372803,
      "num_input_tokens_seen": 439598688,
      "step": 13790,
      "train_runtime": 36068.5069,
      "train_tokens_per_second": 12187.881
    },
    {
      "epoch": 0.9288311338540264,
      "grad_norm": 0.7318609376027997,
      "learning_rate": 5.558320951138853e-06,
      "loss": 0.3627305507659912,
      "num_input_tokens_seen": 439761200,
      "step": 13795,
      "train_runtime": 36081.8154,
      "train_tokens_per_second": 12187.89
    },
    {
      "epoch": 0.9291677888499865,
      "grad_norm": 0.7543206884172933,
      "learning_rate": 5.5556933271347046e-06,
      "loss": 0.3809455394744873,
      "num_input_tokens_seen": 439921784,
      "step": 13800,
      "train_runtime": 36094.6929,
      "train_tokens_per_second": 12187.991
    },
    {
      "epoch": 0.9295044438459467,
      "grad_norm": 0.7249638777934448,
      "learning_rate": 5.55306554773269e-06,
      "loss": 0.3896027088165283,
      "num_input_tokens_seen": 440079400,
      "step": 13805,
      "train_runtime": 36107.5854,
      "train_tokens_per_second": 12188.004
    },
    {
      "epoch": 0.9298410988419068,
      "grad_norm": 0.8577222631813816,
      "learning_rate": 5.550437613667658e-06,
      "loss": 0.3538863182067871,
      "num_input_tokens_seen": 440233488,
      "step": 13810,
      "train_runtime": 36121.4544,
      "train_tokens_per_second": 12187.59
    },
    {
      "epoch": 0.9301777538378669,
      "grad_norm": 0.7679644627129287,
      "learning_rate": 5.547809525674501e-06,
      "loss": 0.3617534160614014,
      "num_input_tokens_seen": 440389392,
      "step": 13815,
      "train_runtime": 36134.7055,
      "train_tokens_per_second": 12187.435
    },
    {
      "epoch": 0.9305144088338271,
      "grad_norm": 1.1484592077230544,
      "learning_rate": 5.545181284488159e-06,
      "loss": 0.3943178176879883,
      "num_input_tokens_seen": 440552216,
      "step": 13820,
      "train_runtime": 36147.4655,
      "train_tokens_per_second": 12187.638
    },
    {
      "epoch": 0.9308510638297872,
      "grad_norm": 0.8318210445063801,
      "learning_rate": 5.5425528908436085e-06,
      "loss": 0.37734689712524416,
      "num_input_tokens_seen": 440710096,
      "step": 13825,
      "train_runtime": 36161.2689,
      "train_tokens_per_second": 12187.352
    },
    {
      "epoch": 0.9311877188257474,
      "grad_norm": 0.8153835492590567,
      "learning_rate": 5.539924345475873e-06,
      "loss": 0.38255393505096436,
      "num_input_tokens_seen": 440863472,
      "step": 13830,
      "train_runtime": 36174.7345,
      "train_tokens_per_second": 12187.055
    },
    {
      "epoch": 0.9315243738217075,
      "grad_norm": 0.8290801093520833,
      "learning_rate": 5.5372956491200145e-06,
      "loss": 0.3836904764175415,
      "num_input_tokens_seen": 441017848,
      "step": 13835,
      "train_runtime": 36187.5446,
      "train_tokens_per_second": 12187.007
    },
    {
      "epoch": 0.9318610288176676,
      "grad_norm": 0.7545905605006498,
      "learning_rate": 5.534666802511143e-06,
      "loss": 0.36342949867248536,
      "num_input_tokens_seen": 441172752,
      "step": 13840,
      "train_runtime": 36200.7846,
      "train_tokens_per_second": 12186.828
    },
    {
      "epoch": 0.9321976838136278,
      "grad_norm": 0.9370753732456154,
      "learning_rate": 5.532037806384404e-06,
      "loss": 0.3723859071731567,
      "num_input_tokens_seen": 441333832,
      "step": 13845,
      "train_runtime": 36213.9333,
      "train_tokens_per_second": 12186.852
    },
    {
      "epoch": 0.9325343388095879,
      "grad_norm": 0.7693952993648859,
      "learning_rate": 5.529408661474989e-06,
      "loss": 0.35491232872009276,
      "num_input_tokens_seen": 441485480,
      "step": 13850,
      "train_runtime": 36226.7617,
      "train_tokens_per_second": 12186.722
    },
    {
      "epoch": 0.932870993805548,
      "grad_norm": 0.7914960218939375,
      "learning_rate": 5.526779368518129e-06,
      "loss": 0.404754114151001,
      "num_input_tokens_seen": 441647440,
      "step": 13855,
      "train_runtime": 36239.3257,
      "train_tokens_per_second": 12186.966
    },
    {
      "epoch": 0.9332076488015082,
      "grad_norm": 0.8928356628841586,
      "learning_rate": 5.5241499282491e-06,
      "loss": 0.37383761405944826,
      "num_input_tokens_seen": 441803144,
      "step": 13860,
      "train_runtime": 36253.1819,
      "train_tokens_per_second": 12186.603
    },
    {
      "epoch": 0.9335443037974683,
      "grad_norm": 0.7142641768444611,
      "learning_rate": 5.5215203414032135e-06,
      "loss": 0.35165064334869384,
      "num_input_tokens_seen": 441966280,
      "step": 13865,
      "train_runtime": 36265.9786,
      "train_tokens_per_second": 12186.801
    },
    {
      "epoch": 0.9338809587934285,
      "grad_norm": 0.7606356934359481,
      "learning_rate": 5.518890608715828e-06,
      "loss": 0.36032867431640625,
      "num_input_tokens_seen": 442126040,
      "step": 13870,
      "train_runtime": 36279.1756,
      "train_tokens_per_second": 12186.772
    },
    {
      "epoch": 0.9342176137893886,
      "grad_norm": 0.9170878942581523,
      "learning_rate": 5.516260730922337e-06,
      "loss": 0.39864609241485593,
      "num_input_tokens_seen": 442280072,
      "step": 13875,
      "train_runtime": 36292.3268,
      "train_tokens_per_second": 12186.6
    },
    {
      "epoch": 0.9345542687853488,
      "grad_norm": 0.7759815527790396,
      "learning_rate": 5.51363070875818e-06,
      "loss": 0.3764288663864136,
      "num_input_tokens_seen": 442438424,
      "step": 13880,
      "train_runtime": 36305.5866,
      "train_tokens_per_second": 12186.511
    },
    {
      "epoch": 0.9348909237813089,
      "grad_norm": 0.7103101909517316,
      "learning_rate": 5.511000542958832e-06,
      "loss": 0.39668152332305906,
      "num_input_tokens_seen": 442596432,
      "step": 13885,
      "train_runtime": 36319.6998,
      "train_tokens_per_second": 12186.126
    },
    {
      "epoch": 0.935227578777269,
      "grad_norm": 0.8891080198861476,
      "learning_rate": 5.508370234259812e-06,
      "loss": 0.3612593412399292,
      "num_input_tokens_seen": 442754888,
      "step": 13890,
      "train_runtime": 36332.166,
      "train_tokens_per_second": 12186.306
    },
    {
      "epoch": 0.9355642337732292,
      "grad_norm": 0.7746743052198768,
      "learning_rate": 5.505739783396678e-06,
      "loss": 0.4050482749938965,
      "num_input_tokens_seen": 442913432,
      "step": 13895,
      "train_runtime": 36345.2963,
      "train_tokens_per_second": 12186.266
    },
    {
      "epoch": 0.9359008887691893,
      "grad_norm": 0.7806922017303211,
      "learning_rate": 5.503109191105026e-06,
      "loss": 0.36899418830871583,
      "num_input_tokens_seen": 443068464,
      "step": 13900,
      "train_runtime": 36358.7129,
      "train_tokens_per_second": 12186.033
    },
    {
      "epoch": 0.9362375437651494,
      "grad_norm": 0.7668651109460208,
      "learning_rate": 5.500478458120493e-06,
      "loss": 0.4057442665100098,
      "num_input_tokens_seen": 443227784,
      "step": 13905,
      "train_runtime": 36371.4379,
      "train_tokens_per_second": 12186.15
    },
    {
      "epoch": 0.9365741987611096,
      "grad_norm": 0.7440293900675143,
      "learning_rate": 5.497847585178754e-06,
      "loss": 0.3907855272293091,
      "num_input_tokens_seen": 443387080,
      "step": 13910,
      "train_runtime": 36384.4614,
      "train_tokens_per_second": 12186.166
    },
    {
      "epoch": 0.9369108537570697,
      "grad_norm": 0.7255660960702708,
      "learning_rate": 5.495216573015526e-06,
      "loss": 0.3533169746398926,
      "num_input_tokens_seen": 443544512,
      "step": 13915,
      "train_runtime": 36397.2578,
      "train_tokens_per_second": 12186.207
    },
    {
      "epoch": 0.9372475087530299,
      "grad_norm": 0.7382644356718363,
      "learning_rate": 5.492585422366562e-06,
      "loss": 0.3474785089492798,
      "num_input_tokens_seen": 443704008,
      "step": 13920,
      "train_runtime": 36409.9181,
      "train_tokens_per_second": 12186.35
    },
    {
      "epoch": 0.93758416374899,
      "grad_norm": 0.7005126443786297,
      "learning_rate": 5.489954133967657e-06,
      "loss": 0.3417153835296631,
      "num_input_tokens_seen": 443858624,
      "step": 13925,
      "train_runtime": 36422.5608,
      "train_tokens_per_second": 12186.365
    },
    {
      "epoch": 0.9379208187449501,
      "grad_norm": 0.7610057894605636,
      "learning_rate": 5.48732270855464e-06,
      "loss": 0.3548431873321533,
      "num_input_tokens_seen": 444022416,
      "step": 13930,
      "train_runtime": 36435.6483,
      "train_tokens_per_second": 12186.483
    },
    {
      "epoch": 0.9382574737409103,
      "grad_norm": 0.76856369287446,
      "learning_rate": 5.48469114686338e-06,
      "loss": 0.3636350393295288,
      "num_input_tokens_seen": 444176648,
      "step": 13935,
      "train_runtime": 36448.5689,
      "train_tokens_per_second": 12186.395
    },
    {
      "epoch": 0.9385941287368704,
      "grad_norm": 0.7710704926418497,
      "learning_rate": 5.482059449629788e-06,
      "loss": 0.35806918144226074,
      "num_input_tokens_seen": 444337888,
      "step": 13940,
      "train_runtime": 36461.0497,
      "train_tokens_per_second": 12186.646
    },
    {
      "epoch": 0.9389307837328306,
      "grad_norm": 0.7246643438859957,
      "learning_rate": 5.479427617589808e-06,
      "loss": 0.3702289581298828,
      "num_input_tokens_seen": 444497160,
      "step": 13945,
      "train_runtime": 36473.9183,
      "train_tokens_per_second": 12186.713
    },
    {
      "epoch": 0.9392674387287907,
      "grad_norm": 0.7112055160792288,
      "learning_rate": 5.4767956514794226e-06,
      "loss": 0.38546218872070315,
      "num_input_tokens_seen": 444657920,
      "step": 13950,
      "train_runtime": 36486.3546,
      "train_tokens_per_second": 12186.965
    },
    {
      "epoch": 0.9396040937247508,
      "grad_norm": 0.8755228518876836,
      "learning_rate": 5.474163552034655e-06,
      "loss": 0.36105003356933596,
      "num_input_tokens_seen": 444811744,
      "step": 13955,
      "train_runtime": 36499.4943,
      "train_tokens_per_second": 12186.792
    },
    {
      "epoch": 0.939940748720711,
      "grad_norm": 0.7641643222459481,
      "learning_rate": 5.4715313199915595e-06,
      "loss": 0.3539999485015869,
      "num_input_tokens_seen": 444967144,
      "step": 13960,
      "train_runtime": 36513.6935,
      "train_tokens_per_second": 12186.309
    },
    {
      "epoch": 0.9402774037166711,
      "grad_norm": 0.8671036623458517,
      "learning_rate": 5.468898956086236e-06,
      "loss": 0.37435474395751955,
      "num_input_tokens_seen": 445127160,
      "step": 13965,
      "train_runtime": 36526.4978,
      "train_tokens_per_second": 12186.418
    },
    {
      "epoch": 0.9406140587126313,
      "grad_norm": 0.7977875793230513,
      "learning_rate": 5.466266461054815e-06,
      "loss": 0.3693050146102905,
      "num_input_tokens_seen": 445289696,
      "step": 13970,
      "train_runtime": 36538.9516,
      "train_tokens_per_second": 12186.712
    },
    {
      "epoch": 0.9409507137085914,
      "grad_norm": 0.7499401867807944,
      "learning_rate": 5.463633835633464e-06,
      "loss": 0.391118597984314,
      "num_input_tokens_seen": 445452896,
      "step": 13975,
      "train_runtime": 36551.7812,
      "train_tokens_per_second": 12186.9
    },
    {
      "epoch": 0.9412873687045515,
      "grad_norm": 0.7362568406494412,
      "learning_rate": 5.46100108055839e-06,
      "loss": 0.37929835319519045,
      "num_input_tokens_seen": 445613848,
      "step": 13980,
      "train_runtime": 36564.6549,
      "train_tokens_per_second": 12187.011
    },
    {
      "epoch": 0.9416240237005117,
      "grad_norm": 0.8221134354682008,
      "learning_rate": 5.458368196565833e-06,
      "loss": 0.3321723222732544,
      "num_input_tokens_seen": 445775584,
      "step": 13985,
      "train_runtime": 36577.6663,
      "train_tokens_per_second": 12187.097
    },
    {
      "epoch": 0.9419606786964718,
      "grad_norm": 0.8558306169627611,
      "learning_rate": 5.455735184392073e-06,
      "loss": 0.38138484954833984,
      "num_input_tokens_seen": 445939096,
      "step": 13990,
      "train_runtime": 36590.0973,
      "train_tokens_per_second": 12187.426
    },
    {
      "epoch": 0.942297333692432,
      "grad_norm": 0.8116076487236336,
      "learning_rate": 5.4531020447734204e-06,
      "loss": 0.38126182556152344,
      "num_input_tokens_seen": 446096856,
      "step": 13995,
      "train_runtime": 36603.1819,
      "train_tokens_per_second": 12187.379
    },
    {
      "epoch": 0.9426339886883921,
      "grad_norm": 2.010805132193222,
      "learning_rate": 5.450468778446226e-06,
      "loss": 0.37644472122192385,
      "num_input_tokens_seen": 446260696,
      "step": 14000,
      "train_runtime": 36615.6118,
      "train_tokens_per_second": 12187.717
    },
    {
      "epoch": 0.9429706436843522,
      "grad_norm": 0.8343198443956905,
      "learning_rate": 5.447835386146875e-06,
      "loss": 0.3727182626724243,
      "num_input_tokens_seen": 446423848,
      "step": 14005,
      "train_runtime": 36628.4716,
      "train_tokens_per_second": 12187.892
    },
    {
      "epoch": 0.9433072986803124,
      "grad_norm": 0.7317344712550558,
      "learning_rate": 5.445201868611784e-06,
      "loss": 0.4103872776031494,
      "num_input_tokens_seen": 446584104,
      "step": 14010,
      "train_runtime": 36641.9557,
      "train_tokens_per_second": 12187.78
    },
    {
      "epoch": 0.9436439536762725,
      "grad_norm": 0.7317534250849597,
      "learning_rate": 5.442568226577413e-06,
      "loss": 0.3761645555496216,
      "num_input_tokens_seen": 446745576,
      "step": 14015,
      "train_runtime": 36654.9035,
      "train_tokens_per_second": 12187.88
    },
    {
      "epoch": 0.9439806086722327,
      "grad_norm": 0.7354748681825876,
      "learning_rate": 5.439934460780247e-06,
      "loss": 0.34859304428100585,
      "num_input_tokens_seen": 446905720,
      "step": 14020,
      "train_runtime": 36667.7324,
      "train_tokens_per_second": 12187.984
    },
    {
      "epoch": 0.9443172636681928,
      "grad_norm": 0.6699010344372787,
      "learning_rate": 5.437300571956813e-06,
      "loss": 0.3796972751617432,
      "num_input_tokens_seen": 447059320,
      "step": 14025,
      "train_runtime": 36680.4761,
      "train_tokens_per_second": 12187.937
    },
    {
      "epoch": 0.9446539186641529,
      "grad_norm": 0.7416459126571048,
      "learning_rate": 5.434666560843667e-06,
      "loss": 0.35377705097198486,
      "num_input_tokens_seen": 447213656,
      "step": 14030,
      "train_runtime": 36693.1772,
      "train_tokens_per_second": 12187.924
    },
    {
      "epoch": 0.9449905736601131,
      "grad_norm": 0.7297523647373726,
      "learning_rate": 5.432032428177405e-06,
      "loss": 0.38315491676330565,
      "num_input_tokens_seen": 447376432,
      "step": 14035,
      "train_runtime": 36705.9213,
      "train_tokens_per_second": 12188.127
    },
    {
      "epoch": 0.9453272286560732,
      "grad_norm": 0.82855793269814,
      "learning_rate": 5.429398174694651e-06,
      "loss": 0.34437713623046873,
      "num_input_tokens_seen": 447539112,
      "step": 14040,
      "train_runtime": 36718.6177,
      "train_tokens_per_second": 12188.343
    },
    {
      "epoch": 0.9456638836520334,
      "grad_norm": 0.7225058758957088,
      "learning_rate": 5.426763801132067e-06,
      "loss": 0.3724466562271118,
      "num_input_tokens_seen": 447689232,
      "step": 14045,
      "train_runtime": 36732.1853,
      "train_tokens_per_second": 12187.928
    },
    {
      "epoch": 0.9460005386479935,
      "grad_norm": 0.8120492077889117,
      "learning_rate": 5.424129308226347e-06,
      "loss": 0.3902392864227295,
      "num_input_tokens_seen": 447843968,
      "step": 14050,
      "train_runtime": 36744.8917,
      "train_tokens_per_second": 12187.925
    },
    {
      "epoch": 0.9463371936439536,
      "grad_norm": 0.77097565782903,
      "learning_rate": 5.421494696714219e-06,
      "loss": 0.4099125862121582,
      "num_input_tokens_seen": 448007808,
      "step": 14055,
      "train_runtime": 36757.3229,
      "train_tokens_per_second": 12188.26
    },
    {
      "epoch": 0.9466738486399138,
      "grad_norm": 0.7053564537083229,
      "learning_rate": 5.418859967332442e-06,
      "loss": 0.35971038341522216,
      "num_input_tokens_seen": 448169648,
      "step": 14060,
      "train_runtime": 36769.859,
      "train_tokens_per_second": 12188.506
    },
    {
      "epoch": 0.9470105036358739,
      "grad_norm": 0.8779938524669136,
      "learning_rate": 5.416225120817811e-06,
      "loss": 0.38056182861328125,
      "num_input_tokens_seen": 448319600,
      "step": 14065,
      "train_runtime": 36783.4362,
      "train_tokens_per_second": 12188.084
    },
    {
      "epoch": 0.947347158631834,
      "grad_norm": 0.8294238448858673,
      "learning_rate": 5.413590157907153e-06,
      "loss": 0.3597753047943115,
      "num_input_tokens_seen": 448483440,
      "step": 14070,
      "train_runtime": 36795.8644,
      "train_tokens_per_second": 12188.42
    },
    {
      "epoch": 0.9476838136277942,
      "grad_norm": 0.7471990852830855,
      "learning_rate": 5.4109550793373255e-06,
      "loss": 0.3628887414932251,
      "num_input_tokens_seen": 448644736,
      "step": 14075,
      "train_runtime": 36808.3291,
      "train_tokens_per_second": 12188.674
    },
    {
      "epoch": 0.9480204686237543,
      "grad_norm": 0.8181142574573709,
      "learning_rate": 5.40831988584522e-06,
      "loss": 0.39128866195678713,
      "num_input_tokens_seen": 448804280,
      "step": 14080,
      "train_runtime": 36821.1039,
      "train_tokens_per_second": 12188.779
    },
    {
      "epoch": 0.9483571236197145,
      "grad_norm": 0.7581642721164937,
      "learning_rate": 5.40568457816776e-06,
      "loss": 0.381651496887207,
      "num_input_tokens_seen": 448967728,
      "step": 14085,
      "train_runtime": 36834.1092,
      "train_tokens_per_second": 12188.912
    },
    {
      "epoch": 0.9486937786156746,
      "grad_norm": 0.7008786642035699,
      "learning_rate": 5.403049157041903e-06,
      "loss": 0.36566188335418703,
      "num_input_tokens_seen": 449127152,
      "step": 14090,
      "train_runtime": 36846.916,
      "train_tokens_per_second": 12189.003
    },
    {
      "epoch": 0.9490304336116347,
      "grad_norm": 0.8712206615664322,
      "learning_rate": 5.400413623204635e-06,
      "loss": 0.3482259750366211,
      "num_input_tokens_seen": 449288008,
      "step": 14095,
      "train_runtime": 36860.673,
      "train_tokens_per_second": 12188.817
    },
    {
      "epoch": 0.9493670886075949,
      "grad_norm": 0.8470507877600656,
      "learning_rate": 5.397777977392973e-06,
      "loss": 0.4060727596282959,
      "num_input_tokens_seen": 449450856,
      "step": 14100,
      "train_runtime": 36874.7797,
      "train_tokens_per_second": 12188.571
    },
    {
      "epoch": 0.949703743603555,
      "grad_norm": 0.7607946082993723,
      "learning_rate": 5.395142220343967e-06,
      "loss": 0.37915027141571045,
      "num_input_tokens_seen": 449612920,
      "step": 14105,
      "train_runtime": 36888.0555,
      "train_tokens_per_second": 12188.577
    },
    {
      "epoch": 0.9500403985995152,
      "grad_norm": 0.7446290397140839,
      "learning_rate": 5.3925063527947e-06,
      "loss": 0.3664482355117798,
      "num_input_tokens_seen": 449775456,
      "step": 14110,
      "train_runtime": 36902.7499,
      "train_tokens_per_second": 12188.128
    },
    {
      "epoch": 0.9503770535954753,
      "grad_norm": 0.7709488654203768,
      "learning_rate": 5.389870375482283e-06,
      "loss": 0.40068492889404295,
      "num_input_tokens_seen": 449932416,
      "step": 14115,
      "train_runtime": 36916.7268,
      "train_tokens_per_second": 12187.766
    },
    {
      "epoch": 0.9507137085914354,
      "grad_norm": 0.9440710909792275,
      "learning_rate": 5.387234289143859e-06,
      "loss": 0.35057199001312256,
      "num_input_tokens_seen": 450089272,
      "step": 14120,
      "train_runtime": 36930.1936,
      "train_tokens_per_second": 12187.569
    },
    {
      "epoch": 0.9510503635873956,
      "grad_norm": 0.7244910811376039,
      "learning_rate": 5.384598094516601e-06,
      "loss": 0.3898542642593384,
      "num_input_tokens_seen": 450252216,
      "step": 14125,
      "train_runtime": 36943.6998,
      "train_tokens_per_second": 12187.524
    },
    {
      "epoch": 0.9513870185833557,
      "grad_norm": 0.7296658717889744,
      "learning_rate": 5.381961792337712e-06,
      "loss": 0.3796407461166382,
      "num_input_tokens_seen": 450414720,
      "step": 14130,
      "train_runtime": 36956.3757,
      "train_tokens_per_second": 12187.741
    },
    {
      "epoch": 0.9517236735793159,
      "grad_norm": 0.7183164253815125,
      "learning_rate": 5.3793253833444245e-06,
      "loss": 0.3474994897842407,
      "num_input_tokens_seen": 450573848,
      "step": 14135,
      "train_runtime": 36969.0658,
      "train_tokens_per_second": 12187.861
    },
    {
      "epoch": 0.952060328575276,
      "grad_norm": 0.6884243012361292,
      "learning_rate": 5.3766888682740045e-06,
      "loss": 0.3536869525909424,
      "num_input_tokens_seen": 450734904,
      "step": 14140,
      "train_runtime": 36982.3916,
      "train_tokens_per_second": 12187.825
    },
    {
      "epoch": 0.9523969835712361,
      "grad_norm": 0.8188414535712991,
      "learning_rate": 5.374052247863745e-06,
      "loss": 0.37145175933837893,
      "num_input_tokens_seen": 450891392,
      "step": 14145,
      "train_runtime": 36995.6293,
      "train_tokens_per_second": 12187.694
    },
    {
      "epoch": 0.9527336385671963,
      "grad_norm": 0.767305436419478,
      "learning_rate": 5.371415522850966e-06,
      "loss": 0.37569870948791506,
      "num_input_tokens_seen": 451049032,
      "step": 14150,
      "train_runtime": 37008.8009,
      "train_tokens_per_second": 12187.615
    },
    {
      "epoch": 0.9530702935631564,
      "grad_norm": 0.9797885723961564,
      "learning_rate": 5.368778693973019e-06,
      "loss": 0.3529491901397705,
      "num_input_tokens_seen": 451206776,
      "step": 14155,
      "train_runtime": 37021.8381,
      "train_tokens_per_second": 12187.584
    },
    {
      "epoch": 0.9534069485591166,
      "grad_norm": 0.7335182006097138,
      "learning_rate": 5.366141761967287e-06,
      "loss": 0.38477230072021484,
      "num_input_tokens_seen": 451364320,
      "step": 14160,
      "train_runtime": 37034.3662,
      "train_tokens_per_second": 12187.716
    },
    {
      "epoch": 0.9537436035550767,
      "grad_norm": 0.8292175762414618,
      "learning_rate": 5.363504727571179e-06,
      "loss": 0.38037211894989015,
      "num_input_tokens_seen": 451523952,
      "step": 14165,
      "train_runtime": 37047.701,
      "train_tokens_per_second": 12187.638
    },
    {
      "epoch": 0.954080258551037,
      "grad_norm": 0.8744536988474335,
      "learning_rate": 5.360867591522133e-06,
      "loss": 0.42474966049194335,
      "num_input_tokens_seen": 451687792,
      "step": 14170,
      "train_runtime": 37060.1278,
      "train_tokens_per_second": 12187.972
    },
    {
      "epoch": 0.9544169135469971,
      "grad_norm": 0.7647512759883638,
      "learning_rate": 5.358230354557617e-06,
      "loss": 0.34644064903259275,
      "num_input_tokens_seen": 451848696,
      "step": 14175,
      "train_runtime": 37073.6017,
      "train_tokens_per_second": 12187.882
    },
    {
      "epoch": 0.9547535685429572,
      "grad_norm": 0.809870693160124,
      "learning_rate": 5.355593017415123e-06,
      "loss": 0.4143869400024414,
      "num_input_tokens_seen": 452009784,
      "step": 14180,
      "train_runtime": 37086.5086,
      "train_tokens_per_second": 12187.984
    },
    {
      "epoch": 0.9550902235389174,
      "grad_norm": 0.7491096506251071,
      "learning_rate": 5.352955580832176e-06,
      "loss": 0.3671841859817505,
      "num_input_tokens_seen": 452173624,
      "step": 14185,
      "train_runtime": 37098.9293,
      "train_tokens_per_second": 12188.32
    },
    {
      "epoch": 0.9554268785348775,
      "grad_norm": 0.7185791041884471,
      "learning_rate": 5.350318045546326e-06,
      "loss": 0.36811280250549316,
      "num_input_tokens_seen": 452332592,
      "step": 14190,
      "train_runtime": 37111.5189,
      "train_tokens_per_second": 12188.469
    },
    {
      "epoch": 0.9557635335308377,
      "grad_norm": 0.78134114785335,
      "learning_rate": 5.347680412295152e-06,
      "loss": 0.3897176027297974,
      "num_input_tokens_seen": 452492584,
      "step": 14195,
      "train_runtime": 37124.9231,
      "train_tokens_per_second": 12188.378
    },
    {
      "epoch": 0.9561001885267978,
      "grad_norm": 0.886941867472269,
      "learning_rate": 5.34504268181626e-06,
      "loss": 0.3777491807937622,
      "num_input_tokens_seen": 452650872,
      "step": 14200,
      "train_runtime": 37138.1786,
      "train_tokens_per_second": 12188.289
    },
    {
      "epoch": 0.9564368435227579,
      "grad_norm": 0.936749729452858,
      "learning_rate": 5.342404854847282e-06,
      "loss": 0.3859711170196533,
      "num_input_tokens_seen": 452804328,
      "step": 14205,
      "train_runtime": 37151.5339,
      "train_tokens_per_second": 12188.039
    },
    {
      "epoch": 0.9567734985187181,
      "grad_norm": 0.6951983340438338,
      "learning_rate": 5.3397669321258776e-06,
      "loss": 0.3589979648590088,
      "num_input_tokens_seen": 452966984,
      "step": 14210,
      "train_runtime": 37164.2394,
      "train_tokens_per_second": 12188.248
    },
    {
      "epoch": 0.9571101535146782,
      "grad_norm": 0.7434936287166044,
      "learning_rate": 5.337128914389734e-06,
      "loss": 0.40025577545166013,
      "num_input_tokens_seen": 453128488,
      "step": 14215,
      "train_runtime": 37177.0684,
      "train_tokens_per_second": 12188.387
    },
    {
      "epoch": 0.9574468085106383,
      "grad_norm": 0.764591281125932,
      "learning_rate": 5.334490802376565e-06,
      "loss": 0.373136305809021,
      "num_input_tokens_seen": 453291728,
      "step": 14220,
      "train_runtime": 37189.525,
      "train_tokens_per_second": 12188.694
    },
    {
      "epoch": 0.9577834635065985,
      "grad_norm": 1.2439273878082822,
      "learning_rate": 5.33185259682411e-06,
      "loss": 0.3976832628250122,
      "num_input_tokens_seen": 453444056,
      "step": 14225,
      "train_runtime": 37202.2116,
      "train_tokens_per_second": 12188.632
    },
    {
      "epoch": 0.9581201185025586,
      "grad_norm": 0.7540906865696856,
      "learning_rate": 5.329214298470134e-06,
      "loss": 0.39126389026641845,
      "num_input_tokens_seen": 453604512,
      "step": 14230,
      "train_runtime": 37215.6913,
      "train_tokens_per_second": 12188.528
    },
    {
      "epoch": 0.9584567734985188,
      "grad_norm": 0.7417173621803747,
      "learning_rate": 5.326575908052428e-06,
      "loss": 0.3592424154281616,
      "num_input_tokens_seen": 453767328,
      "step": 14235,
      "train_runtime": 37228.3566,
      "train_tokens_per_second": 12188.755
    },
    {
      "epoch": 0.9587934284944789,
      "grad_norm": 1.112988824368141,
      "learning_rate": 5.323937426308813e-06,
      "loss": 0.42380571365356445,
      "num_input_tokens_seen": 453929080,
      "step": 14240,
      "train_runtime": 37240.8956,
      "train_tokens_per_second": 12188.995
    },
    {
      "epoch": 0.959130083490439,
      "grad_norm": 0.74821466402817,
      "learning_rate": 5.321298853977128e-06,
      "loss": 0.3365647315979004,
      "num_input_tokens_seen": 454092920,
      "step": 14245,
      "train_runtime": 37253.3311,
      "train_tokens_per_second": 12189.324
    },
    {
      "epoch": 0.9594667384863992,
      "grad_norm": 0.7225322558169128,
      "learning_rate": 5.318660191795244e-06,
      "loss": 0.3519944667816162,
      "num_input_tokens_seen": 454248608,
      "step": 14250,
      "train_runtime": 37266.3091,
      "train_tokens_per_second": 12189.257
    },
    {
      "epoch": 0.9598033934823593,
      "grad_norm": 0.7153740342072372,
      "learning_rate": 5.316021440501053e-06,
      "loss": 0.3227083206176758,
      "num_input_tokens_seen": 454410032,
      "step": 14255,
      "train_runtime": 37281.321,
      "train_tokens_per_second": 12188.678
    },
    {
      "epoch": 0.9601400484783195,
      "grad_norm": 0.7389520545340884,
      "learning_rate": 5.313382600832474e-06,
      "loss": 0.3425132751464844,
      "num_input_tokens_seen": 454572528,
      "step": 14260,
      "train_runtime": 37294.8662,
      "train_tokens_per_second": 12188.609
    },
    {
      "epoch": 0.9604767034742796,
      "grad_norm": 0.7852045345588873,
      "learning_rate": 5.310743673527451e-06,
      "loss": 0.39432587623596194,
      "num_input_tokens_seen": 454733200,
      "step": 14265,
      "train_runtime": 37307.3568,
      "train_tokens_per_second": 12188.835
    },
    {
      "epoch": 0.9608133584702397,
      "grad_norm": 0.7307382606692295,
      "learning_rate": 5.3081046593239496e-06,
      "loss": 0.3800446271896362,
      "num_input_tokens_seen": 454891192,
      "step": 14270,
      "train_runtime": 37319.7706,
      "train_tokens_per_second": 12189.014
    },
    {
      "epoch": 0.9611500134661999,
      "grad_norm": 0.7237620345275972,
      "learning_rate": 5.305465558959964e-06,
      "loss": 0.36459693908691404,
      "num_input_tokens_seen": 455051208,
      "step": 14275,
      "train_runtime": 37333.6012,
      "train_tokens_per_second": 12188.784
    },
    {
      "epoch": 0.96148666846216,
      "grad_norm": 0.7863378298144611,
      "learning_rate": 5.302826373173506e-06,
      "loss": 0.3564426898956299,
      "num_input_tokens_seen": 455210928,
      "step": 14280,
      "train_runtime": 37346.2719,
      "train_tokens_per_second": 12188.926
    },
    {
      "epoch": 0.9618233234581202,
      "grad_norm": 0.755744196548147,
      "learning_rate": 5.3001871027026194e-06,
      "loss": 0.36427080631256104,
      "num_input_tokens_seen": 455368904,
      "step": 14285,
      "train_runtime": 37359.7785,
      "train_tokens_per_second": 12188.747
    },
    {
      "epoch": 0.9621599784540803,
      "grad_norm": 0.7502650304796661,
      "learning_rate": 5.2975477482853685e-06,
      "loss": 0.36396446228027346,
      "num_input_tokens_seen": 455529648,
      "step": 14290,
      "train_runtime": 37372.203,
      "train_tokens_per_second": 12188.996
    },
    {
      "epoch": 0.9624966334500404,
      "grad_norm": 0.8439977122593634,
      "learning_rate": 5.294908310659838e-06,
      "loss": 0.35645992755889894,
      "num_input_tokens_seen": 455689008,
      "step": 14295,
      "train_runtime": 37384.6388,
      "train_tokens_per_second": 12189.205
    },
    {
      "epoch": 0.9628332884460006,
      "grad_norm": 0.8232400167706919,
      "learning_rate": 5.292268790564138e-06,
      "loss": 0.35867953300476074,
      "num_input_tokens_seen": 455848688,
      "step": 14300,
      "train_runtime": 37397.309,
      "train_tokens_per_second": 12189.345
    },
    {
      "epoch": 0.9631699434419607,
      "grad_norm": 0.790494087299572,
      "learning_rate": 5.289629188736403e-06,
      "loss": 0.40677671432495116,
      "num_input_tokens_seen": 456011208,
      "step": 14305,
      "train_runtime": 37411.2715,
      "train_tokens_per_second": 12189.14
    },
    {
      "epoch": 0.9635065984379209,
      "grad_norm": 0.8591506703062213,
      "learning_rate": 5.286989505914788e-06,
      "loss": 0.3871395349502563,
      "num_input_tokens_seen": 456174568,
      "step": 14310,
      "train_runtime": 37424.2371,
      "train_tokens_per_second": 12189.282
    },
    {
      "epoch": 0.963843253433881,
      "grad_norm": 0.951215245991792,
      "learning_rate": 5.284349742837475e-06,
      "loss": 0.37456276416778567,
      "num_input_tokens_seen": 456336416,
      "step": 14315,
      "train_runtime": 37436.7693,
      "train_tokens_per_second": 12189.524
    },
    {
      "epoch": 0.9641799084298411,
      "grad_norm": 0.9728174147500613,
      "learning_rate": 5.281709900242662e-06,
      "loss": 0.36783158779144287,
      "num_input_tokens_seen": 456496664,
      "step": 14320,
      "train_runtime": 37449.5352,
      "train_tokens_per_second": 12189.648
    },
    {
      "epoch": 0.9645165634258013,
      "grad_norm": 0.7539586731921601,
      "learning_rate": 5.279069978868574e-06,
      "loss": 0.3605900049209595,
      "num_input_tokens_seen": 456658304,
      "step": 14325,
      "train_runtime": 37462.6009,
      "train_tokens_per_second": 12189.712
    },
    {
      "epoch": 0.9648532184217614,
      "grad_norm": 0.789045301350713,
      "learning_rate": 5.276429979453455e-06,
      "loss": 0.39236452579498293,
      "num_input_tokens_seen": 456817752,
      "step": 14330,
      "train_runtime": 37475.9762,
      "train_tokens_per_second": 12189.616
    },
    {
      "epoch": 0.9651898734177216,
      "grad_norm": 0.7037960674999219,
      "learning_rate": 5.273789902735575e-06,
      "loss": 0.35866966247558596,
      "num_input_tokens_seen": 456977016,
      "step": 14335,
      "train_runtime": 37488.8683,
      "train_tokens_per_second": 12189.672
    },
    {
      "epoch": 0.9655265284136817,
      "grad_norm": 0.773957684548207,
      "learning_rate": 5.27114974945322e-06,
      "loss": 0.35486769676208496,
      "num_input_tokens_seen": 457135240,
      "step": 14340,
      "train_runtime": 37502.3174,
      "train_tokens_per_second": 12189.52
    },
    {
      "epoch": 0.9658631834096418,
      "grad_norm": 0.8146027035613111,
      "learning_rate": 5.2685095203447046e-06,
      "loss": 0.36144609451293946,
      "num_input_tokens_seen": 457296672,
      "step": 14345,
      "train_runtime": 37515.9357,
      "train_tokens_per_second": 12189.398
    },
    {
      "epoch": 0.966199838405602,
      "grad_norm": 0.7307889812805859,
      "learning_rate": 5.265869216148357e-06,
      "loss": 0.36154971122741697,
      "num_input_tokens_seen": 457455128,
      "step": 14350,
      "train_runtime": 37528.6014,
      "train_tokens_per_second": 12189.506
    },
    {
      "epoch": 0.9665364934015621,
      "grad_norm": 0.7871377891361506,
      "learning_rate": 5.263228837602532e-06,
      "loss": 0.4133151054382324,
      "num_input_tokens_seen": 457611312,
      "step": 14355,
      "train_runtime": 37541.4637,
      "train_tokens_per_second": 12189.49
    },
    {
      "epoch": 0.9668731483975223,
      "grad_norm": 0.7438574811560466,
      "learning_rate": 5.2605883854456e-06,
      "loss": 0.37948946952819823,
      "num_input_tokens_seen": 457774000,
      "step": 14360,
      "train_runtime": 37554.4747,
      "train_tokens_per_second": 12189.599
    },
    {
      "epoch": 0.9672098033934824,
      "grad_norm": 0.6945289211089826,
      "learning_rate": 5.257947860415957e-06,
      "loss": 0.32939133644104,
      "num_input_tokens_seen": 457933224,
      "step": 14365,
      "train_runtime": 37567.3265,
      "train_tokens_per_second": 12189.668
    },
    {
      "epoch": 0.9675464583894425,
      "grad_norm": 0.8365326863548435,
      "learning_rate": 5.255307263252019e-06,
      "loss": 0.34754252433776855,
      "num_input_tokens_seen": 458094632,
      "step": 14370,
      "train_runtime": 37580.431,
      "train_tokens_per_second": 12189.712
    },
    {
      "epoch": 0.9678831133854027,
      "grad_norm": 0.6993116447208811,
      "learning_rate": 5.2526665946922185e-06,
      "loss": 0.39890940189361573,
      "num_input_tokens_seen": 458256368,
      "step": 14375,
      "train_runtime": 37592.9606,
      "train_tokens_per_second": 12189.952
    },
    {
      "epoch": 0.9682197683813628,
      "grad_norm": 0.7583886440316113,
      "learning_rate": 5.25002585547501e-06,
      "loss": 0.36263489723205566,
      "num_input_tokens_seen": 458416944,
      "step": 14380,
      "train_runtime": 37605.9782,
      "train_tokens_per_second": 12190.002
    },
    {
      "epoch": 0.968556423377323,
      "grad_norm": 0.7455960611864481,
      "learning_rate": 5.2473850463388675e-06,
      "loss": 0.39397149085998534,
      "num_input_tokens_seen": 458577400,
      "step": 14385,
      "train_runtime": 37618.3855,
      "train_tokens_per_second": 12190.247
    },
    {
      "epoch": 0.9688930783732831,
      "grad_norm": 0.7884269754008423,
      "learning_rate": 5.244744168022288e-06,
      "loss": 0.36410582065582275,
      "num_input_tokens_seen": 458740008,
      "step": 14390,
      "train_runtime": 37631.0766,
      "train_tokens_per_second": 12190.457
    },
    {
      "epoch": 0.9692297333692432,
      "grad_norm": 0.7777869069441501,
      "learning_rate": 5.24210322126378e-06,
      "loss": 0.3689519166946411,
      "num_input_tokens_seen": 458903352,
      "step": 14395,
      "train_runtime": 37644.1715,
      "train_tokens_per_second": 12190.555
    },
    {
      "epoch": 0.9695663883652034,
      "grad_norm": 0.7742942403465678,
      "learning_rate": 5.23946220680188e-06,
      "loss": 0.37396509647369386,
      "num_input_tokens_seen": 459067192,
      "step": 14400,
      "train_runtime": 37656.6105,
      "train_tokens_per_second": 12190.879
    },
    {
      "epoch": 0.9699030433611635,
      "grad_norm": 0.8079182267709639,
      "learning_rate": 5.236821125375136e-06,
      "loss": 0.35969550609588624,
      "num_input_tokens_seen": 459228472,
      "step": 14405,
      "train_runtime": 37669.881,
      "train_tokens_per_second": 12190.866
    },
    {
      "epoch": 0.9702396983571236,
      "grad_norm": 0.6655218005996368,
      "learning_rate": 5.234179977722119e-06,
      "loss": 0.3288028955459595,
      "num_input_tokens_seen": 459392312,
      "step": 14410,
      "train_runtime": 37682.3158,
      "train_tokens_per_second": 12191.191
    },
    {
      "epoch": 0.9705763533530838,
      "grad_norm": 0.7013689432600881,
      "learning_rate": 5.23153876458142e-06,
      "loss": 0.36847169399261476,
      "num_input_tokens_seen": 459550776,
      "step": 14415,
      "train_runtime": 37696.8265,
      "train_tokens_per_second": 12190.702
    },
    {
      "epoch": 0.9709130083490439,
      "grad_norm": 0.7787213569762704,
      "learning_rate": 5.2288974866916435e-06,
      "loss": 0.37161428928375245,
      "num_input_tokens_seen": 459699472,
      "step": 14420,
      "train_runtime": 37709.4211,
      "train_tokens_per_second": 12190.574
    },
    {
      "epoch": 0.9712496633450041,
      "grad_norm": 0.8239083471099061,
      "learning_rate": 5.226256144791414e-06,
      "loss": 0.379974365234375,
      "num_input_tokens_seen": 459854680,
      "step": 14425,
      "train_runtime": 37722.8214,
      "train_tokens_per_second": 12190.358
    },
    {
      "epoch": 0.9715863183409642,
      "grad_norm": 0.7716519508546801,
      "learning_rate": 5.223614739619375e-06,
      "loss": 0.37413322925567627,
      "num_input_tokens_seen": 460009640,
      "step": 14430,
      "train_runtime": 37736.5311,
      "train_tokens_per_second": 12190.035
    },
    {
      "epoch": 0.9719229733369243,
      "grad_norm": 0.6951826925866688,
      "learning_rate": 5.2209732719141855e-06,
      "loss": 0.34517014026641846,
      "num_input_tokens_seen": 460168216,
      "step": 14435,
      "train_runtime": 37749.5691,
      "train_tokens_per_second": 12190.026
    },
    {
      "epoch": 0.9722596283328845,
      "grad_norm": 0.8089390327054088,
      "learning_rate": 5.218331742414528e-06,
      "loss": 0.35672996044158933,
      "num_input_tokens_seen": 460329360,
      "step": 14440,
      "train_runtime": 37762.6676,
      "train_tokens_per_second": 12190.065
    },
    {
      "epoch": 0.9725962833288446,
      "grad_norm": 0.7634757472096132,
      "learning_rate": 5.215690151859094e-06,
      "loss": 0.3562602519989014,
      "num_input_tokens_seen": 460485672,
      "step": 14445,
      "train_runtime": 37775.7036,
      "train_tokens_per_second": 12189.996
    },
    {
      "epoch": 0.9729329383248048,
      "grad_norm": 0.7933560250898755,
      "learning_rate": 5.213048500986597e-06,
      "loss": 0.38979640007019045,
      "num_input_tokens_seen": 460646552,
      "step": 14450,
      "train_runtime": 37789.1612,
      "train_tokens_per_second": 12189.912
    },
    {
      "epoch": 0.9732695933207649,
      "grad_norm": 0.8272756708843483,
      "learning_rate": 5.210406790535766e-06,
      "loss": 0.3738735198974609,
      "num_input_tokens_seen": 460804272,
      "step": 14455,
      "train_runtime": 37802.3389,
      "train_tokens_per_second": 12189.835
    },
    {
      "epoch": 0.973606248316725,
      "grad_norm": 0.7441848802672192,
      "learning_rate": 5.207765021245346e-06,
      "loss": 0.37014117240905764,
      "num_input_tokens_seen": 460963520,
      "step": 14460,
      "train_runtime": 37815.7726,
      "train_tokens_per_second": 12189.716
    },
    {
      "epoch": 0.9739429033126852,
      "grad_norm": 0.8125375048886795,
      "learning_rate": 5.2051231938541015e-06,
      "loss": 0.3655364513397217,
      "num_input_tokens_seen": 461118152,
      "step": 14465,
      "train_runtime": 37829.2046,
      "train_tokens_per_second": 12189.475
    },
    {
      "epoch": 0.9742795583086453,
      "grad_norm": 0.7688422566461453,
      "learning_rate": 5.202481309100808e-06,
      "loss": 0.36936194896698,
      "num_input_tokens_seen": 461276184,
      "step": 14470,
      "train_runtime": 37842.6202,
      "train_tokens_per_second": 12189.33
    },
    {
      "epoch": 0.9746162133046055,
      "grad_norm": 0.7809438496801016,
      "learning_rate": 5.199839367724266e-06,
      "loss": 0.3820170879364014,
      "num_input_tokens_seen": 461437400,
      "step": 14475,
      "train_runtime": 37855.0747,
      "train_tokens_per_second": 12189.578
    },
    {
      "epoch": 0.9749528683005656,
      "grad_norm": 0.7382035432021151,
      "learning_rate": 5.197197370463278e-06,
      "loss": 0.3520524024963379,
      "num_input_tokens_seen": 461601240,
      "step": 14480,
      "train_runtime": 37867.5138,
      "train_tokens_per_second": 12189.901
    },
    {
      "epoch": 0.9752895232965257,
      "grad_norm": 0.9968966691462924,
      "learning_rate": 5.194555318056675e-06,
      "loss": 0.40593700408935546,
      "num_input_tokens_seen": 461765080,
      "step": 14485,
      "train_runtime": 37879.9541,
      "train_tokens_per_second": 12190.223
    },
    {
      "epoch": 0.9756261782924859,
      "grad_norm": 0.9263491316623089,
      "learning_rate": 5.191913211243299e-06,
      "loss": 0.3960269927978516,
      "num_input_tokens_seen": 461925784,
      "step": 14490,
      "train_runtime": 37892.3912,
      "train_tokens_per_second": 12190.463
    },
    {
      "epoch": 0.975962833288446,
      "grad_norm": 0.9232103319973263,
      "learning_rate": 5.1892710507620055e-06,
      "loss": 0.3641959190368652,
      "num_input_tokens_seen": 462086600,
      "step": 14495,
      "train_runtime": 37905.917,
      "train_tokens_per_second": 12190.355
    },
    {
      "epoch": 0.9762994882844062,
      "grad_norm": 0.7201328186407614,
      "learning_rate": 5.186628837351666e-06,
      "loss": 0.3365805149078369,
      "num_input_tokens_seen": 462247552,
      "step": 14500,
      "train_runtime": 37918.8323,
      "train_tokens_per_second": 12190.448
    },
    {
      "epoch": 0.9766361432803663,
      "grad_norm": 0.7336594502036727,
      "learning_rate": 5.183986571751167e-06,
      "loss": 0.37297930717468264,
      "num_input_tokens_seen": 462410056,
      "step": 14505,
      "train_runtime": 37932.1738,
      "train_tokens_per_second": 12190.444
    },
    {
      "epoch": 0.9769727982763264,
      "grad_norm": 1.671714332768747,
      "learning_rate": 5.181344254699408e-06,
      "loss": 0.396826958656311,
      "num_input_tokens_seen": 462569928,
      "step": 14510,
      "train_runtime": 37945.0329,
      "train_tokens_per_second": 12190.526
    },
    {
      "epoch": 0.9773094532722866,
      "grad_norm": 0.7223086551779808,
      "learning_rate": 5.178701886935309e-06,
      "loss": 0.3570199489593506,
      "num_input_tokens_seen": 462724368,
      "step": 14515,
      "train_runtime": 37958.3798,
      "train_tokens_per_second": 12190.309
    },
    {
      "epoch": 0.9776461082682467,
      "grad_norm": 0.6831554958682025,
      "learning_rate": 5.176059469197796e-06,
      "loss": 0.3520961284637451,
      "num_input_tokens_seen": 462885360,
      "step": 14520,
      "train_runtime": 37971.3489,
      "train_tokens_per_second": 12190.385
    },
    {
      "epoch": 0.9779827632642069,
      "grad_norm": 0.8049938496894542,
      "learning_rate": 5.173417002225815e-06,
      "loss": 0.34856243133544923,
      "num_input_tokens_seen": 463044496,
      "step": 14525,
      "train_runtime": 37984.6605,
      "train_tokens_per_second": 12190.302
    },
    {
      "epoch": 0.978319418260167,
      "grad_norm": 1.0130538246511267,
      "learning_rate": 5.170774486758322e-06,
      "loss": 0.37352752685546875,
      "num_input_tokens_seen": 463193728,
      "step": 14530,
      "train_runtime": 37997.9777,
      "train_tokens_per_second": 12189.957
    },
    {
      "epoch": 0.9786560732561271,
      "grad_norm": 0.7060297629483385,
      "learning_rate": 5.168131923534287e-06,
      "loss": 0.38727126121520994,
      "num_input_tokens_seen": 463354016,
      "step": 14535,
      "train_runtime": 38011.1479,
      "train_tokens_per_second": 12189.951
    },
    {
      "epoch": 0.9789927282520873,
      "grad_norm": 0.7725840183103372,
      "learning_rate": 5.165489313292696e-06,
      "loss": 0.36117095947265626,
      "num_input_tokens_seen": 463512296,
      "step": 14540,
      "train_runtime": 38024.1813,
      "train_tokens_per_second": 12189.935
    },
    {
      "epoch": 0.9793293832480474,
      "grad_norm": 0.741042917241812,
      "learning_rate": 5.162846656772548e-06,
      "loss": 0.3356083869934082,
      "num_input_tokens_seen": 463670272,
      "step": 14545,
      "train_runtime": 38036.6662,
      "train_tokens_per_second": 12190.087
    },
    {
      "epoch": 0.9796660382440076,
      "grad_norm": 0.8454127740063642,
      "learning_rate": 5.16020395471285e-06,
      "loss": 0.3562184810638428,
      "num_input_tokens_seen": 463829952,
      "step": 14550,
      "train_runtime": 38049.6801,
      "train_tokens_per_second": 12190.114
    },
    {
      "epoch": 0.9800026932399677,
      "grad_norm": 0.7490317449064234,
      "learning_rate": 5.157561207852628e-06,
      "loss": 0.376031231880188,
      "num_input_tokens_seen": 463982872,
      "step": 14555,
      "train_runtime": 38062.8747,
      "train_tokens_per_second": 12189.906
    },
    {
      "epoch": 0.9803393482359278,
      "grad_norm": 0.7952757370643082,
      "learning_rate": 5.154918416930915e-06,
      "loss": 0.3518802642822266,
      "num_input_tokens_seen": 464141192,
      "step": 14560,
      "train_runtime": 38076.4912,
      "train_tokens_per_second": 12189.705
    },
    {
      "epoch": 0.980676003231888,
      "grad_norm": 0.8383011603568105,
      "learning_rate": 5.152275582686761e-06,
      "loss": 0.4002518653869629,
      "num_input_tokens_seen": 464298328,
      "step": 14565,
      "train_runtime": 38089.2328,
      "train_tokens_per_second": 12189.753
    },
    {
      "epoch": 0.9810126582278481,
      "grad_norm": 0.8069517483258946,
      "learning_rate": 5.149632705859227e-06,
      "loss": 0.3675529956817627,
      "num_input_tokens_seen": 464460232,
      "step": 14570,
      "train_runtime": 38102.4546,
      "train_tokens_per_second": 12189.772
    },
    {
      "epoch": 0.9813493132238083,
      "grad_norm": 0.8626272001319256,
      "learning_rate": 5.146989787187383e-06,
      "loss": 0.3813780307769775,
      "num_input_tokens_seen": 464618776,
      "step": 14575,
      "train_runtime": 38114.9992,
      "train_tokens_per_second": 12189.92
    },
    {
      "epoch": 0.9816859682197684,
      "grad_norm": 0.7804321206014594,
      "learning_rate": 5.144346827410314e-06,
      "loss": 0.37446858882904055,
      "num_input_tokens_seen": 464777600,
      "step": 14580,
      "train_runtime": 38127.5605,
      "train_tokens_per_second": 12190.069
    },
    {
      "epoch": 0.9820226232157285,
      "grad_norm": 0.8622149732520641,
      "learning_rate": 5.141703827267117e-06,
      "loss": 0.39135141372680665,
      "num_input_tokens_seen": 464936560,
      "step": 14585,
      "train_runtime": 38140.1023,
      "train_tokens_per_second": 12190.228
    },
    {
      "epoch": 0.9823592782116887,
      "grad_norm": 0.7511951111785298,
      "learning_rate": 5.139060787496893e-06,
      "loss": 0.384206485748291,
      "num_input_tokens_seen": 465097040,
      "step": 14590,
      "train_runtime": 38152.7256,
      "train_tokens_per_second": 12190.401
    },
    {
      "epoch": 0.9826959332076488,
      "grad_norm": 0.9634676325548029,
      "learning_rate": 5.136417708838766e-06,
      "loss": 0.38232452869415284,
      "num_input_tokens_seen": 465252168,
      "step": 14595,
      "train_runtime": 38167.0006,
      "train_tokens_per_second": 12189.906
    },
    {
      "epoch": 0.983032588203609,
      "grad_norm": 0.8437121707370554,
      "learning_rate": 5.13377459203186e-06,
      "loss": 0.3698857069015503,
      "num_input_tokens_seen": 465410864,
      "step": 14600,
      "train_runtime": 38181.3123,
      "train_tokens_per_second": 12189.494
    },
    {
      "epoch": 0.9833692431995691,
      "grad_norm": 0.734436813582297,
      "learning_rate": 5.131131437815318e-06,
      "loss": 0.3412070035934448,
      "num_input_tokens_seen": 465570960,
      "step": 14605,
      "train_runtime": 38194.6815,
      "train_tokens_per_second": 12189.419
    },
    {
      "epoch": 0.9837058981955292,
      "grad_norm": 0.735107304989146,
      "learning_rate": 5.128488246928287e-06,
      "loss": 0.3736592769622803,
      "num_input_tokens_seen": 465732504,
      "step": 14610,
      "train_runtime": 38207.2099,
      "train_tokens_per_second": 12189.65
    },
    {
      "epoch": 0.9840425531914894,
      "grad_norm": 0.772764873286212,
      "learning_rate": 5.125845020109925e-06,
      "loss": 0.37637863159179685,
      "num_input_tokens_seen": 465894856,
      "step": 14615,
      "train_runtime": 38219.8878,
      "train_tokens_per_second": 12189.854
    },
    {
      "epoch": 0.9843792081874495,
      "grad_norm": 0.8526057276579204,
      "learning_rate": 5.123201758099406e-06,
      "loss": 0.37646872997283937,
      "num_input_tokens_seen": 466055048,
      "step": 14620,
      "train_runtime": 38233.3819,
      "train_tokens_per_second": 12189.742
    },
    {
      "epoch": 0.9847158631834096,
      "grad_norm": 0.8615279653418381,
      "learning_rate": 5.12055846163591e-06,
      "loss": 0.38402888774871824,
      "num_input_tokens_seen": 466210464,
      "step": 14625,
      "train_runtime": 38245.8569,
      "train_tokens_per_second": 12189.829
    },
    {
      "epoch": 0.9850525181793698,
      "grad_norm": 0.70316794770979,
      "learning_rate": 5.117915131458623e-06,
      "loss": 0.33484597206115724,
      "num_input_tokens_seen": 466373032,
      "step": 14630,
      "train_runtime": 38259.4654,
      "train_tokens_per_second": 12189.743
    },
    {
      "epoch": 0.9853891731753299,
      "grad_norm": 0.8857117300487848,
      "learning_rate": 5.115271768306745e-06,
      "loss": 0.3731785774230957,
      "num_input_tokens_seen": 466527464,
      "step": 14635,
      "train_runtime": 38272.3041,
      "train_tokens_per_second": 12189.688
    },
    {
      "epoch": 0.9857258281712901,
      "grad_norm": 0.7593815421749512,
      "learning_rate": 5.112628372919485e-06,
      "loss": 0.3848330736160278,
      "num_input_tokens_seen": 466689408,
      "step": 14640,
      "train_runtime": 38285.5776,
      "train_tokens_per_second": 12189.692
    },
    {
      "epoch": 0.9860624831672502,
      "grad_norm": 0.7371794230732585,
      "learning_rate": 5.1099849460360595e-06,
      "loss": 0.3547880411148071,
      "num_input_tokens_seen": 466838856,
      "step": 14645,
      "train_runtime": 38298.7327,
      "train_tokens_per_second": 12189.407
    },
    {
      "epoch": 0.9863991381632103,
      "grad_norm": 0.6764122726709579,
      "learning_rate": 5.107341488395695e-06,
      "loss": 0.3506003856658936,
      "num_input_tokens_seen": 467002480,
      "step": 14650,
      "train_runtime": 38311.7702,
      "train_tokens_per_second": 12189.53
    },
    {
      "epoch": 0.9867357931591705,
      "grad_norm": 0.7471504762959407,
      "learning_rate": 5.104698000737626e-06,
      "loss": 0.37791857719421384,
      "num_input_tokens_seen": 467163640,
      "step": 14655,
      "train_runtime": 38324.7813,
      "train_tokens_per_second": 12189.597
    },
    {
      "epoch": 0.9870724481551306,
      "grad_norm": 0.7498307704092784,
      "learning_rate": 5.102054483801094e-06,
      "loss": 0.3794479131698608,
      "num_input_tokens_seen": 467327280,
      "step": 14660,
      "train_runtime": 38337.8242,
      "train_tokens_per_second": 12189.718
    },
    {
      "epoch": 0.9874091031510908,
      "grad_norm": 0.6887394754180528,
      "learning_rate": 5.099410938325351e-06,
      "loss": 0.3883463621139526,
      "num_input_tokens_seen": 467487784,
      "step": 14665,
      "train_runtime": 38350.8446,
      "train_tokens_per_second": 12189.765
    },
    {
      "epoch": 0.9877457581470509,
      "grad_norm": 0.7888553019094406,
      "learning_rate": 5.096767365049657e-06,
      "loss": 0.35828289985656736,
      "num_input_tokens_seen": 467642104,
      "step": 14670,
      "train_runtime": 38363.8098,
      "train_tokens_per_second": 12189.668
    },
    {
      "epoch": 0.988082413143011,
      "grad_norm": 0.6316462645305534,
      "learning_rate": 5.094123764713277e-06,
      "loss": 0.33677384853363035,
      "num_input_tokens_seen": 467805240,
      "step": 14675,
      "train_runtime": 38376.6303,
      "train_tokens_per_second": 12189.847
    },
    {
      "epoch": 0.9884190681389712,
      "grad_norm": 0.7988930417641884,
      "learning_rate": 5.091480138055487e-06,
      "loss": 0.38098456859588625,
      "num_input_tokens_seen": 467965872,
      "step": 14680,
      "train_runtime": 38389.459,
      "train_tokens_per_second": 12189.957
    },
    {
      "epoch": 0.9887557231349313,
      "grad_norm": 0.8573503854534872,
      "learning_rate": 5.088836485815568e-06,
      "loss": 0.3684356689453125,
      "num_input_tokens_seen": 468116992,
      "step": 14685,
      "train_runtime": 38402.4444,
      "train_tokens_per_second": 12189.771
    },
    {
      "epoch": 0.9890923781308915,
      "grad_norm": 0.7733438064653952,
      "learning_rate": 5.086192808732809e-06,
      "loss": 0.36145625114440916,
      "num_input_tokens_seen": 468270736,
      "step": 14690,
      "train_runtime": 38415.0685,
      "train_tokens_per_second": 12189.767
    },
    {
      "epoch": 0.9894290331268516,
      "grad_norm": 0.8379262214307905,
      "learning_rate": 5.083549107546505e-06,
      "loss": 0.37652878761291503,
      "num_input_tokens_seen": 468423360,
      "step": 14695,
      "train_runtime": 38428.3322,
      "train_tokens_per_second": 12189.531
    },
    {
      "epoch": 0.9897656881228117,
      "grad_norm": 0.7716888898498779,
      "learning_rate": 5.080905382995961e-06,
      "loss": 0.3989598274230957,
      "num_input_tokens_seen": 468585256,
      "step": 14700,
      "train_runtime": 38440.8723,
      "train_tokens_per_second": 12189.766
    },
    {
      "epoch": 0.9901023431187719,
      "grad_norm": 0.7035478532135641,
      "learning_rate": 5.0782616358204835e-06,
      "loss": 0.38455634117126464,
      "num_input_tokens_seen": 468744608,
      "step": 14705,
      "train_runtime": 38454.0633,
      "train_tokens_per_second": 12189.729
    },
    {
      "epoch": 0.990438998114732,
      "grad_norm": 0.8230253775506062,
      "learning_rate": 5.075617866759389e-06,
      "loss": 0.3496450901031494,
      "num_input_tokens_seen": 468900016,
      "step": 14710,
      "train_runtime": 38466.8983,
      "train_tokens_per_second": 12189.702
    },
    {
      "epoch": 0.9907756531106922,
      "grad_norm": 0.7644415289687652,
      "learning_rate": 5.0729740765519995e-06,
      "loss": 0.41901111602783203,
      "num_input_tokens_seen": 469058616,
      "step": 14715,
      "train_runtime": 38480.1259,
      "train_tokens_per_second": 12189.633
    },
    {
      "epoch": 0.9911123081066523,
      "grad_norm": 0.7935556940637902,
      "learning_rate": 5.070330265937643e-06,
      "loss": 0.36202635765075686,
      "num_input_tokens_seen": 469213944,
      "step": 14720,
      "train_runtime": 38493.5655,
      "train_tokens_per_second": 12189.412
    },
    {
      "epoch": 0.9914489631026124,
      "grad_norm": 0.7777524920648146,
      "learning_rate": 5.067686435655651e-06,
      "loss": 0.38873977661132814,
      "num_input_tokens_seen": 469377120,
      "step": 14725,
      "train_runtime": 38507.1609,
      "train_tokens_per_second": 12189.346
    },
    {
      "epoch": 0.9917856180985726,
      "grad_norm": 0.767938122684092,
      "learning_rate": 5.065042586445363e-06,
      "loss": 0.38827083110809324,
      "num_input_tokens_seen": 469531408,
      "step": 14730,
      "train_runtime": 38520.1831,
      "train_tokens_per_second": 12189.231
    },
    {
      "epoch": 0.9921222730945327,
      "grad_norm": 0.7672822762349635,
      "learning_rate": 5.0623987190461245e-06,
      "loss": 0.33096330165863036,
      "num_input_tokens_seen": 469687784,
      "step": 14735,
      "train_runtime": 38533.0715,
      "train_tokens_per_second": 12189.212
    },
    {
      "epoch": 0.9924589280904929,
      "grad_norm": 0.8592535902194339,
      "learning_rate": 5.059754834197281e-06,
      "loss": 0.38752386569976804,
      "num_input_tokens_seen": 469847608,
      "step": 14740,
      "train_runtime": 38546.3236,
      "train_tokens_per_second": 12189.168
    },
    {
      "epoch": 0.992795583086453,
      "grad_norm": 0.6701916289320892,
      "learning_rate": 5.05711093263819e-06,
      "loss": 0.3065890073776245,
      "num_input_tokens_seen": 470008320,
      "step": 14745,
      "train_runtime": 38560.0845,
      "train_tokens_per_second": 12188.986
    },
    {
      "epoch": 0.9931322380824131,
      "grad_norm": 0.8121691015014327,
      "learning_rate": 5.054467015108208e-06,
      "loss": 0.40102643966674806,
      "num_input_tokens_seen": 470171736,
      "step": 14750,
      "train_runtime": 38573.0821,
      "train_tokens_per_second": 12189.115
    },
    {
      "epoch": 0.9934688930783733,
      "grad_norm": 0.7849767621136483,
      "learning_rate": 5.051823082346701e-06,
      "loss": 0.3718144178390503,
      "num_input_tokens_seen": 470329800,
      "step": 14755,
      "train_runtime": 38585.9969,
      "train_tokens_per_second": 12189.132
    },
    {
      "epoch": 0.9938055480743334,
      "grad_norm": 0.8431159544435034,
      "learning_rate": 5.049179135093033e-06,
      "loss": 0.37436881065368655,
      "num_input_tokens_seen": 470490288,
      "step": 14760,
      "train_runtime": 38599.5296,
      "train_tokens_per_second": 12189.016
    },
    {
      "epoch": 0.9941422030702936,
      "grad_norm": 0.7264065460372751,
      "learning_rate": 5.046535174086575e-06,
      "loss": 0.382991099357605,
      "num_input_tokens_seen": 470650976,
      "step": 14765,
      "train_runtime": 38612.3934,
      "train_tokens_per_second": 12189.117
    },
    {
      "epoch": 0.9944788580662537,
      "grad_norm": 0.8081427587146074,
      "learning_rate": 5.043891200066706e-06,
      "loss": 0.36737618446350095,
      "num_input_tokens_seen": 470812552,
      "step": 14770,
      "train_runtime": 38625.6988,
      "train_tokens_per_second": 12189.101
    },
    {
      "epoch": 0.9948155130622138,
      "grad_norm": 0.7468630070744658,
      "learning_rate": 5.041247213772802e-06,
      "loss": 0.3550856113433838,
      "num_input_tokens_seen": 470976392,
      "step": 14775,
      "train_runtime": 38638.1326,
      "train_tokens_per_second": 12189.419
    },
    {
      "epoch": 0.995152168058174,
      "grad_norm": 0.8661324517569222,
      "learning_rate": 5.038603215944247e-06,
      "loss": 0.37668414115905763,
      "num_input_tokens_seen": 471125648,
      "step": 14780,
      "train_runtime": 38651.6107,
      "train_tokens_per_second": 12189.03
    },
    {
      "epoch": 0.9954888230541341,
      "grad_norm": 0.7998243342432169,
      "learning_rate": 5.0359592073204254e-06,
      "loss": 0.3421839714050293,
      "num_input_tokens_seen": 471281056,
      "step": 14785,
      "train_runtime": 38664.3506,
      "train_tokens_per_second": 12189.033
    },
    {
      "epoch": 0.9958254780500942,
      "grad_norm": 0.8347736664971019,
      "learning_rate": 5.033315188640725e-06,
      "loss": 0.3480769395828247,
      "num_input_tokens_seen": 471444288,
      "step": 14790,
      "train_runtime": 38677.2441,
      "train_tokens_per_second": 12189.19
    },
    {
      "epoch": 0.9961621330460544,
      "grad_norm": 0.7088835818485234,
      "learning_rate": 5.03067116064454e-06,
      "loss": 0.3531078815460205,
      "num_input_tokens_seen": 471607544,
      "step": 14795,
      "train_runtime": 38690.3087,
      "train_tokens_per_second": 12189.294
    },
    {
      "epoch": 0.9964987880420145,
      "grad_norm": 0.7102750769536917,
      "learning_rate": 5.028027124071262e-06,
      "loss": 0.3675748348236084,
      "num_input_tokens_seen": 471763328,
      "step": 14800,
      "train_runtime": 38703.2442,
      "train_tokens_per_second": 12189.245
    },
    {
      "epoch": 0.9968354430379747,
      "grad_norm": 0.7142184067704911,
      "learning_rate": 5.025383079660289e-06,
      "loss": 0.3480186700820923,
      "num_input_tokens_seen": 471926824,
      "step": 14805,
      "train_runtime": 38716.2587,
      "train_tokens_per_second": 12189.371
    },
    {
      "epoch": 0.9971720980339348,
      "grad_norm": 0.772594813278471,
      "learning_rate": 5.0227390281510166e-06,
      "loss": 0.3718601703643799,
      "num_input_tokens_seen": 472085760,
      "step": 14810,
      "train_runtime": 38728.836,
      "train_tokens_per_second": 12189.516
    },
    {
      "epoch": 0.997508753029895,
      "grad_norm": 0.7265510590897472,
      "learning_rate": 5.020094970282849e-06,
      "loss": 0.3606245517730713,
      "num_input_tokens_seen": 472238096,
      "step": 14815,
      "train_runtime": 38742.0492,
      "train_tokens_per_second": 12189.291
    },
    {
      "epoch": 0.9978454080258551,
      "grad_norm": 0.7933018562893929,
      "learning_rate": 5.017450906795187e-06,
      "loss": 0.3589968204498291,
      "num_input_tokens_seen": 472399544,
      "step": 14820,
      "train_runtime": 38754.5831,
      "train_tokens_per_second": 12189.514
    },
    {
      "epoch": 0.9981820630218152,
      "grad_norm": 0.7360374145233007,
      "learning_rate": 5.014806838427434e-06,
      "loss": 0.37145347595214845,
      "num_input_tokens_seen": 472557768,
      "step": 14825,
      "train_runtime": 38769.5693,
      "train_tokens_per_second": 12188.884
    },
    {
      "epoch": 0.9985187180177754,
      "grad_norm": 0.737675632861731,
      "learning_rate": 5.0121627659189964e-06,
      "loss": 0.36044859886169434,
      "num_input_tokens_seen": 472719752,
      "step": 14830,
      "train_runtime": 38782.0732,
      "train_tokens_per_second": 12189.131
    },
    {
      "epoch": 0.9988553730137355,
      "grad_norm": 0.9154770931403123,
      "learning_rate": 5.009518690009281e-06,
      "loss": 0.36890339851379395,
      "num_input_tokens_seen": 472873712,
      "step": 14835,
      "train_runtime": 38794.8734,
      "train_tokens_per_second": 12189.077
    },
    {
      "epoch": 0.9991920280096956,
      "grad_norm": 0.7973189235618524,
      "learning_rate": 5.006874611437692e-06,
      "loss": 0.3721126079559326,
      "num_input_tokens_seen": 473036384,
      "step": 14840,
      "train_runtime": 38807.7055,
      "train_tokens_per_second": 12189.239
    },
    {
      "epoch": 0.9995286830056558,
      "grad_norm": 0.7066298729868201,
      "learning_rate": 5.004230530943641e-06,
      "loss": 0.35857548713684084,
      "num_input_tokens_seen": 473199600,
      "step": 14845,
      "train_runtime": 38821.3327,
      "train_tokens_per_second": 12189.164
    },
    {
      "epoch": 0.9998653380016159,
      "grad_norm": 0.7674343390337574,
      "learning_rate": 5.001586449266536e-06,
      "loss": 0.3374613046646118,
      "num_input_tokens_seen": 473361976,
      "step": 14850,
      "train_runtime": 38833.9612,
      "train_tokens_per_second": 12189.382
    },
    {
      "epoch": 1.000201992997576,
      "grad_norm": 0.6705871939026546,
      "learning_rate": 4.998942367145787e-06,
      "loss": 0.34558203220367434,
      "num_input_tokens_seen": 473519176,
      "step": 14855,
      "train_runtime": 39005.0494,
      "train_tokens_per_second": 12139.946
    },
    {
      "epoch": 1.0005386479935363,
      "grad_norm": 0.7966159468716607,
      "learning_rate": 4.996298285320798e-06,
      "loss": 0.2983903408050537,
      "num_input_tokens_seen": 473670848,
      "step": 14860,
      "train_runtime": 39017.8807,
      "train_tokens_per_second": 12139.841
    },
    {
      "epoch": 1.0008753029894963,
      "grad_norm": 0.8062317848022944,
      "learning_rate": 4.993654204530983e-06,
      "loss": 0.3015679121017456,
      "num_input_tokens_seen": 473830880,
      "step": 14865,
      "train_runtime": 39030.7995,
      "train_tokens_per_second": 12139.922
    },
    {
      "epoch": 1.0012119579854566,
      "grad_norm": 0.7780150686596056,
      "learning_rate": 4.99101012551575e-06,
      "loss": 0.30467169284820556,
      "num_input_tokens_seen": 473979264,
      "step": 14870,
      "train_runtime": 39043.9053,
      "train_tokens_per_second": 12139.648
    },
    {
      "epoch": 1.0015486129814166,
      "grad_norm": 0.7867978083039806,
      "learning_rate": 4.9883660490145075e-06,
      "loss": 0.3058297157287598,
      "num_input_tokens_seen": 474143104,
      "step": 14875,
      "train_runtime": 39056.3373,
      "train_tokens_per_second": 12139.979
    },
    {
      "epoch": 1.0018852679773769,
      "grad_norm": 0.7340867135672074,
      "learning_rate": 4.985721975766661e-06,
      "loss": 0.3142852306365967,
      "num_input_tokens_seen": 474301104,
      "step": 14880,
      "train_runtime": 39069.2671,
      "train_tokens_per_second": 12140.005
    },
    {
      "epoch": 1.002221922973337,
      "grad_norm": 0.8211587526473608,
      "learning_rate": 4.983077906511619e-06,
      "loss": 0.3184364795684814,
      "num_input_tokens_seen": 474459776,
      "step": 14885,
      "train_runtime": 39082.3504,
      "train_tokens_per_second": 12140.001
    },
    {
      "epoch": 1.0025585779692971,
      "grad_norm": 0.7927471846307965,
      "learning_rate": 4.980433841988786e-06,
      "loss": 0.29326457977294923,
      "num_input_tokens_seen": 474619304,
      "step": 14890,
      "train_runtime": 39095.1809,
      "train_tokens_per_second": 12140.097
    },
    {
      "epoch": 1.0028952329652572,
      "grad_norm": 0.8056356008728264,
      "learning_rate": 4.977789782937568e-06,
      "loss": 0.30637235641479493,
      "num_input_tokens_seen": 474782752,
      "step": 14895,
      "train_runtime": 39108.1726,
      "train_tokens_per_second": 12140.244
    },
    {
      "epoch": 1.0032318879612174,
      "grad_norm": 0.8271742083544236,
      "learning_rate": 4.975145730097364e-06,
      "loss": 0.30994338989257814,
      "num_input_tokens_seen": 474937832,
      "step": 14900,
      "train_runtime": 39121.7025,
      "train_tokens_per_second": 12140.009
    },
    {
      "epoch": 1.0035685429571775,
      "grad_norm": 0.7572057918138807,
      "learning_rate": 4.97250168420758e-06,
      "loss": 0.30400514602661133,
      "num_input_tokens_seen": 475093640,
      "step": 14905,
      "train_runtime": 39134.5567,
      "train_tokens_per_second": 12140.003
    },
    {
      "epoch": 1.0039051979531377,
      "grad_norm": 0.8184971479869119,
      "learning_rate": 4.96985764600761e-06,
      "loss": 0.3087057590484619,
      "num_input_tokens_seen": 475257480,
      "step": 14910,
      "train_runtime": 39146.9774,
      "train_tokens_per_second": 12140.337
    },
    {
      "epoch": 1.0042418529490977,
      "grad_norm": 0.7551672835062946,
      "learning_rate": 4.967213616236855e-06,
      "loss": 0.31442787647247317,
      "num_input_tokens_seen": 475414480,
      "step": 14915,
      "train_runtime": 39159.9859,
      "train_tokens_per_second": 12140.313
    },
    {
      "epoch": 1.004578507945058,
      "grad_norm": 0.8907511821677292,
      "learning_rate": 4.964569595634706e-06,
      "loss": 0.31869916915893554,
      "num_input_tokens_seen": 475576136,
      "step": 14920,
      "train_runtime": 39172.4171,
      "train_tokens_per_second": 12140.587
    },
    {
      "epoch": 1.004915162941018,
      "grad_norm": 0.8038716472571167,
      "learning_rate": 4.961925584940559e-06,
      "loss": 0.3332134008407593,
      "num_input_tokens_seen": 475737936,
      "step": 14925,
      "train_runtime": 39184.9648,
      "train_tokens_per_second": 12140.828
    },
    {
      "epoch": 1.0052518179369783,
      "grad_norm": 0.796286764759513,
      "learning_rate": 4.959281584893799e-06,
      "loss": 0.27721872329711916,
      "num_input_tokens_seen": 475896760,
      "step": 14930,
      "train_runtime": 39197.544,
      "train_tokens_per_second": 12140.984
    },
    {
      "epoch": 1.0055884729329383,
      "grad_norm": 1.0596097827467876,
      "learning_rate": 4.956637596233814e-06,
      "loss": 0.31590662002563474,
      "num_input_tokens_seen": 476051992,
      "step": 14935,
      "train_runtime": 39211.4987,
      "train_tokens_per_second": 12140.622
    },
    {
      "epoch": 1.0059251279288985,
      "grad_norm": 0.7825866276852012,
      "learning_rate": 4.953993619699986e-06,
      "loss": 0.30376486778259276,
      "num_input_tokens_seen": 476210664,
      "step": 14940,
      "train_runtime": 39224.5654,
      "train_tokens_per_second": 12140.623
    },
    {
      "epoch": 1.0062617829248586,
      "grad_norm": 0.763076546094469,
      "learning_rate": 4.951349656031698e-06,
      "loss": 0.29827275276184084,
      "num_input_tokens_seen": 476372464,
      "step": 14945,
      "train_runtime": 39237.8088,
      "train_tokens_per_second": 12140.649
    },
    {
      "epoch": 1.0065984379208188,
      "grad_norm": 0.8114419181000545,
      "learning_rate": 4.948705705968322e-06,
      "loss": 0.3202513694763184,
      "num_input_tokens_seen": 476533456,
      "step": 14950,
      "train_runtime": 39250.8261,
      "train_tokens_per_second": 12140.724
    },
    {
      "epoch": 1.0069350929167789,
      "grad_norm": 0.718462429090509,
      "learning_rate": 4.946061770249233e-06,
      "loss": 0.303941535949707,
      "num_input_tokens_seen": 476695024,
      "step": 14955,
      "train_runtime": 39264.5568,
      "train_tokens_per_second": 12140.594
    },
    {
      "epoch": 1.007271747912739,
      "grad_norm": 0.7391776582057842,
      "learning_rate": 4.9434178496137955e-06,
      "loss": 0.32828574180603026,
      "num_input_tokens_seen": 476855264,
      "step": 14960,
      "train_runtime": 39277.6671,
      "train_tokens_per_second": 12140.621
    },
    {
      "epoch": 1.0076084029086991,
      "grad_norm": 0.7250506471342612,
      "learning_rate": 4.940773944801378e-06,
      "loss": 0.30356998443603517,
      "num_input_tokens_seen": 477014400,
      "step": 14965,
      "train_runtime": 39291.2399,
      "train_tokens_per_second": 12140.477
    },
    {
      "epoch": 1.0079450579046594,
      "grad_norm": 0.9205406447341752,
      "learning_rate": 4.938130056551337e-06,
      "loss": 0.32108659744262696,
      "num_input_tokens_seen": 477174616,
      "step": 14970,
      "train_runtime": 39303.921,
      "train_tokens_per_second": 12140.636
    },
    {
      "epoch": 1.0082817129006194,
      "grad_norm": 0.7214209925335233,
      "learning_rate": 4.9354861856030305e-06,
      "loss": 0.3137290239334106,
      "num_input_tokens_seen": 477333264,
      "step": 14975,
      "train_runtime": 39316.381,
      "train_tokens_per_second": 12140.824
    },
    {
      "epoch": 1.0086183678965797,
      "grad_norm": 0.8269121707839991,
      "learning_rate": 4.932842332695803e-06,
      "loss": 0.2959961175918579,
      "num_input_tokens_seen": 477489120,
      "step": 14980,
      "train_runtime": 39330.4954,
      "train_tokens_per_second": 12140.43
    },
    {
      "epoch": 1.0089550228925397,
      "grad_norm": 0.7952572939709217,
      "learning_rate": 4.9301984985690055e-06,
      "loss": 0.3175344467163086,
      "num_input_tokens_seen": 477652832,
      "step": 14985,
      "train_runtime": 39343.481,
      "train_tokens_per_second": 12140.584
    },
    {
      "epoch": 1.0092916778885,
      "grad_norm": 0.8855264278942446,
      "learning_rate": 4.927554683961974e-06,
      "loss": 0.3260321617126465,
      "num_input_tokens_seen": 477810192,
      "step": 14990,
      "train_runtime": 39356.2329,
      "train_tokens_per_second": 12140.649
    },
    {
      "epoch": 1.00962833288446,
      "grad_norm": 0.7564543412301691,
      "learning_rate": 4.924910889614046e-06,
      "loss": 0.2932924747467041,
      "num_input_tokens_seen": 477971256,
      "step": 14995,
      "train_runtime": 39370.0613,
      "train_tokens_per_second": 12140.475
    },
    {
      "epoch": 1.0099649878804202,
      "grad_norm": 0.8076110068815173,
      "learning_rate": 4.922267116264546e-06,
      "loss": 0.32979135513305663,
      "num_input_tokens_seen": 478123976,
      "step": 15000,
      "train_runtime": 39382.8246,
      "train_tokens_per_second": 12140.419
    },
    {
      "epoch": 1.0103016428763802,
      "grad_norm": 0.8172184811454559,
      "learning_rate": 4.919623364652801e-06,
      "loss": 0.29823894500732423,
      "num_input_tokens_seen": 478281480,
      "step": 15005,
      "train_runtime": 39395.7931,
      "train_tokens_per_second": 12140.42
    },
    {
      "epoch": 1.0106382978723405,
      "grad_norm": 0.7882461689682837,
      "learning_rate": 4.916979635518124e-06,
      "loss": 0.32759990692138674,
      "num_input_tokens_seen": 478435192,
      "step": 15010,
      "train_runtime": 39408.7817,
      "train_tokens_per_second": 12140.319
    },
    {
      "epoch": 1.0109749528683005,
      "grad_norm": 0.7904949057510042,
      "learning_rate": 4.914335929599828e-06,
      "loss": 0.3193901777267456,
      "num_input_tokens_seen": 478598288,
      "step": 15015,
      "train_runtime": 39421.6536,
      "train_tokens_per_second": 12140.492
    },
    {
      "epoch": 1.0113116078642608,
      "grad_norm": 0.89903611592992,
      "learning_rate": 4.9116922476372155e-06,
      "loss": 0.3195690870285034,
      "num_input_tokens_seen": 478759896,
      "step": 15020,
      "train_runtime": 39435.3711,
      "train_tokens_per_second": 12140.367
    },
    {
      "epoch": 1.0116482628602208,
      "grad_norm": 0.8455563216714693,
      "learning_rate": 4.909048590369586e-06,
      "loss": 0.30799388885498047,
      "num_input_tokens_seen": 478919224,
      "step": 15025,
      "train_runtime": 39448.5885,
      "train_tokens_per_second": 12140.339
    },
    {
      "epoch": 1.011984917856181,
      "grad_norm": 0.771178353861204,
      "learning_rate": 4.906404958536227e-06,
      "loss": 0.30479068756103517,
      "num_input_tokens_seen": 479080912,
      "step": 15030,
      "train_runtime": 39461.1228,
      "train_tokens_per_second": 12140.58
    },
    {
      "epoch": 1.012321572852141,
      "grad_norm": 0.7323087139637398,
      "learning_rate": 4.903761352876422e-06,
      "loss": 0.32252049446105957,
      "num_input_tokens_seen": 479236208,
      "step": 15035,
      "train_runtime": 39474.2336,
      "train_tokens_per_second": 12140.482
    },
    {
      "epoch": 1.0126582278481013,
      "grad_norm": 0.8654964782312182,
      "learning_rate": 4.9011177741294485e-06,
      "loss": 0.30508058071136473,
      "num_input_tokens_seen": 479390208,
      "step": 15040,
      "train_runtime": 39487.4524,
      "train_tokens_per_second": 12140.317
    },
    {
      "epoch": 1.0129948828440614,
      "grad_norm": 0.7440129822001812,
      "learning_rate": 4.898474223034576e-06,
      "loss": 0.3174762725830078,
      "num_input_tokens_seen": 479553752,
      "step": 15045,
      "train_runtime": 39500.4856,
      "train_tokens_per_second": 12140.452
    },
    {
      "epoch": 1.0133315378400216,
      "grad_norm": 0.8558713265492676,
      "learning_rate": 4.895830700331061e-06,
      "loss": 0.32155675888061525,
      "num_input_tokens_seen": 479711504,
      "step": 15050,
      "train_runtime": 39514.3392,
      "train_tokens_per_second": 12140.188
    },
    {
      "epoch": 1.0136681928359816,
      "grad_norm": 0.845004201463467,
      "learning_rate": 4.893187206758158e-06,
      "loss": 0.3127194166183472,
      "num_input_tokens_seen": 479870504,
      "step": 15055,
      "train_runtime": 39527.5774,
      "train_tokens_per_second": 12140.145
    },
    {
      "epoch": 1.014004847831942,
      "grad_norm": 0.9235731007514596,
      "learning_rate": 4.890543743055115e-06,
      "loss": 0.2902012586593628,
      "num_input_tokens_seen": 480030408,
      "step": 15060,
      "train_runtime": 39540.0963,
      "train_tokens_per_second": 12140.345
    },
    {
      "epoch": 1.014341502827902,
      "grad_norm": 0.7785770715607198,
      "learning_rate": 4.887900309961163e-06,
      "loss": 0.3177165985107422,
      "num_input_tokens_seen": 480190616,
      "step": 15065,
      "train_runtime": 39553.7345,
      "train_tokens_per_second": 12140.209
    },
    {
      "epoch": 1.0146781578238622,
      "grad_norm": 0.7488225820561665,
      "learning_rate": 4.885256908215532e-06,
      "loss": 0.3219867706298828,
      "num_input_tokens_seen": 480347736,
      "step": 15070,
      "train_runtime": 39566.5653,
      "train_tokens_per_second": 12140.243
    },
    {
      "epoch": 1.0150148128198222,
      "grad_norm": 0.7641437941572442,
      "learning_rate": 4.88261353855744e-06,
      "loss": 0.30176339149475095,
      "num_input_tokens_seen": 480510640,
      "step": 15075,
      "train_runtime": 39580.0783,
      "train_tokens_per_second": 12140.214
    },
    {
      "epoch": 1.0153514678157824,
      "grad_norm": 0.7541865739481328,
      "learning_rate": 4.8799702017261e-06,
      "loss": 0.26839122772216795,
      "num_input_tokens_seen": 480668744,
      "step": 15080,
      "train_runtime": 39592.846,
      "train_tokens_per_second": 12140.293
    },
    {
      "epoch": 1.0156881228117425,
      "grad_norm": 0.8337459210320616,
      "learning_rate": 4.877326898460709e-06,
      "loss": 0.3106881856918335,
      "num_input_tokens_seen": 480829912,
      "step": 15085,
      "train_runtime": 39606.0808,
      "train_tokens_per_second": 12140.305
    },
    {
      "epoch": 1.0160247778077027,
      "grad_norm": 0.7239097954927662,
      "learning_rate": 4.87468362950046e-06,
      "loss": 0.27660608291625977,
      "num_input_tokens_seen": 480987872,
      "step": 15090,
      "train_runtime": 39619.4693,
      "train_tokens_per_second": 12140.19
    },
    {
      "epoch": 1.0163614328036628,
      "grad_norm": 0.7957268375717846,
      "learning_rate": 4.872040395584533e-06,
      "loss": 0.31780261993408204,
      "num_input_tokens_seen": 481144784,
      "step": 15095,
      "train_runtime": 39632.65,
      "train_tokens_per_second": 12140.111
    },
    {
      "epoch": 1.016698087799623,
      "grad_norm": 0.8244179066039832,
      "learning_rate": 4.869397197452104e-06,
      "loss": 0.27200262546539306,
      "num_input_tokens_seen": 481305144,
      "step": 15100,
      "train_runtime": 39645.3403,
      "train_tokens_per_second": 12140.27
    },
    {
      "epoch": 1.017034742795583,
      "grad_norm": 0.9390239650385678,
      "learning_rate": 4.866754035842329e-06,
      "loss": 0.29276602268218993,
      "num_input_tokens_seen": 481465384,
      "step": 15105,
      "train_runtime": 39659.2215,
      "train_tokens_per_second": 12140.061
    },
    {
      "epoch": 1.0173713977915433,
      "grad_norm": 0.8400657661915847,
      "learning_rate": 4.8641109114943666e-06,
      "loss": 0.29243738651275636,
      "num_input_tokens_seen": 481622480,
      "step": 15110,
      "train_runtime": 39671.9192,
      "train_tokens_per_second": 12140.136
    },
    {
      "epoch": 1.0177080527875033,
      "grad_norm": 0.7967285330560733,
      "learning_rate": 4.861467825147352e-06,
      "loss": 0.29627945423126223,
      "num_input_tokens_seen": 481781328,
      "step": 15115,
      "train_runtime": 39685.4654,
      "train_tokens_per_second": 12139.994
    },
    {
      "epoch": 1.0180447077834636,
      "grad_norm": 0.8288297329746581,
      "learning_rate": 4.858824777540419e-06,
      "loss": 0.29321043491363524,
      "num_input_tokens_seen": 481940288,
      "step": 15120,
      "train_runtime": 39698.0466,
      "train_tokens_per_second": 12140.151
    },
    {
      "epoch": 1.0183813627794236,
      "grad_norm": 0.816647911107894,
      "learning_rate": 4.856181769412686e-06,
      "loss": 0.3050651788711548,
      "num_input_tokens_seen": 482097384,
      "step": 15125,
      "train_runtime": 39711.9621,
      "train_tokens_per_second": 12139.853
    },
    {
      "epoch": 1.0187180177753838,
      "grad_norm": 0.8161940351398661,
      "learning_rate": 4.8535388015032644e-06,
      "loss": 0.3015030860900879,
      "num_input_tokens_seen": 482256688,
      "step": 15130,
      "train_runtime": 39725.415,
      "train_tokens_per_second": 12139.752
    },
    {
      "epoch": 1.0190546727713439,
      "grad_norm": 0.6881612122333698,
      "learning_rate": 4.850895874551248e-06,
      "loss": 0.3095236301422119,
      "num_input_tokens_seen": 482419000,
      "step": 15135,
      "train_runtime": 39738.4578,
      "train_tokens_per_second": 12139.852
    },
    {
      "epoch": 1.0193913277673041,
      "grad_norm": 0.7944609897360215,
      "learning_rate": 4.848252989295726e-06,
      "loss": 0.33499579429626464,
      "num_input_tokens_seen": 482575296,
      "step": 15140,
      "train_runtime": 39751.452,
      "train_tokens_per_second": 12139.816
    },
    {
      "epoch": 1.0197279827632642,
      "grad_norm": 0.7329336936586256,
      "learning_rate": 4.845610146475771e-06,
      "loss": 0.27798492908477784,
      "num_input_tokens_seen": 482734648,
      "step": 15145,
      "train_runtime": 39764.6239,
      "train_tokens_per_second": 12139.802
    },
    {
      "epoch": 1.0200646377592244,
      "grad_norm": 0.8686166400123077,
      "learning_rate": 4.842967346830447e-06,
      "loss": 0.3024160385131836,
      "num_input_tokens_seen": 482898008,
      "step": 15150,
      "train_runtime": 39777.6078,
      "train_tokens_per_second": 12139.946
    },
    {
      "epoch": 1.0204012927551844,
      "grad_norm": 0.8767784526793144,
      "learning_rate": 4.840324591098803e-06,
      "loss": 0.30704617500305176,
      "num_input_tokens_seen": 483059112,
      "step": 15155,
      "train_runtime": 39791.3706,
      "train_tokens_per_second": 12139.796
    },
    {
      "epoch": 1.0207379477511447,
      "grad_norm": 0.7904418653230822,
      "learning_rate": 4.83768188001988e-06,
      "loss": 0.3019458293914795,
      "num_input_tokens_seen": 483217504,
      "step": 15160,
      "train_runtime": 39804.5503,
      "train_tokens_per_second": 12139.755
    },
    {
      "epoch": 1.0210746027471047,
      "grad_norm": 0.7834581111448851,
      "learning_rate": 4.835039214332699e-06,
      "loss": 0.290245509147644,
      "num_input_tokens_seen": 483376688,
      "step": 15165,
      "train_runtime": 39817.4418,
      "train_tokens_per_second": 12139.823
    },
    {
      "epoch": 1.021411257743065,
      "grad_norm": 0.7200605971762654,
      "learning_rate": 4.832396594776277e-06,
      "loss": 0.2989678382873535,
      "num_input_tokens_seen": 483540280,
      "step": 15170,
      "train_runtime": 39829.8639,
      "train_tokens_per_second": 12140.144
    },
    {
      "epoch": 1.021747912739025,
      "grad_norm": 0.7271107576178001,
      "learning_rate": 4.82975402208961e-06,
      "loss": 0.30151023864746096,
      "num_input_tokens_seen": 483703224,
      "step": 15175,
      "train_runtime": 39844.1958,
      "train_tokens_per_second": 12139.867
    },
    {
      "epoch": 1.0220845677349852,
      "grad_norm": 0.8344344470827905,
      "learning_rate": 4.827111497011692e-06,
      "loss": 0.28261580467224123,
      "num_input_tokens_seen": 483867064,
      "step": 15180,
      "train_runtime": 39856.6263,
      "train_tokens_per_second": 12140.191
    },
    {
      "epoch": 1.0224212227309453,
      "grad_norm": 0.8614747317203577,
      "learning_rate": 4.82446902028149e-06,
      "loss": 0.30556159019470214,
      "num_input_tokens_seen": 484026848,
      "step": 15185,
      "train_runtime": 39869.3087,
      "train_tokens_per_second": 12140.337
    },
    {
      "epoch": 1.0227578777269055,
      "grad_norm": 0.8957681172242815,
      "learning_rate": 4.821826592637968e-06,
      "loss": 0.319193172454834,
      "num_input_tokens_seen": 484186120,
      "step": 15190,
      "train_runtime": 39881.961,
      "train_tokens_per_second": 12140.479
    },
    {
      "epoch": 1.0230945327228655,
      "grad_norm": 0.8000588457774874,
      "learning_rate": 4.8191842148200694e-06,
      "loss": 0.3015614986419678,
      "num_input_tokens_seen": 484349360,
      "step": 15195,
      "train_runtime": 39894.8505,
      "train_tokens_per_second": 12140.649
    },
    {
      "epoch": 1.0234311877188258,
      "grad_norm": 0.8023934231191953,
      "learning_rate": 4.8165418875667315e-06,
      "loss": 0.294821572303772,
      "num_input_tokens_seen": 484511680,
      "step": 15200,
      "train_runtime": 39907.4656,
      "train_tokens_per_second": 12140.878
    },
    {
      "epoch": 1.0237678427147858,
      "grad_norm": 0.6950426551421778,
      "learning_rate": 4.813899611616867e-06,
      "loss": 0.3134977102279663,
      "num_input_tokens_seen": 484672552,
      "step": 15205,
      "train_runtime": 39921.3082,
      "train_tokens_per_second": 12140.698
    },
    {
      "epoch": 1.024104497710746,
      "grad_norm": 0.8182813180711925,
      "learning_rate": 4.8112573877093855e-06,
      "loss": 0.2890610218048096,
      "num_input_tokens_seen": 484835872,
      "step": 15210,
      "train_runtime": 39934.5117,
      "train_tokens_per_second": 12140.774
    },
    {
      "epoch": 1.024441152706706,
      "grad_norm": 0.8090918860339908,
      "learning_rate": 4.808615216583171e-06,
      "loss": 0.30469717979431155,
      "num_input_tokens_seen": 484993168,
      "step": 15215,
      "train_runtime": 39948.1935,
      "train_tokens_per_second": 12140.553
    },
    {
      "epoch": 1.0247778077026664,
      "grad_norm": 0.9146973303528697,
      "learning_rate": 4.805973098977102e-06,
      "loss": 0.30773401260375977,
      "num_input_tokens_seen": 485152816,
      "step": 15220,
      "train_runtime": 39962.2664,
      "train_tokens_per_second": 12140.273
    },
    {
      "epoch": 1.0251144626986264,
      "grad_norm": 0.8529999387159083,
      "learning_rate": 4.803331035630037e-06,
      "loss": 0.31885361671447754,
      "num_input_tokens_seen": 485313400,
      "step": 15225,
      "train_runtime": 39974.7027,
      "train_tokens_per_second": 12140.513
    },
    {
      "epoch": 1.0254511176945866,
      "grad_norm": 0.7727144658216679,
      "learning_rate": 4.800689027280822e-06,
      "loss": 0.2875771760940552,
      "num_input_tokens_seen": 485466152,
      "step": 15230,
      "train_runtime": 39987.2924,
      "train_tokens_per_second": 12140.511
    },
    {
      "epoch": 1.0257877726905467,
      "grad_norm": 0.8244214262196446,
      "learning_rate": 4.798047074668283e-06,
      "loss": 0.2990741491317749,
      "num_input_tokens_seen": 485626072,
      "step": 15235,
      "train_runtime": 40000.8223,
      "train_tokens_per_second": 12140.402
    },
    {
      "epoch": 1.026124427686507,
      "grad_norm": 0.7415401930431901,
      "learning_rate": 4.795405178531236e-06,
      "loss": 0.2889739751815796,
      "num_input_tokens_seen": 485780944,
      "step": 15240,
      "train_runtime": 40014.1338,
      "train_tokens_per_second": 12140.234
    },
    {
      "epoch": 1.026461082682467,
      "grad_norm": 0.8211435189469937,
      "learning_rate": 4.792763339608478e-06,
      "loss": 0.3030578374862671,
      "num_input_tokens_seen": 485934984,
      "step": 15245,
      "train_runtime": 40026.7785,
      "train_tokens_per_second": 12140.247
    },
    {
      "epoch": 1.0267977376784272,
      "grad_norm": 0.8978167915807699,
      "learning_rate": 4.790121558638793e-06,
      "loss": 0.2705661773681641,
      "num_input_tokens_seen": 486096664,
      "step": 15250,
      "train_runtime": 40039.599,
      "train_tokens_per_second": 12140.398
    },
    {
      "epoch": 1.0271343926743872,
      "grad_norm": 0.7609769763577684,
      "learning_rate": 4.787479836360944e-06,
      "loss": 0.3211428165435791,
      "num_input_tokens_seen": 486252624,
      "step": 15255,
      "train_runtime": 40052.2741,
      "train_tokens_per_second": 12140.45
    },
    {
      "epoch": 1.0274710476703475,
      "grad_norm": 0.7844643495384684,
      "learning_rate": 4.784838173513682e-06,
      "loss": 0.2951428651809692,
      "num_input_tokens_seen": 486411344,
      "step": 15260,
      "train_runtime": 40065.3555,
      "train_tokens_per_second": 12140.447
    },
    {
      "epoch": 1.0278077026663075,
      "grad_norm": 0.8478434989614024,
      "learning_rate": 4.782196570835737e-06,
      "loss": 0.2943997383117676,
      "num_input_tokens_seen": 486571448,
      "step": 15265,
      "train_runtime": 40077.9033,
      "train_tokens_per_second": 12140.641
    },
    {
      "epoch": 1.0281443576622677,
      "grad_norm": 0.8195626198569597,
      "learning_rate": 4.779555029065827e-06,
      "loss": 0.2975254774093628,
      "num_input_tokens_seen": 486724128,
      "step": 15270,
      "train_runtime": 40090.9795,
      "train_tokens_per_second": 12140.49
    },
    {
      "epoch": 1.0284810126582278,
      "grad_norm": 0.8116973535849498,
      "learning_rate": 4.776913548942649e-06,
      "loss": 0.30239067077636717,
      "num_input_tokens_seen": 486885704,
      "step": 15275,
      "train_runtime": 40103.5686,
      "train_tokens_per_second": 12140.708
    },
    {
      "epoch": 1.028817667654188,
      "grad_norm": 0.8172848907843281,
      "learning_rate": 4.774272131204889e-06,
      "loss": 0.3130920886993408,
      "num_input_tokens_seen": 487048192,
      "step": 15280,
      "train_runtime": 40116.5235,
      "train_tokens_per_second": 12140.837
    },
    {
      "epoch": 1.029154322650148,
      "grad_norm": 0.8923513541854086,
      "learning_rate": 4.7716307765912045e-06,
      "loss": 0.3016470670700073,
      "num_input_tokens_seen": 487210360,
      "step": 15285,
      "train_runtime": 40129.1641,
      "train_tokens_per_second": 12141.054
    },
    {
      "epoch": 1.0294909776461083,
      "grad_norm": 0.7992772255253157,
      "learning_rate": 4.768989485840246e-06,
      "loss": 0.3130969047546387,
      "num_input_tokens_seen": 487372048,
      "step": 15290,
      "train_runtime": 40142.0443,
      "train_tokens_per_second": 12141.187
    },
    {
      "epoch": 1.0298276326420683,
      "grad_norm": 0.8729193651340412,
      "learning_rate": 4.766348259690641e-06,
      "loss": 0.2877023220062256,
      "num_input_tokens_seen": 487529832,
      "step": 15295,
      "train_runtime": 40155.2168,
      "train_tokens_per_second": 12141.133
    },
    {
      "epoch": 1.0301642876380286,
      "grad_norm": 0.8206803317553585,
      "learning_rate": 4.763707098881002e-06,
      "loss": 0.30493950843811035,
      "num_input_tokens_seen": 487687704,
      "step": 15300,
      "train_runtime": 40168.2925,
      "train_tokens_per_second": 12141.111
    },
    {
      "epoch": 1.0305009426339886,
      "grad_norm": 0.8659920427933887,
      "learning_rate": 4.761066004149917e-06,
      "loss": 0.33200578689575194,
      "num_input_tokens_seen": 487843720,
      "step": 15305,
      "train_runtime": 40180.8918,
      "train_tokens_per_second": 12141.187
    },
    {
      "epoch": 1.0308375976299489,
      "grad_norm": 1.0594039325887077,
      "learning_rate": 4.758424976235964e-06,
      "loss": 0.32883241176605227,
      "num_input_tokens_seen": 488004552,
      "step": 15310,
      "train_runtime": 40194.3335,
      "train_tokens_per_second": 12141.128
    },
    {
      "epoch": 1.031174252625909,
      "grad_norm": 0.8979303497711981,
      "learning_rate": 4.755784015877694e-06,
      "loss": 0.3260646343231201,
      "num_input_tokens_seen": 488166296,
      "step": 15315,
      "train_runtime": 40206.8749,
      "train_tokens_per_second": 12141.364
    },
    {
      "epoch": 1.0315109076218691,
      "grad_norm": 0.8334221380814353,
      "learning_rate": 4.753143123813646e-06,
      "loss": 0.3162816524505615,
      "num_input_tokens_seen": 488322496,
      "step": 15320,
      "train_runtime": 40220.0663,
      "train_tokens_per_second": 12141.265
    },
    {
      "epoch": 1.0318475626178292,
      "grad_norm": 0.8623512113245309,
      "learning_rate": 4.7505023007823365e-06,
      "loss": 0.3234898328781128,
      "num_input_tokens_seen": 488482368,
      "step": 15325,
      "train_runtime": 40233.3571,
      "train_tokens_per_second": 12141.228
    },
    {
      "epoch": 1.0321842176137894,
      "grad_norm": 0.7508707514226148,
      "learning_rate": 4.747861547522264e-06,
      "loss": 0.2953406810760498,
      "num_input_tokens_seen": 488641888,
      "step": 15330,
      "train_runtime": 40245.7897,
      "train_tokens_per_second": 12141.441
    },
    {
      "epoch": 1.0325208726097495,
      "grad_norm": 0.7153252782703782,
      "learning_rate": 4.745220864771904e-06,
      "loss": 0.28333287239074706,
      "num_input_tokens_seen": 488795424,
      "step": 15335,
      "train_runtime": 40258.7318,
      "train_tokens_per_second": 12141.352
    },
    {
      "epoch": 1.0328575276057097,
      "grad_norm": 0.7183356579215424,
      "learning_rate": 4.742580253269718e-06,
      "loss": 0.31073908805847167,
      "num_input_tokens_seen": 488953024,
      "step": 15340,
      "train_runtime": 40272.2166,
      "train_tokens_per_second": 12141.2
    },
    {
      "epoch": 1.0331941826016697,
      "grad_norm": 0.8045495036028653,
      "learning_rate": 4.739939713754141e-06,
      "loss": 0.2809459209442139,
      "num_input_tokens_seen": 489110264,
      "step": 15345,
      "train_runtime": 40284.7736,
      "train_tokens_per_second": 12141.318
    },
    {
      "epoch": 1.03353083759763,
      "grad_norm": 0.8062160834455826,
      "learning_rate": 4.737299246963597e-06,
      "loss": 0.30010926723480225,
      "num_input_tokens_seen": 489271592,
      "step": 15350,
      "train_runtime": 40297.9144,
      "train_tokens_per_second": 12141.363
    },
    {
      "epoch": 1.03386749259359,
      "grad_norm": 0.8086162086935216,
      "learning_rate": 4.734658853636479e-06,
      "loss": 0.3008315086364746,
      "num_input_tokens_seen": 489430112,
      "step": 15355,
      "train_runtime": 40311.8248,
      "train_tokens_per_second": 12141.105
    },
    {
      "epoch": 1.0342041475895503,
      "grad_norm": 0.8431828814749335,
      "learning_rate": 4.732018534511167e-06,
      "loss": 0.33202786445617677,
      "num_input_tokens_seen": 489587040,
      "step": 15360,
      "train_runtime": 40324.4524,
      "train_tokens_per_second": 12141.195
    },
    {
      "epoch": 1.0345408025855103,
      "grad_norm": 0.817599883158712,
      "learning_rate": 4.729378290326017e-06,
      "loss": 0.2919569492340088,
      "num_input_tokens_seen": 489745608,
      "step": 15365,
      "train_runtime": 40338.6418,
      "train_tokens_per_second": 12140.855
    },
    {
      "epoch": 1.0348774575814705,
      "grad_norm": 0.7162527438810427,
      "learning_rate": 4.726738121819365e-06,
      "loss": 0.31473674774169924,
      "num_input_tokens_seen": 489902456,
      "step": 15370,
      "train_runtime": 40351.8533,
      "train_tokens_per_second": 12140.767
    },
    {
      "epoch": 1.0352141125774306,
      "grad_norm": 1.3334678646746396,
      "learning_rate": 4.7240980297295255e-06,
      "loss": 0.2772502422332764,
      "num_input_tokens_seen": 490058768,
      "step": 15375,
      "train_runtime": 40365.3463,
      "train_tokens_per_second": 12140.581
    },
    {
      "epoch": 1.0355507675733908,
      "grad_norm": 0.6610740765076342,
      "learning_rate": 4.721458014794795e-06,
      "loss": 0.3005315065383911,
      "num_input_tokens_seen": 490218904,
      "step": 15380,
      "train_runtime": 40378.5379,
      "train_tokens_per_second": 12140.581
    },
    {
      "epoch": 1.0358874225693508,
      "grad_norm": 0.9759511504699067,
      "learning_rate": 4.718818077753439e-06,
      "loss": 0.3232064723968506,
      "num_input_tokens_seen": 490381560,
      "step": 15385,
      "train_runtime": 40392.4697,
      "train_tokens_per_second": 12140.42
    },
    {
      "epoch": 1.036224077565311,
      "grad_norm": 0.7827665132811843,
      "learning_rate": 4.716178219343712e-06,
      "loss": 0.3020668029785156,
      "num_input_tokens_seen": 490538648,
      "step": 15390,
      "train_runtime": 40405.6835,
      "train_tokens_per_second": 12140.338
    },
    {
      "epoch": 1.0365607325612711,
      "grad_norm": 0.8852796795043774,
      "learning_rate": 4.713538440303841e-06,
      "loss": 0.3057693004608154,
      "num_input_tokens_seen": 490687424,
      "step": 15395,
      "train_runtime": 40418.9586,
      "train_tokens_per_second": 12140.031
    },
    {
      "epoch": 1.0368973875572314,
      "grad_norm": 0.7954186500254895,
      "learning_rate": 4.7108987413720335e-06,
      "loss": 0.3214967966079712,
      "num_input_tokens_seen": 490843712,
      "step": 15400,
      "train_runtime": 40431.484,
      "train_tokens_per_second": 12140.136
    },
    {
      "epoch": 1.0372340425531914,
      "grad_norm": 0.8506904845143792,
      "learning_rate": 4.708259123286469e-06,
      "loss": 0.3077725410461426,
      "num_input_tokens_seen": 490998184,
      "step": 15405,
      "train_runtime": 40444.6349,
      "train_tokens_per_second": 12140.008
    },
    {
      "epoch": 1.0375706975491517,
      "grad_norm": 0.8879981892697617,
      "learning_rate": 4.7056195867853105e-06,
      "loss": 0.318404483795166,
      "num_input_tokens_seen": 491159920,
      "step": 15410,
      "train_runtime": 40457.6275,
      "train_tokens_per_second": 12140.107
    },
    {
      "epoch": 1.0379073525451117,
      "grad_norm": 0.7731614759214138,
      "learning_rate": 4.702980132606696e-06,
      "loss": 0.28397555351257325,
      "num_input_tokens_seen": 491321184,
      "step": 15415,
      "train_runtime": 40470.599,
      "train_tokens_per_second": 12140.2
    },
    {
      "epoch": 1.038244007541072,
      "grad_norm": 0.803677194220429,
      "learning_rate": 4.700340761488741e-06,
      "loss": 0.2894084692001343,
      "num_input_tokens_seen": 491482376,
      "step": 15420,
      "train_runtime": 40484.2137,
      "train_tokens_per_second": 12140.099
    },
    {
      "epoch": 1.038580662537032,
      "grad_norm": 0.7286042600692628,
      "learning_rate": 4.6977014741695345e-06,
      "loss": 0.29176557064056396,
      "num_input_tokens_seen": 491643248,
      "step": 15425,
      "train_runtime": 40496.7104,
      "train_tokens_per_second": 12140.326
    },
    {
      "epoch": 1.0389173175329922,
      "grad_norm": 0.7395624413326506,
      "learning_rate": 4.69506227138715e-06,
      "loss": 0.31354827880859376,
      "num_input_tokens_seen": 491805800,
      "step": 15430,
      "train_runtime": 40509.3861,
      "train_tokens_per_second": 12140.539
    },
    {
      "epoch": 1.0392539725289522,
      "grad_norm": 0.8039203309499465,
      "learning_rate": 4.6924231538796265e-06,
      "loss": 0.32450170516967775,
      "num_input_tokens_seen": 491967504,
      "step": 15435,
      "train_runtime": 40522.391,
      "train_tokens_per_second": 12140.634
    },
    {
      "epoch": 1.0395906275249125,
      "grad_norm": 0.8032907221283301,
      "learning_rate": 4.6897841223849885e-06,
      "loss": 0.3096402883529663,
      "num_input_tokens_seen": 492127240,
      "step": 15440,
      "train_runtime": 40535.9591,
      "train_tokens_per_second": 12140.511
    },
    {
      "epoch": 1.0399272825208725,
      "grad_norm": 0.749873526646356,
      "learning_rate": 4.68714517764123e-06,
      "loss": 0.2843073606491089,
      "num_input_tokens_seen": 492286680,
      "step": 15445,
      "train_runtime": 40548.5266,
      "train_tokens_per_second": 12140.68
    },
    {
      "epoch": 1.0402639375168328,
      "grad_norm": 0.6587176431930899,
      "learning_rate": 4.684506320386327e-06,
      "loss": 0.28523907661437986,
      "num_input_tokens_seen": 492444520,
      "step": 15450,
      "train_runtime": 40561.3609,
      "train_tokens_per_second": 12140.73
    },
    {
      "epoch": 1.0406005925127928,
      "grad_norm": 0.7303682422559145,
      "learning_rate": 4.6818675513582235e-06,
      "loss": 0.28906548023223877,
      "num_input_tokens_seen": 492608048,
      "step": 15455,
      "train_runtime": 40574.3712,
      "train_tokens_per_second": 12140.867
    },
    {
      "epoch": 1.040937247508753,
      "grad_norm": 0.8194671774703502,
      "learning_rate": 4.679228871294846e-06,
      "loss": 0.27915356159210203,
      "num_input_tokens_seen": 492768152,
      "step": 15460,
      "train_runtime": 40587.525,
      "train_tokens_per_second": 12140.877
    },
    {
      "epoch": 1.041273902504713,
      "grad_norm": 0.7888920505638816,
      "learning_rate": 4.67659028093409e-06,
      "loss": 0.31377453804016114,
      "num_input_tokens_seen": 492928560,
      "step": 15465,
      "train_runtime": 40600.8218,
      "train_tokens_per_second": 12140.852
    },
    {
      "epoch": 1.0416105575006733,
      "grad_norm": 0.8654971039360112,
      "learning_rate": 4.673951781013834e-06,
      "loss": 0.33304409980773925,
      "num_input_tokens_seen": 493085824,
      "step": 15470,
      "train_runtime": 40613.5563,
      "train_tokens_per_second": 12140.917
    },
    {
      "epoch": 1.0419472124966334,
      "grad_norm": 0.7575923199417002,
      "learning_rate": 4.67131337227192e-06,
      "loss": 0.3072779655456543,
      "num_input_tokens_seen": 493247632,
      "step": 15475,
      "train_runtime": 40627.283,
      "train_tokens_per_second": 12140.798
    },
    {
      "epoch": 1.0422838674925936,
      "grad_norm": 0.7684046271365783,
      "learning_rate": 4.668675055446175e-06,
      "loss": 0.31185169219970704,
      "num_input_tokens_seen": 493406584,
      "step": 15480,
      "train_runtime": 40640.9476,
      "train_tokens_per_second": 12140.627
    },
    {
      "epoch": 1.0426205224885536,
      "grad_norm": 0.9795255158061947,
      "learning_rate": 4.666036831274392e-06,
      "loss": 0.27005825042724607,
      "num_input_tokens_seen": 493561712,
      "step": 15485,
      "train_runtime": 40653.6877,
      "train_tokens_per_second": 12140.638
    },
    {
      "epoch": 1.0429571774845139,
      "grad_norm": 1.4538015681097096,
      "learning_rate": 4.663398700494346e-06,
      "loss": 0.3166783332824707,
      "num_input_tokens_seen": 493718920,
      "step": 15490,
      "train_runtime": 40666.7895,
      "train_tokens_per_second": 12140.593
    },
    {
      "epoch": 1.043293832480474,
      "grad_norm": 0.8044585270951892,
      "learning_rate": 4.66076066384378e-06,
      "loss": 0.3103484153747559,
      "num_input_tokens_seen": 493878224,
      "step": 15495,
      "train_runtime": 40679.3423,
      "train_tokens_per_second": 12140.762
    },
    {
      "epoch": 1.0436304874764342,
      "grad_norm": 0.6685193897646251,
      "learning_rate": 4.658122722060413e-06,
      "loss": 0.31473257541656496,
      "num_input_tokens_seen": 494038504,
      "step": 15500,
      "train_runtime": 40692.9579,
      "train_tokens_per_second": 12140.639
    },
    {
      "epoch": 1.0439671424723942,
      "grad_norm": 0.7640360760465873,
      "learning_rate": 4.655484875881935e-06,
      "loss": 0.32730019092559814,
      "num_input_tokens_seen": 494198512,
      "step": 15505,
      "train_runtime": 40706.1495,
      "train_tokens_per_second": 12140.635
    },
    {
      "epoch": 1.0443037974683544,
      "grad_norm": 0.836737033993298,
      "learning_rate": 4.652847126046014e-06,
      "loss": 0.3173715114593506,
      "num_input_tokens_seen": 494349400,
      "step": 15510,
      "train_runtime": 40720.3218,
      "train_tokens_per_second": 12140.115
    },
    {
      "epoch": 1.0446404524643145,
      "grad_norm": 0.7870292674168577,
      "learning_rate": 4.650209473290286e-06,
      "loss": 0.32094688415527345,
      "num_input_tokens_seen": 494509480,
      "step": 15515,
      "train_runtime": 40733.3751,
      "train_tokens_per_second": 12140.155
    },
    {
      "epoch": 1.0449771074602747,
      "grad_norm": 0.8205299701544333,
      "learning_rate": 4.647571918352366e-06,
      "loss": 0.3088592529296875,
      "num_input_tokens_seen": 494667656,
      "step": 15520,
      "train_runtime": 40745.9057,
      "train_tokens_per_second": 12140.303
    },
    {
      "epoch": 1.0453137624562348,
      "grad_norm": 0.7961879693116772,
      "learning_rate": 4.644934461969833e-06,
      "loss": 0.3034135103225708,
      "num_input_tokens_seen": 494829920,
      "step": 15525,
      "train_runtime": 40758.3436,
      "train_tokens_per_second": 12140.58
    },
    {
      "epoch": 1.045650417452195,
      "grad_norm": 0.7129139091123158,
      "learning_rate": 4.642297104880247e-06,
      "loss": 0.2811610698699951,
      "num_input_tokens_seen": 494991504,
      "step": 15530,
      "train_runtime": 40772.0416,
      "train_tokens_per_second": 12140.464
    },
    {
      "epoch": 1.045987072448155,
      "grad_norm": 0.7970021500580954,
      "learning_rate": 4.639659847821133e-06,
      "loss": 0.2915218353271484,
      "num_input_tokens_seen": 495151544,
      "step": 15535,
      "train_runtime": 40785.6407,
      "train_tokens_per_second": 12140.34
    },
    {
      "epoch": 1.0463237274441153,
      "grad_norm": 0.8432601612760606,
      "learning_rate": 4.6370226915299945e-06,
      "loss": 0.3025413274765015,
      "num_input_tokens_seen": 495309792,
      "step": 15540,
      "train_runtime": 40799.0033,
      "train_tokens_per_second": 12140.242
    },
    {
      "epoch": 1.0466603824400753,
      "grad_norm": 0.7736505353016837,
      "learning_rate": 4.634385636744302e-06,
      "loss": 0.31579957008361814,
      "num_input_tokens_seen": 495472608,
      "step": 15545,
      "train_runtime": 40811.6696,
      "train_tokens_per_second": 12140.464
    },
    {
      "epoch": 1.0469970374360356,
      "grad_norm": 0.8542988788107384,
      "learning_rate": 4.631748684201503e-06,
      "loss": 0.3162078380584717,
      "num_input_tokens_seen": 495632320,
      "step": 15550,
      "train_runtime": 40824.0897,
      "train_tokens_per_second": 12140.683
    },
    {
      "epoch": 1.0473336924319956,
      "grad_norm": 0.7574058631996403,
      "learning_rate": 4.629111834639008e-06,
      "loss": 0.28935353755950927,
      "num_input_tokens_seen": 495792040,
      "step": 15555,
      "train_runtime": 40836.7542,
      "train_tokens_per_second": 12140.829
    },
    {
      "epoch": 1.0476703474279558,
      "grad_norm": 0.7752268236907609,
      "learning_rate": 4.626475088794206e-06,
      "loss": 0.29108147621154784,
      "num_input_tokens_seen": 495952568,
      "step": 15560,
      "train_runtime": 40849.184,
      "train_tokens_per_second": 12141.064
    },
    {
      "epoch": 1.0480070024239159,
      "grad_norm": 0.8281505396948898,
      "learning_rate": 4.623838447404454e-06,
      "loss": 0.3071067094802856,
      "num_input_tokens_seen": 496114448,
      "step": 15565,
      "train_runtime": 40862.3416,
      "train_tokens_per_second": 12141.116
    },
    {
      "epoch": 1.0483436574198761,
      "grad_norm": 0.7021233706746992,
      "learning_rate": 4.621201911207082e-06,
      "loss": 0.3025949478149414,
      "num_input_tokens_seen": 496271224,
      "step": 15570,
      "train_runtime": 40875.5846,
      "train_tokens_per_second": 12141.018
    },
    {
      "epoch": 1.0486803124158361,
      "grad_norm": 0.7382981012676141,
      "learning_rate": 4.618565480939387e-06,
      "loss": 0.3078681468963623,
      "num_input_tokens_seen": 496434328,
      "step": 15575,
      "train_runtime": 40888.4466,
      "train_tokens_per_second": 12141.188
    },
    {
      "epoch": 1.0490169674117964,
      "grad_norm": 0.8784045966814769,
      "learning_rate": 4.6159291573386395e-06,
      "loss": 0.30956554412841797,
      "num_input_tokens_seen": 496592456,
      "step": 15580,
      "train_runtime": 40901.0196,
      "train_tokens_per_second": 12141.322
    },
    {
      "epoch": 1.0493536224077564,
      "grad_norm": 0.7080734204547475,
      "learning_rate": 4.613292941142077e-06,
      "loss": 0.309919548034668,
      "num_input_tokens_seen": 496751808,
      "step": 15585,
      "train_runtime": 40913.4755,
      "train_tokens_per_second": 12141.521
    },
    {
      "epoch": 1.0496902774037167,
      "grad_norm": 0.8020327765750802,
      "learning_rate": 4.610656833086912e-06,
      "loss": 0.30821919441223145,
      "num_input_tokens_seen": 496905200,
      "step": 15590,
      "train_runtime": 40926.6312,
      "train_tokens_per_second": 12141.366
    },
    {
      "epoch": 1.0500269323996767,
      "grad_norm": 0.9070846713266137,
      "learning_rate": 4.60802083391032e-06,
      "loss": 0.3466046333312988,
      "num_input_tokens_seen": 497064136,
      "step": 15595,
      "train_runtime": 40939.7696,
      "train_tokens_per_second": 12141.352
    },
    {
      "epoch": 1.050363587395637,
      "grad_norm": 0.7913533398088878,
      "learning_rate": 4.605384944349454e-06,
      "loss": 0.277728796005249,
      "num_input_tokens_seen": 497221256,
      "step": 15600,
      "train_runtime": 40952.8472,
      "train_tokens_per_second": 12141.311
    },
    {
      "epoch": 1.050700242391597,
      "grad_norm": 0.7437937794846986,
      "learning_rate": 4.602749165141429e-06,
      "loss": 0.2947253942489624,
      "num_input_tokens_seen": 497377224,
      "step": 15605,
      "train_runtime": 40966.7459,
      "train_tokens_per_second": 12140.999
    },
    {
      "epoch": 1.0510368973875572,
      "grad_norm": 0.7639512433659383,
      "learning_rate": 4.600113497023333e-06,
      "loss": 0.32112808227539064,
      "num_input_tokens_seen": 497536024,
      "step": 15610,
      "train_runtime": 40979.9151,
      "train_tokens_per_second": 12140.973
    },
    {
      "epoch": 1.0513735523835175,
      "grad_norm": 0.7825943430599259,
      "learning_rate": 4.597477940732222e-06,
      "loss": 0.29928107261657716,
      "num_input_tokens_seen": 497691088,
      "step": 15615,
      "train_runtime": 40992.8746,
      "train_tokens_per_second": 12140.917
    },
    {
      "epoch": 1.0517102073794775,
      "grad_norm": 0.8224364231495325,
      "learning_rate": 4.594842497005124e-06,
      "loss": 0.29639034271240233,
      "num_input_tokens_seen": 497851400,
      "step": 15620,
      "train_runtime": 41005.7774,
      "train_tokens_per_second": 12141.006
    },
    {
      "epoch": 1.0520468623754375,
      "grad_norm": 0.71963143504583,
      "learning_rate": 4.5922071665790264e-06,
      "loss": 0.3243539810180664,
      "num_input_tokens_seen": 498008624,
      "step": 15625,
      "train_runtime": 41018.8641,
      "train_tokens_per_second": 12140.966
    },
    {
      "epoch": 1.0523835173713978,
      "grad_norm": 0.7394806230325817,
      "learning_rate": 4.589571950190897e-06,
      "loss": 0.3089911460876465,
      "num_input_tokens_seen": 498172464,
      "step": 15630,
      "train_runtime": 41031.3015,
      "train_tokens_per_second": 12141.279
    },
    {
      "epoch": 1.052720172367358,
      "grad_norm": 0.7693978980750562,
      "learning_rate": 4.586936848577661e-06,
      "loss": 0.3002563238143921,
      "num_input_tokens_seen": 498333256,
      "step": 15635,
      "train_runtime": 41044.9279,
      "train_tokens_per_second": 12141.165
    },
    {
      "epoch": 1.053056827363318,
      "grad_norm": 0.7220698823164534,
      "learning_rate": 4.584301862476219e-06,
      "loss": 0.3249997138977051,
      "num_input_tokens_seen": 498491424,
      "step": 15640,
      "train_runtime": 41057.9386,
      "train_tokens_per_second": 12141.17
    },
    {
      "epoch": 1.053393482359278,
      "grad_norm": 0.7580345413979345,
      "learning_rate": 4.581666992623435e-06,
      "loss": 0.2922648906707764,
      "num_input_tokens_seen": 498655264,
      "step": 15645,
      "train_runtime": 41070.395,
      "train_tokens_per_second": 12141.477
    },
    {
      "epoch": 1.0537301373552383,
      "grad_norm": 0.8604028925426824,
      "learning_rate": 4.579032239756144e-06,
      "loss": 0.30808842182159424,
      "num_input_tokens_seen": 498814264,
      "step": 15650,
      "train_runtime": 41083.889,
      "train_tokens_per_second": 12141.359
    },
    {
      "epoch": 1.0540667923511986,
      "grad_norm": 0.7604796423457374,
      "learning_rate": 4.576397604611143e-06,
      "loss": 0.3018012285232544,
      "num_input_tokens_seen": 498975768,
      "step": 15655,
      "train_runtime": 41097.1878,
      "train_tokens_per_second": 12141.36
    },
    {
      "epoch": 1.0544034473471586,
      "grad_norm": 0.7729971816117196,
      "learning_rate": 4.573763087925202e-06,
      "loss": 0.304763650894165,
      "num_input_tokens_seen": 499137976,
      "step": 15660,
      "train_runtime": 41110.3329,
      "train_tokens_per_second": 12141.424
    },
    {
      "epoch": 1.0547401023431189,
      "grad_norm": 0.8166783729445071,
      "learning_rate": 4.571128690435053e-06,
      "loss": 0.3164165258407593,
      "num_input_tokens_seen": 499299984,
      "step": 15665,
      "train_runtime": 41122.9174,
      "train_tokens_per_second": 12141.648
    },
    {
      "epoch": 1.055076757339079,
      "grad_norm": 0.9523824673439621,
      "learning_rate": 4.568494412877401e-06,
      "loss": 0.3166008472442627,
      "num_input_tokens_seen": 499454520,
      "step": 15670,
      "train_runtime": 41136.268,
      "train_tokens_per_second": 12141.464
    },
    {
      "epoch": 1.0554134123350392,
      "grad_norm": 0.8604205059395552,
      "learning_rate": 4.565860255988908e-06,
      "loss": 0.30739483833312986,
      "num_input_tokens_seen": 499610736,
      "step": 15675,
      "train_runtime": 41149.4833,
      "train_tokens_per_second": 12141.361
    },
    {
      "epoch": 1.0557500673309992,
      "grad_norm": 0.8599882474328476,
      "learning_rate": 4.563226220506212e-06,
      "loss": 0.3091917037963867,
      "num_input_tokens_seen": 499774576,
      "step": 15680,
      "train_runtime": 41161.932,
      "train_tokens_per_second": 12141.67
    },
    {
      "epoch": 1.0560867223269594,
      "grad_norm": 0.7166199216375662,
      "learning_rate": 4.5605923071659085e-06,
      "loss": 0.3108429193496704,
      "num_input_tokens_seen": 499932088,
      "step": 15685,
      "train_runtime": 41175.2927,
      "train_tokens_per_second": 12141.555
    },
    {
      "epoch": 1.0564233773229195,
      "grad_norm": 0.7783368747610865,
      "learning_rate": 4.557958516704566e-06,
      "loss": 0.29545657634735106,
      "num_input_tokens_seen": 500093016,
      "step": 15690,
      "train_runtime": 41188.8206,
      "train_tokens_per_second": 12141.475
    },
    {
      "epoch": 1.0567600323188797,
      "grad_norm": 0.7639467038269939,
      "learning_rate": 4.5553248498587144e-06,
      "loss": 0.30799593925476076,
      "num_input_tokens_seen": 500249704,
      "step": 15695,
      "train_runtime": 41201.8753,
      "train_tokens_per_second": 12141.43
    },
    {
      "epoch": 1.0570966873148397,
      "grad_norm": 0.855452169072172,
      "learning_rate": 4.552691307364853e-06,
      "loss": 0.31041455268859863,
      "num_input_tokens_seen": 500409136,
      "step": 15700,
      "train_runtime": 41215.9419,
      "train_tokens_per_second": 12141.155
    },
    {
      "epoch": 1.0574333423108,
      "grad_norm": 0.9157896787688669,
      "learning_rate": 4.5500578899594375e-06,
      "loss": 0.3277078866958618,
      "num_input_tokens_seen": 500570952,
      "step": 15705,
      "train_runtime": 41229.81,
      "train_tokens_per_second": 12140.996
    },
    {
      "epoch": 1.05776999730676,
      "grad_norm": 0.7489539433497033,
      "learning_rate": 4.5474245983789e-06,
      "loss": 0.3013786315917969,
      "num_input_tokens_seen": 500727568,
      "step": 15710,
      "train_runtime": 41242.2364,
      "train_tokens_per_second": 12141.135
    },
    {
      "epoch": 1.0581066523027203,
      "grad_norm": 0.740308006414045,
      "learning_rate": 4.54479143335963e-06,
      "loss": 0.29260926246643065,
      "num_input_tokens_seen": 500890432,
      "step": 15715,
      "train_runtime": 41255.0153,
      "train_tokens_per_second": 12141.322
    },
    {
      "epoch": 1.0584433072986803,
      "grad_norm": 0.8392210769341023,
      "learning_rate": 4.5421583956379855e-06,
      "loss": 0.317919921875,
      "num_input_tokens_seen": 501049616,
      "step": 15720,
      "train_runtime": 41268.4443,
      "train_tokens_per_second": 12141.229
    },
    {
      "epoch": 1.0587799622946406,
      "grad_norm": 0.7869549951404201,
      "learning_rate": 4.539525485950284e-06,
      "loss": 0.2929553508758545,
      "num_input_tokens_seen": 501213432,
      "step": 15725,
      "train_runtime": 41281.545,
      "train_tokens_per_second": 12141.344
    },
    {
      "epoch": 1.0591166172906006,
      "grad_norm": 0.9557886922130165,
      "learning_rate": 4.536892705032815e-06,
      "loss": 0.32071738243103026,
      "num_input_tokens_seen": 501372832,
      "step": 15730,
      "train_runtime": 41294.1911,
      "train_tokens_per_second": 12141.486
    },
    {
      "epoch": 1.0594532722865608,
      "grad_norm": 0.8043401353079279,
      "learning_rate": 4.534260053621821e-06,
      "loss": 0.29186296463012695,
      "num_input_tokens_seen": 501533064,
      "step": 15735,
      "train_runtime": 41307.8357,
      "train_tokens_per_second": 12141.354
    },
    {
      "epoch": 1.0597899272825209,
      "grad_norm": 0.8026413405454789,
      "learning_rate": 4.531627532453519e-06,
      "loss": 0.27665510177612307,
      "num_input_tokens_seen": 501691976,
      "step": 15740,
      "train_runtime": 41320.6017,
      "train_tokens_per_second": 12141.449
    },
    {
      "epoch": 1.0601265822784811,
      "grad_norm": 0.7889160677144242,
      "learning_rate": 4.5289951422640826e-06,
      "loss": 0.32256455421447755,
      "num_input_tokens_seen": 501847936,
      "step": 15745,
      "train_runtime": 41333.609,
      "train_tokens_per_second": 12141.401
    },
    {
      "epoch": 1.0604632372744411,
      "grad_norm": 0.826341886805559,
      "learning_rate": 4.526362883789655e-06,
      "loss": 0.29686803817749025,
      "num_input_tokens_seen": 502010752,
      "step": 15750,
      "train_runtime": 41346.5267,
      "train_tokens_per_second": 12141.546
    },
    {
      "epoch": 1.0607998922704014,
      "grad_norm": 0.9112805532666093,
      "learning_rate": 4.5237307577663345e-06,
      "loss": 0.33019020557403567,
      "num_input_tokens_seen": 502172528,
      "step": 15755,
      "train_runtime": 41359.0966,
      "train_tokens_per_second": 12141.767
    },
    {
      "epoch": 1.0611365472663614,
      "grad_norm": 0.6938062080422588,
      "learning_rate": 4.52109876493019e-06,
      "loss": 0.3061249256134033,
      "num_input_tokens_seen": 502334584,
      "step": 15760,
      "train_runtime": 41372.9119,
      "train_tokens_per_second": 12141.63
    },
    {
      "epoch": 1.0614732022623217,
      "grad_norm": 0.7370196937125292,
      "learning_rate": 4.518466906017249e-06,
      "loss": 0.3247066974639893,
      "num_input_tokens_seen": 502495160,
      "step": 15765,
      "train_runtime": 41386.4269,
      "train_tokens_per_second": 12141.545
    },
    {
      "epoch": 1.0618098572582817,
      "grad_norm": 0.8087408045747353,
      "learning_rate": 4.515835181763503e-06,
      "loss": 0.29228289127349855,
      "num_input_tokens_seen": 502654008,
      "step": 15770,
      "train_runtime": 41398.9834,
      "train_tokens_per_second": 12141.699
    },
    {
      "epoch": 1.062146512254242,
      "grad_norm": 0.7414272672545331,
      "learning_rate": 4.513203592904903e-06,
      "loss": 0.3032099962234497,
      "num_input_tokens_seen": 502815408,
      "step": 15775,
      "train_runtime": 41412.0057,
      "train_tokens_per_second": 12141.779
    },
    {
      "epoch": 1.062483167250202,
      "grad_norm": 0.7887934841677506,
      "learning_rate": 4.510572140177368e-06,
      "loss": 0.31247117519378664,
      "num_input_tokens_seen": 502979248,
      "step": 15780,
      "train_runtime": 41424.4474,
      "train_tokens_per_second": 12142.087
    },
    {
      "epoch": 1.0628198222461622,
      "grad_norm": 0.817120390119069,
      "learning_rate": 4.507940824316771e-06,
      "loss": 0.3199119567871094,
      "num_input_tokens_seen": 503138144,
      "step": 15785,
      "train_runtime": 41437.8429,
      "train_tokens_per_second": 12141.997
    },
    {
      "epoch": 1.0631564772421223,
      "grad_norm": 0.7163399295997693,
      "learning_rate": 4.505309646058954e-06,
      "loss": 0.3010636568069458,
      "num_input_tokens_seen": 503300400,
      "step": 15790,
      "train_runtime": 41451.2585,
      "train_tokens_per_second": 12141.981
    },
    {
      "epoch": 1.0634931322380825,
      "grad_norm": 0.8698374743160174,
      "learning_rate": 4.502678606139716e-06,
      "loss": 0.2943506956100464,
      "num_input_tokens_seen": 503459832,
      "step": 15795,
      "train_runtime": 41464.3194,
      "train_tokens_per_second": 12142.002
    },
    {
      "epoch": 1.0638297872340425,
      "grad_norm": 0.683418469449124,
      "learning_rate": 4.500047705294823e-06,
      "loss": 0.27575345039367677,
      "num_input_tokens_seen": 503622704,
      "step": 15800,
      "train_runtime": 41477.4484,
      "train_tokens_per_second": 12142.085
    },
    {
      "epoch": 1.0641664422300028,
      "grad_norm": 0.7433492874399087,
      "learning_rate": 4.497416944259993e-06,
      "loss": 0.30148606300354003,
      "num_input_tokens_seen": 503784960,
      "step": 15805,
      "train_runtime": 41491.405,
      "train_tokens_per_second": 12141.911
    },
    {
      "epoch": 1.0645030972259628,
      "grad_norm": 0.784045647285606,
      "learning_rate": 4.494786323770912e-06,
      "loss": 0.3191340684890747,
      "num_input_tokens_seen": 503946344,
      "step": 15810,
      "train_runtime": 41504.4859,
      "train_tokens_per_second": 12141.973
    },
    {
      "epoch": 1.064839752221923,
      "grad_norm": 0.8038057768448763,
      "learning_rate": 4.492155844563224e-06,
      "loss": 0.2987980842590332,
      "num_input_tokens_seen": 504106392,
      "step": 15815,
      "train_runtime": 41518.0631,
      "train_tokens_per_second": 12141.857
    },
    {
      "epoch": 1.065176407217883,
      "grad_norm": 0.8278255813891302,
      "learning_rate": 4.489525507372537e-06,
      "loss": 0.3317086696624756,
      "num_input_tokens_seen": 504260776,
      "step": 15820,
      "train_runtime": 41530.5376,
      "train_tokens_per_second": 12141.927
    },
    {
      "epoch": 1.0655130622138433,
      "grad_norm": 0.7605081211399658,
      "learning_rate": 4.486895312934411e-06,
      "loss": 0.29726648330688477,
      "num_input_tokens_seen": 504419984,
      "step": 15825,
      "train_runtime": 41543.5078,
      "train_tokens_per_second": 12141.969
    },
    {
      "epoch": 1.0658497172098034,
      "grad_norm": 0.8892114411078401,
      "learning_rate": 4.484265261984378e-06,
      "loss": 0.35415384769439695,
      "num_input_tokens_seen": 504583480,
      "step": 15830,
      "train_runtime": 41556.5287,
      "train_tokens_per_second": 12142.099
    },
    {
      "epoch": 1.0661863722057636,
      "grad_norm": 0.7911529635978167,
      "learning_rate": 4.4816353552579165e-06,
      "loss": 0.29717953205108644,
      "num_input_tokens_seen": 504740904,
      "step": 15835,
      "train_runtime": 41569.5041,
      "train_tokens_per_second": 12142.096
    },
    {
      "epoch": 1.0665230272017237,
      "grad_norm": 0.8164744324364961,
      "learning_rate": 4.479005593490477e-06,
      "loss": 0.3261962175369263,
      "num_input_tokens_seen": 504899536,
      "step": 15840,
      "train_runtime": 41582.1907,
      "train_tokens_per_second": 12142.206
    },
    {
      "epoch": 1.066859682197684,
      "grad_norm": 0.7472282273296579,
      "learning_rate": 4.476375977417461e-06,
      "loss": 0.2812601089477539,
      "num_input_tokens_seen": 505061560,
      "step": 15845,
      "train_runtime": 41594.7742,
      "train_tokens_per_second": 12142.428
    },
    {
      "epoch": 1.067196337193644,
      "grad_norm": 0.8945826388832476,
      "learning_rate": 4.473746507774235e-06,
      "loss": 0.30530753135681155,
      "num_input_tokens_seen": 505224600,
      "step": 15850,
      "train_runtime": 41607.6248,
      "train_tokens_per_second": 12142.597
    },
    {
      "epoch": 1.0675329921896042,
      "grad_norm": 0.796737877670132,
      "learning_rate": 4.471117185296117e-06,
      "loss": 0.28793628215789796,
      "num_input_tokens_seen": 505384632,
      "step": 15855,
      "train_runtime": 41620.5553,
      "train_tokens_per_second": 12142.669
    },
    {
      "epoch": 1.0678696471855642,
      "grad_norm": 0.7141899690343959,
      "learning_rate": 4.468488010718394e-06,
      "loss": 0.2760047435760498,
      "num_input_tokens_seen": 505540816,
      "step": 15860,
      "train_runtime": 41634.3455,
      "train_tokens_per_second": 12142.399
    },
    {
      "epoch": 1.0682063021815245,
      "grad_norm": 0.8149575204075473,
      "learning_rate": 4.465858984776302e-06,
      "loss": 0.3093507528305054,
      "num_input_tokens_seen": 505698672,
      "step": 15865,
      "train_runtime": 41647.2736,
      "train_tokens_per_second": 12142.42
    },
    {
      "epoch": 1.0685429571774845,
      "grad_norm": 0.7756964961522265,
      "learning_rate": 4.463230108205044e-06,
      "loss": 0.30608930587768557,
      "num_input_tokens_seen": 505849368,
      "step": 15870,
      "train_runtime": 41661.6765,
      "train_tokens_per_second": 12141.839
    },
    {
      "epoch": 1.0688796121734447,
      "grad_norm": 0.7299853686580681,
      "learning_rate": 4.460601381739772e-06,
      "loss": 0.30599393844604494,
      "num_input_tokens_seen": 506012200,
      "step": 15875,
      "train_runtime": 41675.1316,
      "train_tokens_per_second": 12141.826
    },
    {
      "epoch": 1.0692162671694048,
      "grad_norm": 0.8044127197515359,
      "learning_rate": 4.457972806115607e-06,
      "loss": 0.3312252044677734,
      "num_input_tokens_seen": 506171584,
      "step": 15880,
      "train_runtime": 41687.606,
      "train_tokens_per_second": 12142.016
    },
    {
      "epoch": 1.069552922165365,
      "grad_norm": 0.7622947813623563,
      "learning_rate": 4.455344382067616e-06,
      "loss": 0.29750833511352537,
      "num_input_tokens_seen": 506326216,
      "step": 15885,
      "train_runtime": 41700.1445,
      "train_tokens_per_second": 12142.073
    },
    {
      "epoch": 1.069889577161325,
      "grad_norm": 0.7900556405084451,
      "learning_rate": 4.452716110330832e-06,
      "loss": 0.3218410015106201,
      "num_input_tokens_seen": 506488256,
      "step": 15890,
      "train_runtime": 41713.1883,
      "train_tokens_per_second": 12142.161
    },
    {
      "epoch": 1.0702262321572853,
      "grad_norm": 0.7674155895751068,
      "learning_rate": 4.450087991640242e-06,
      "loss": 0.2956411123275757,
      "num_input_tokens_seen": 506642552,
      "step": 15895,
      "train_runtime": 41726.8616,
      "train_tokens_per_second": 12141.88
    },
    {
      "epoch": 1.0705628871532453,
      "grad_norm": 0.8400634289493644,
      "learning_rate": 4.447460026730794e-06,
      "loss": 0.31752021312713624,
      "num_input_tokens_seen": 506802544,
      "step": 15900,
      "train_runtime": 41740.0769,
      "train_tokens_per_second": 12141.869
    },
    {
      "epoch": 1.0708995421492056,
      "grad_norm": 0.8674125441207241,
      "learning_rate": 4.444832216337384e-06,
      "loss": 0.2998190879821777,
      "num_input_tokens_seen": 506961352,
      "step": 15905,
      "train_runtime": 41753.6472,
      "train_tokens_per_second": 12141.726
    },
    {
      "epoch": 1.0712361971451656,
      "grad_norm": 1.0146359541815608,
      "learning_rate": 4.442204561194877e-06,
      "loss": 0.30346121788024905,
      "num_input_tokens_seen": 507119544,
      "step": 15910,
      "train_runtime": 41766.8454,
      "train_tokens_per_second": 12141.677
    },
    {
      "epoch": 1.0715728521411259,
      "grad_norm": 0.7726994426895054,
      "learning_rate": 4.439577062038085e-06,
      "loss": 0.29607155323028567,
      "num_input_tokens_seen": 507283384,
      "step": 15915,
      "train_runtime": 41779.2663,
      "train_tokens_per_second": 12141.989
    },
    {
      "epoch": 1.0719095071370859,
      "grad_norm": 0.7931230438107604,
      "learning_rate": 4.436949719601782e-06,
      "loss": 0.32483975887298583,
      "num_input_tokens_seen": 507438968,
      "step": 15920,
      "train_runtime": 41792.2795,
      "train_tokens_per_second": 12141.931
    },
    {
      "epoch": 1.0722461621330461,
      "grad_norm": 0.8944388649128798,
      "learning_rate": 4.434322534620692e-06,
      "loss": 0.32542123794555666,
      "num_input_tokens_seen": 507591744,
      "step": 15925,
      "train_runtime": 41805.472,
      "train_tokens_per_second": 12141.754
    },
    {
      "epoch": 1.0725828171290062,
      "grad_norm": 0.7009415269611473,
      "learning_rate": 4.431695507829504e-06,
      "loss": 0.2663717269897461,
      "num_input_tokens_seen": 507752888,
      "step": 15930,
      "train_runtime": 41818.5581,
      "train_tokens_per_second": 12141.808
    },
    {
      "epoch": 1.0729194721249664,
      "grad_norm": 0.8945073836205689,
      "learning_rate": 4.429068639962853e-06,
      "loss": 0.3137186050415039,
      "num_input_tokens_seen": 507906136,
      "step": 15935,
      "train_runtime": 41831.3185,
      "train_tokens_per_second": 12141.767
    },
    {
      "epoch": 1.0732561271209264,
      "grad_norm": 0.7892931047698717,
      "learning_rate": 4.426441931755337e-06,
      "loss": 0.3003633260726929,
      "num_input_tokens_seen": 508068256,
      "step": 15940,
      "train_runtime": 41843.9013,
      "train_tokens_per_second": 12141.991
    },
    {
      "epoch": 1.0735927821168867,
      "grad_norm": 0.7050125545508502,
      "learning_rate": 4.4238153839415045e-06,
      "loss": 0.30908546447753904,
      "num_input_tokens_seen": 508229096,
      "step": 15945,
      "train_runtime": 41856.3796,
      "train_tokens_per_second": 12142.213
    },
    {
      "epoch": 1.0739294371128467,
      "grad_norm": 0.8246728697405513,
      "learning_rate": 4.421188997255865e-06,
      "loss": 0.3237894058227539,
      "num_input_tokens_seen": 508380424,
      "step": 15950,
      "train_runtime": 41869.1635,
      "train_tokens_per_second": 12142.12
    },
    {
      "epoch": 1.074266092108807,
      "grad_norm": 0.8006298435114627,
      "learning_rate": 4.418562772432874e-06,
      "loss": 0.31391019821166993,
      "num_input_tokens_seen": 508539688,
      "step": 15955,
      "train_runtime": 41882.3198,
      "train_tokens_per_second": 12142.109
    },
    {
      "epoch": 1.074602747104767,
      "grad_norm": 0.8411378775283455,
      "learning_rate": 4.415936710206952e-06,
      "loss": 0.30562567710876465,
      "num_input_tokens_seen": 508702216,
      "step": 15960,
      "train_runtime": 41894.9891,
      "train_tokens_per_second": 12142.316
    },
    {
      "epoch": 1.0749394021007272,
      "grad_norm": 0.7928299260507526,
      "learning_rate": 4.413310811312465e-06,
      "loss": 0.3055277824401855,
      "num_input_tokens_seen": 508860928,
      "step": 15965,
      "train_runtime": 41908.3745,
      "train_tokens_per_second": 12142.225
    },
    {
      "epoch": 1.0752760570966873,
      "grad_norm": 0.7318352729499958,
      "learning_rate": 4.41068507648374e-06,
      "loss": 0.2734399318695068,
      "num_input_tokens_seen": 509021128,
      "step": 15970,
      "train_runtime": 41921.1686,
      "train_tokens_per_second": 12142.341
    },
    {
      "epoch": 1.0756127120926475,
      "grad_norm": 0.7689755499687533,
      "learning_rate": 4.408059506455053e-06,
      "loss": 0.29433891773223875,
      "num_input_tokens_seen": 509184576,
      "step": 15975,
      "train_runtime": 41934.1652,
      "train_tokens_per_second": 12142.476
    },
    {
      "epoch": 1.0759493670886076,
      "grad_norm": 0.7597709482532323,
      "learning_rate": 4.40543410196064e-06,
      "loss": 0.31100993156433104,
      "num_input_tokens_seen": 509345904,
      "step": 15980,
      "train_runtime": 41947.0494,
      "train_tokens_per_second": 12142.592
    },
    {
      "epoch": 1.0762860220845678,
      "grad_norm": 0.8336391826154143,
      "learning_rate": 4.402808863734681e-06,
      "loss": 0.3164000272750854,
      "num_input_tokens_seen": 509507320,
      "step": 15985,
      "train_runtime": 41961.2752,
      "train_tokens_per_second": 12142.322
    },
    {
      "epoch": 1.0766226770805278,
      "grad_norm": 0.6956734429204328,
      "learning_rate": 4.40018379251132e-06,
      "loss": 0.30722713470458984,
      "num_input_tokens_seen": 509668672,
      "step": 15990,
      "train_runtime": 41974.1459,
      "train_tokens_per_second": 12142.443
    },
    {
      "epoch": 1.076959332076488,
      "grad_norm": 0.8787725042611539,
      "learning_rate": 4.397558889024649e-06,
      "loss": 0.3348447322845459,
      "num_input_tokens_seen": 509826056,
      "step": 15995,
      "train_runtime": 41987.2304,
      "train_tokens_per_second": 12142.407
    },
    {
      "epoch": 1.0772959870724481,
      "grad_norm": 0.7063604512091014,
      "learning_rate": 4.394934154008715e-06,
      "loss": 0.3141545295715332,
      "num_input_tokens_seen": 509983328,
      "step": 16000,
      "train_runtime": 41999.6493,
      "train_tokens_per_second": 12142.562
    },
    {
      "epoch": 1.0776326420684084,
      "grad_norm": 0.7582125772597686,
      "learning_rate": 4.392309588197514e-06,
      "loss": 0.3068138837814331,
      "num_input_tokens_seen": 510143712,
      "step": 16005,
      "train_runtime": 42012.8988,
      "train_tokens_per_second": 12142.55
    },
    {
      "epoch": 1.0779692970643684,
      "grad_norm": 0.7331137660688943,
      "learning_rate": 4.389685192324999e-06,
      "loss": 0.29303736686706544,
      "num_input_tokens_seen": 510302152,
      "step": 16010,
      "train_runtime": 42026.1676,
      "train_tokens_per_second": 12142.486
    },
    {
      "epoch": 1.0783059520603286,
      "grad_norm": 0.72484844650925,
      "learning_rate": 4.387060967125073e-06,
      "loss": 0.32088570594787597,
      "num_input_tokens_seen": 510455976,
      "step": 16015,
      "train_runtime": 42039.7395,
      "train_tokens_per_second": 12142.225
    },
    {
      "epoch": 1.0786426070562887,
      "grad_norm": 0.8753196315765692,
      "learning_rate": 4.384436913331596e-06,
      "loss": 0.33355059623718264,
      "num_input_tokens_seen": 510617360,
      "step": 16020,
      "train_runtime": 42052.5935,
      "train_tokens_per_second": 12142.351
    },
    {
      "epoch": 1.078979262052249,
      "grad_norm": 0.755121324850921,
      "learning_rate": 4.38181303167837e-06,
      "loss": 0.29300589561462403,
      "num_input_tokens_seen": 510779032,
      "step": 16025,
      "train_runtime": 42065.7711,
      "train_tokens_per_second": 12142.391
    },
    {
      "epoch": 1.079315917048209,
      "grad_norm": 0.8011959734887001,
      "learning_rate": 4.379189322899159e-06,
      "loss": 0.28904061317443847,
      "num_input_tokens_seen": 510932624,
      "step": 16030,
      "train_runtime": 42079.1941,
      "train_tokens_per_second": 12142.168
    },
    {
      "epoch": 1.0796525720441692,
      "grad_norm": 0.8032161697039722,
      "learning_rate": 4.376565787727676e-06,
      "loss": 0.3017615079879761,
      "num_input_tokens_seen": 511092704,
      "step": 16035,
      "train_runtime": 42092.6723,
      "train_tokens_per_second": 12142.083
    },
    {
      "epoch": 1.0799892270401292,
      "grad_norm": 0.7768452477087278,
      "learning_rate": 4.373942426897581e-06,
      "loss": 0.30951457023620604,
      "num_input_tokens_seen": 511247584,
      "step": 16040,
      "train_runtime": 42105.2347,
      "train_tokens_per_second": 12142.138
    },
    {
      "epoch": 1.0803258820360895,
      "grad_norm": 0.8152582723648151,
      "learning_rate": 4.37131924114249e-06,
      "loss": 0.3175765037536621,
      "num_input_tokens_seen": 511404032,
      "step": 16045,
      "train_runtime": 42118.3127,
      "train_tokens_per_second": 12142.083
    },
    {
      "epoch": 1.0806625370320495,
      "grad_norm": 0.7479414952160768,
      "learning_rate": 4.368696231195968e-06,
      "loss": 0.3115133047103882,
      "num_input_tokens_seen": 511565848,
      "step": 16050,
      "train_runtime": 42131.1813,
      "train_tokens_per_second": 12142.215
    },
    {
      "epoch": 1.0809991920280098,
      "grad_norm": 0.8664411279547134,
      "learning_rate": 4.366073397791534e-06,
      "loss": 0.289443826675415,
      "num_input_tokens_seen": 511725400,
      "step": 16055,
      "train_runtime": 42145.1265,
      "train_tokens_per_second": 12141.983
    },
    {
      "epoch": 1.0813358470239698,
      "grad_norm": 0.770396154441638,
      "learning_rate": 4.363450741662651e-06,
      "loss": 0.27687194347381594,
      "num_input_tokens_seen": 511886168,
      "step": 16060,
      "train_runtime": 42157.5451,
      "train_tokens_per_second": 12142.22
    },
    {
      "epoch": 1.08167250201993,
      "grad_norm": 0.8068595200622422,
      "learning_rate": 4.360828263542739e-06,
      "loss": 0.28565425872802735,
      "num_input_tokens_seen": 512045872,
      "step": 16065,
      "train_runtime": 42170.8663,
      "train_tokens_per_second": 12142.171
    },
    {
      "epoch": 1.08200915701589,
      "grad_norm": 0.8152651172299057,
      "learning_rate": 4.358205964165164e-06,
      "loss": 0.30882987976074217,
      "num_input_tokens_seen": 512207848,
      "step": 16070,
      "train_runtime": 42184.0081,
      "train_tokens_per_second": 12142.228
    },
    {
      "epoch": 1.0823458120118503,
      "grad_norm": 0.7961550000669346,
      "learning_rate": 4.355583844263247e-06,
      "loss": 0.3348849773406982,
      "num_input_tokens_seen": 512368568,
      "step": 16075,
      "train_runtime": 42196.9026,
      "train_tokens_per_second": 12142.326
    },
    {
      "epoch": 1.0826824670078103,
      "grad_norm": 0.8597139183818611,
      "learning_rate": 4.352961904570251e-06,
      "loss": 0.29843184947967527,
      "num_input_tokens_seen": 512523552,
      "step": 16080,
      "train_runtime": 42209.6572,
      "train_tokens_per_second": 12142.329
    },
    {
      "epoch": 1.0830191220037706,
      "grad_norm": 0.6999582326364607,
      "learning_rate": 4.350340145819397e-06,
      "loss": 0.3177281379699707,
      "num_input_tokens_seen": 512681984,
      "step": 16085,
      "train_runtime": 42223.134,
      "train_tokens_per_second": 12142.206
    },
    {
      "epoch": 1.0833557769997306,
      "grad_norm": 0.7630997784370741,
      "learning_rate": 4.347718568743847e-06,
      "loss": 0.3115131616592407,
      "num_input_tokens_seen": 512842496,
      "step": 16090,
      "train_runtime": 42235.5777,
      "train_tokens_per_second": 12142.429
    },
    {
      "epoch": 1.0836924319956909,
      "grad_norm": 0.796934092352418,
      "learning_rate": 4.34509717407672e-06,
      "loss": 0.32775604724884033,
      "num_input_tokens_seen": 513005936,
      "step": 16095,
      "train_runtime": 42248.5852,
      "train_tokens_per_second": 12142.559
    },
    {
      "epoch": 1.084029086991651,
      "grad_norm": 0.7837319483259637,
      "learning_rate": 4.342475962551081e-06,
      "loss": 0.3085785865783691,
      "num_input_tokens_seen": 513165168,
      "step": 16100,
      "train_runtime": 42261.6237,
      "train_tokens_per_second": 12142.58
    },
    {
      "epoch": 1.0843657419876112,
      "grad_norm": 0.747949236586764,
      "learning_rate": 4.339854934899943e-06,
      "loss": 0.2832313299179077,
      "num_input_tokens_seen": 513324136,
      "step": 16105,
      "train_runtime": 42275.5947,
      "train_tokens_per_second": 12142.328
    },
    {
      "epoch": 1.0847023969835712,
      "grad_norm": 0.6801092182516424,
      "learning_rate": 4.337234091856265e-06,
      "loss": 0.2824375867843628,
      "num_input_tokens_seen": 513485968,
      "step": 16110,
      "train_runtime": 42288.1265,
      "train_tokens_per_second": 12142.557
    },
    {
      "epoch": 1.0850390519795314,
      "grad_norm": 0.7876340660063319,
      "learning_rate": 4.334613434152961e-06,
      "loss": 0.2991558313369751,
      "num_input_tokens_seen": 513639936,
      "step": 16115,
      "train_runtime": 42300.5636,
      "train_tokens_per_second": 12142.626
    },
    {
      "epoch": 1.0853757069754915,
      "grad_norm": 0.858270331322216,
      "learning_rate": 4.331992962522888e-06,
      "loss": 0.33192427158355714,
      "num_input_tokens_seen": 513801240,
      "step": 16120,
      "train_runtime": 42313.2839,
      "train_tokens_per_second": 12142.788
    },
    {
      "epoch": 1.0857123619714517,
      "grad_norm": 0.7623271972473266,
      "learning_rate": 4.329372677698854e-06,
      "loss": 0.31349477767944334,
      "num_input_tokens_seen": 513965080,
      "step": 16125,
      "train_runtime": 42325.7019,
      "train_tokens_per_second": 12143.096
    },
    {
      "epoch": 1.0860490169674117,
      "grad_norm": 0.8325650038048475,
      "learning_rate": 4.3267525804136114e-06,
      "loss": 0.3035757064819336,
      "num_input_tokens_seen": 514121144,
      "step": 16130,
      "train_runtime": 42338.4827,
      "train_tokens_per_second": 12143.117
    },
    {
      "epoch": 1.086385671963372,
      "grad_norm": 0.8215091915349555,
      "learning_rate": 4.324132671399864e-06,
      "loss": 0.31330142021179197,
      "num_input_tokens_seen": 514278736,
      "step": 16135,
      "train_runtime": 42352.2375,
      "train_tokens_per_second": 12142.894
    },
    {
      "epoch": 1.086722326959332,
      "grad_norm": 0.9517379181212042,
      "learning_rate": 4.321512951390258e-06,
      "loss": 0.31854069232940674,
      "num_input_tokens_seen": 514438768,
      "step": 16140,
      "train_runtime": 42365.4203,
      "train_tokens_per_second": 12142.893
    },
    {
      "epoch": 1.0870589819552923,
      "grad_norm": 0.7432116493268239,
      "learning_rate": 4.318893421117393e-06,
      "loss": 0.2987842082977295,
      "num_input_tokens_seen": 514597536,
      "step": 16145,
      "train_runtime": 42378.1645,
      "train_tokens_per_second": 12142.988
    },
    {
      "epoch": 1.0873956369512523,
      "grad_norm": 0.7099860604888883,
      "learning_rate": 4.31627408131381e-06,
      "loss": 0.29923858642578127,
      "num_input_tokens_seen": 514757216,
      "step": 16150,
      "train_runtime": 42391.5934,
      "train_tokens_per_second": 12142.908
    },
    {
      "epoch": 1.0877322919472125,
      "grad_norm": 0.7513437885259707,
      "learning_rate": 4.313654932712001e-06,
      "loss": 0.28290209770202634,
      "num_input_tokens_seen": 514917192,
      "step": 16155,
      "train_runtime": 42404.8049,
      "train_tokens_per_second": 12142.897
    },
    {
      "epoch": 1.0880689469431726,
      "grad_norm": 0.6917614094403428,
      "learning_rate": 4.3110359760444e-06,
      "loss": 0.30543758869171145,
      "num_input_tokens_seen": 515080744,
      "step": 16160,
      "train_runtime": 42417.837,
      "train_tokens_per_second": 12143.022
    },
    {
      "epoch": 1.0884056019391328,
      "grad_norm": 0.7059364529527964,
      "learning_rate": 4.308417212043392e-06,
      "loss": 0.2869384527206421,
      "num_input_tokens_seen": 515237120,
      "step": 16165,
      "train_runtime": 42430.6585,
      "train_tokens_per_second": 12143.038
    },
    {
      "epoch": 1.0887422569350929,
      "grad_norm": 0.7587259656975229,
      "learning_rate": 4.305798641441304e-06,
      "loss": 0.28521573543548584,
      "num_input_tokens_seen": 515397672,
      "step": 16170,
      "train_runtime": 42443.4915,
      "train_tokens_per_second": 12143.15
    },
    {
      "epoch": 1.089078911931053,
      "grad_norm": 0.7671442724109391,
      "learning_rate": 4.303180264970416e-06,
      "loss": 0.3034674644470215,
      "num_input_tokens_seen": 515557768,
      "step": 16175,
      "train_runtime": 42457.2775,
      "train_tokens_per_second": 12142.978
    },
    {
      "epoch": 1.0894155669270131,
      "grad_norm": 0.7247684279515608,
      "learning_rate": 4.3005620833629415e-06,
      "loss": 0.2748046875,
      "num_input_tokens_seen": 515717744,
      "step": 16180,
      "train_runtime": 42470.5221,
      "train_tokens_per_second": 12142.957
    },
    {
      "epoch": 1.0897522219229734,
      "grad_norm": 0.8195842749574394,
      "learning_rate": 4.297944097351053e-06,
      "loss": 0.3098749160766602,
      "num_input_tokens_seen": 515873776,
      "step": 16185,
      "train_runtime": 42483.2906,
      "train_tokens_per_second": 12142.981
    },
    {
      "epoch": 1.0900888769189334,
      "grad_norm": 0.7156167363580491,
      "learning_rate": 4.295326307666856e-06,
      "loss": 0.3206508159637451,
      "num_input_tokens_seen": 516037224,
      "step": 16190,
      "train_runtime": 42496.2986,
      "train_tokens_per_second": 12143.11
    },
    {
      "epoch": 1.0904255319148937,
      "grad_norm": 0.8095871146954722,
      "learning_rate": 4.292708715042411e-06,
      "loss": 0.29314846992492677,
      "num_input_tokens_seen": 516193800,
      "step": 16195,
      "train_runtime": 42508.8235,
      "train_tokens_per_second": 12143.215
    },
    {
      "epoch": 1.0907621869108537,
      "grad_norm": 0.8612207401482795,
      "learning_rate": 4.290091320209718e-06,
      "loss": 0.31816318035125735,
      "num_input_tokens_seen": 516356952,
      "step": 16200,
      "train_runtime": 42521.6828,
      "train_tokens_per_second": 12143.38
    },
    {
      "epoch": 1.091098841906814,
      "grad_norm": 0.7573667487632446,
      "learning_rate": 4.287474123900726e-06,
      "loss": 0.310860800743103,
      "num_input_tokens_seen": 516512480,
      "step": 16205,
      "train_runtime": 42534.8384,
      "train_tokens_per_second": 12143.281
    },
    {
      "epoch": 1.091435496902774,
      "grad_norm": 0.8128234652222646,
      "learning_rate": 4.284857126847321e-06,
      "loss": 0.3073211669921875,
      "num_input_tokens_seen": 516666920,
      "step": 16210,
      "train_runtime": 42547.4742,
      "train_tokens_per_second": 12143.304
    },
    {
      "epoch": 1.0917721518987342,
      "grad_norm": 0.7519137562268962,
      "learning_rate": 4.28224032978134e-06,
      "loss": 0.3210838794708252,
      "num_input_tokens_seen": 516830112,
      "step": 16215,
      "train_runtime": 42560.351,
      "train_tokens_per_second": 12143.465
    },
    {
      "epoch": 1.0921088068946943,
      "grad_norm": 0.7849906032769867,
      "learning_rate": 4.279623733434561e-06,
      "loss": 0.3127107620239258,
      "num_input_tokens_seen": 516991288,
      "step": 16220,
      "train_runtime": 42573.8003,
      "train_tokens_per_second": 12143.414
    },
    {
      "epoch": 1.0924454618906545,
      "grad_norm": 0.7941015624118866,
      "learning_rate": 4.277007338538711e-06,
      "loss": 0.27778594493865966,
      "num_input_tokens_seen": 517154008,
      "step": 16225,
      "train_runtime": 42586.5184,
      "train_tokens_per_second": 12143.609
    },
    {
      "epoch": 1.0927821168866145,
      "grad_norm": 0.6385759084636674,
      "learning_rate": 4.27439114582545e-06,
      "loss": 0.27554757595062257,
      "num_input_tokens_seen": 517316352,
      "step": 16230,
      "train_runtime": 42600.0405,
      "train_tokens_per_second": 12143.565
    },
    {
      "epoch": 1.0931187718825748,
      "grad_norm": 0.869868839253338,
      "learning_rate": 4.271775156026395e-06,
      "loss": 0.32135074138641356,
      "num_input_tokens_seen": 517473040,
      "step": 16235,
      "train_runtime": 42613.4852,
      "train_tokens_per_second": 12143.41
    },
    {
      "epoch": 1.0934554268785348,
      "grad_norm": 0.7097794953608657,
      "learning_rate": 4.269159369873093e-06,
      "loss": 0.3008142948150635,
      "num_input_tokens_seen": 517625936,
      "step": 16240,
      "train_runtime": 42626.7102,
      "train_tokens_per_second": 12143.23
    },
    {
      "epoch": 1.093792081874495,
      "grad_norm": 1.0745335015887238,
      "learning_rate": 4.266543788097043e-06,
      "loss": 0.29512729644775393,
      "num_input_tokens_seen": 517789568,
      "step": 16245,
      "train_runtime": 42640.4063,
      "train_tokens_per_second": 12143.167
    },
    {
      "epoch": 1.094128736870455,
      "grad_norm": 0.8386242349730625,
      "learning_rate": 4.263928411429683e-06,
      "loss": 0.3487131595611572,
      "num_input_tokens_seen": 517945176,
      "step": 16250,
      "train_runtime": 42653.0038,
      "train_tokens_per_second": 12143.229
    },
    {
      "epoch": 1.0944653918664153,
      "grad_norm": 0.8708778387863215,
      "learning_rate": 4.261313240602398e-06,
      "loss": 0.31237106323242186,
      "num_input_tokens_seen": 518102840,
      "step": 16255,
      "train_runtime": 42666.6208,
      "train_tokens_per_second": 12143.048
    },
    {
      "epoch": 1.0948020468623754,
      "grad_norm": 0.8713071305026213,
      "learning_rate": 4.2586982763465064e-06,
      "loss": 0.29385929107666015,
      "num_input_tokens_seen": 518264520,
      "step": 16260,
      "train_runtime": 42679.1445,
      "train_tokens_per_second": 12143.273
    },
    {
      "epoch": 1.0951387018583356,
      "grad_norm": 0.8034382762500902,
      "learning_rate": 4.256083519393281e-06,
      "loss": 0.3068211078643799,
      "num_input_tokens_seen": 518426200,
      "step": 16265,
      "train_runtime": 42692.2797,
      "train_tokens_per_second": 12143.324
    },
    {
      "epoch": 1.0954753568542956,
      "grad_norm": 0.7950086549599764,
      "learning_rate": 4.253468970473925e-06,
      "loss": 0.3150953769683838,
      "num_input_tokens_seen": 518586184,
      "step": 16270,
      "train_runtime": 42705.1534,
      "train_tokens_per_second": 12143.41
    },
    {
      "epoch": 1.095812011850256,
      "grad_norm": 0.8054529714464732,
      "learning_rate": 4.250854630319593e-06,
      "loss": 0.3317426204681396,
      "num_input_tokens_seen": 518742464,
      "step": 16275,
      "train_runtime": 42717.8194,
      "train_tokens_per_second": 12143.468
    },
    {
      "epoch": 1.096148666846216,
      "grad_norm": 0.756324090329135,
      "learning_rate": 4.248240499661373e-06,
      "loss": 0.3024318218231201,
      "num_input_tokens_seen": 518896976,
      "step": 16280,
      "train_runtime": 42731.5139,
      "train_tokens_per_second": 12143.192
    },
    {
      "epoch": 1.0964853218421762,
      "grad_norm": 0.8768147154005344,
      "learning_rate": 4.245626579230302e-06,
      "loss": 0.27411704063415526,
      "num_input_tokens_seen": 519056016,
      "step": 16285,
      "train_runtime": 42744.6587,
      "train_tokens_per_second": 12143.178
    },
    {
      "epoch": 1.0968219768381362,
      "grad_norm": 0.9908850979949622,
      "learning_rate": 4.243012869757351e-06,
      "loss": 0.30057971477508544,
      "num_input_tokens_seen": 519219856,
      "step": 16290,
      "train_runtime": 42757.076,
      "train_tokens_per_second": 12143.484
    },
    {
      "epoch": 1.0971586318340965,
      "grad_norm": 0.7797781971021313,
      "learning_rate": 4.240399371973438e-06,
      "loss": 0.29999120235443116,
      "num_input_tokens_seen": 519382704,
      "step": 16295,
      "train_runtime": 42769.8186,
      "train_tokens_per_second": 12143.673
    },
    {
      "epoch": 1.0974952868300565,
      "grad_norm": 0.859772325255097,
      "learning_rate": 4.2377860866094165e-06,
      "loss": 0.32585830688476564,
      "num_input_tokens_seen": 519543296,
      "step": 16300,
      "train_runtime": 42782.8828,
      "train_tokens_per_second": 12143.719
    },
    {
      "epoch": 1.0978319418260167,
      "grad_norm": 0.8872067994313502,
      "learning_rate": 4.235173014396088e-06,
      "loss": 0.3113248348236084,
      "num_input_tokens_seen": 519702328,
      "step": 16305,
      "train_runtime": 42796.179,
      "train_tokens_per_second": 12143.662
    },
    {
      "epoch": 1.0981685968219768,
      "grad_norm": 0.718673312506372,
      "learning_rate": 4.2325601560641854e-06,
      "loss": 0.3198572874069214,
      "num_input_tokens_seen": 519865744,
      "step": 16310,
      "train_runtime": 42809.1587,
      "train_tokens_per_second": 12143.797
    },
    {
      "epoch": 1.098505251817937,
      "grad_norm": 0.7161498524569786,
      "learning_rate": 4.229947512344389e-06,
      "loss": 0.2932559013366699,
      "num_input_tokens_seen": 520024976,
      "step": 16315,
      "train_runtime": 42822.5452,
      "train_tokens_per_second": 12143.719
    },
    {
      "epoch": 1.098841906813897,
      "grad_norm": 0.8005662693779073,
      "learning_rate": 4.227335083967314e-06,
      "loss": 0.31361708641052244,
      "num_input_tokens_seen": 520181872,
      "step": 16320,
      "train_runtime": 42835.4124,
      "train_tokens_per_second": 12143.734
    },
    {
      "epoch": 1.0991785618098573,
      "grad_norm": 0.7712369108361606,
      "learning_rate": 4.224722871663522e-06,
      "loss": 0.28961164951324464,
      "num_input_tokens_seen": 520339752,
      "step": 16325,
      "train_runtime": 42848.8611,
      "train_tokens_per_second": 12143.608
    },
    {
      "epoch": 1.0995152168058173,
      "grad_norm": 0.7155936063193538,
      "learning_rate": 4.222110876163503e-06,
      "loss": 0.33086836338043213,
      "num_input_tokens_seen": 520502560,
      "step": 16330,
      "train_runtime": 42861.3023,
      "train_tokens_per_second": 12143.881
    },
    {
      "epoch": 1.0998518718017776,
      "grad_norm": 0.8198435314464898,
      "learning_rate": 4.2194990981977e-06,
      "loss": 0.29511992931365966,
      "num_input_tokens_seen": 520662472,
      "step": 16335,
      "train_runtime": 42873.7238,
      "train_tokens_per_second": 12144.093
    },
    {
      "epoch": 1.1001885267977376,
      "grad_norm": 0.8116702693724875,
      "learning_rate": 4.216887538496482e-06,
      "loss": 0.2890969753265381,
      "num_input_tokens_seen": 520821680,
      "step": 16340,
      "train_runtime": 42887.0836,
      "train_tokens_per_second": 12144.022
    },
    {
      "epoch": 1.1005251817936978,
      "grad_norm": 0.8439012904713973,
      "learning_rate": 4.2142761977901695e-06,
      "loss": 0.3554262161254883,
      "num_input_tokens_seen": 520981152,
      "step": 16345,
      "train_runtime": 42900.6133,
      "train_tokens_per_second": 12143.909
    },
    {
      "epoch": 1.1008618367896579,
      "grad_norm": 0.8427584736798379,
      "learning_rate": 4.21166507680901e-06,
      "loss": 0.32249984741210935,
      "num_input_tokens_seen": 521135152,
      "step": 16350,
      "train_runtime": 42913.7727,
      "train_tokens_per_second": 12143.774
    },
    {
      "epoch": 1.1011984917856181,
      "grad_norm": 0.8113832346360442,
      "learning_rate": 4.209054176283202e-06,
      "loss": 0.29596123695373533,
      "num_input_tokens_seen": 521294072,
      "step": 16355,
      "train_runtime": 42926.9982,
      "train_tokens_per_second": 12143.735
    },
    {
      "epoch": 1.1015351467815782,
      "grad_norm": 0.818554904791197,
      "learning_rate": 4.206443496942868e-06,
      "loss": 0.3072291374206543,
      "num_input_tokens_seen": 521450224,
      "step": 16360,
      "train_runtime": 42939.7323,
      "train_tokens_per_second": 12143.77
    },
    {
      "epoch": 1.1018718017775384,
      "grad_norm": 0.8274487066904288,
      "learning_rate": 4.203833039518081e-06,
      "loss": 0.2813434600830078,
      "num_input_tokens_seen": 521607248,
      "step": 16365,
      "train_runtime": 42952.2583,
      "train_tokens_per_second": 12143.884
    },
    {
      "epoch": 1.1022084567734984,
      "grad_norm": 0.8316527632743279,
      "learning_rate": 4.201222804738845e-06,
      "loss": 0.3044417381286621,
      "num_input_tokens_seen": 521769136,
      "step": 16370,
      "train_runtime": 42965.3723,
      "train_tokens_per_second": 12143.945
    },
    {
      "epoch": 1.1025451117694587,
      "grad_norm": 0.8340993615080122,
      "learning_rate": 4.198612793335107e-06,
      "loss": 0.29143385887145995,
      "num_input_tokens_seen": 521929304,
      "step": 16375,
      "train_runtime": 42978.425,
      "train_tokens_per_second": 12143.984
    },
    {
      "epoch": 1.1028817667654187,
      "grad_norm": 0.718624256411508,
      "learning_rate": 4.196003006036743e-06,
      "loss": 0.3010996341705322,
      "num_input_tokens_seen": 522087632,
      "step": 16380,
      "train_runtime": 42991.6727,
      "train_tokens_per_second": 12143.925
    },
    {
      "epoch": 1.103218421761379,
      "grad_norm": 0.8493770822721732,
      "learning_rate": 4.193393443573576e-06,
      "loss": 0.3024613857269287,
      "num_input_tokens_seen": 522247224,
      "step": 16385,
      "train_runtime": 43004.6825,
      "train_tokens_per_second": 12143.962
    },
    {
      "epoch": 1.103555076757339,
      "grad_norm": 0.7905228265671207,
      "learning_rate": 4.19078410667536e-06,
      "loss": 0.32945923805236815,
      "num_input_tokens_seen": 522405720,
      "step": 16390,
      "train_runtime": 43017.601,
      "train_tokens_per_second": 12143.999
    },
    {
      "epoch": 1.1038917317532992,
      "grad_norm": 0.7281244545542417,
      "learning_rate": 4.188174996071788e-06,
      "loss": 0.318757438659668,
      "num_input_tokens_seen": 522553984,
      "step": 16395,
      "train_runtime": 43031.4871,
      "train_tokens_per_second": 12143.526
    },
    {
      "epoch": 1.1042283867492593,
      "grad_norm": 0.6973509107367324,
      "learning_rate": 4.18556611249249e-06,
      "loss": 0.2981884956359863,
      "num_input_tokens_seen": 522715456,
      "step": 16400,
      "train_runtime": 43045.0699,
      "train_tokens_per_second": 12143.445
    },
    {
      "epoch": 1.1045650417452195,
      "grad_norm": 0.9465322310336238,
      "learning_rate": 4.182957456667032e-06,
      "loss": 0.3188304901123047,
      "num_input_tokens_seen": 522877272,
      "step": 16405,
      "train_runtime": 43057.4845,
      "train_tokens_per_second": 12143.702
    },
    {
      "epoch": 1.1049016967411796,
      "grad_norm": 0.7581757107786998,
      "learning_rate": 4.1803490293249145e-06,
      "loss": 0.33195176124572756,
      "num_input_tokens_seen": 523035664,
      "step": 16410,
      "train_runtime": 43070.0534,
      "train_tokens_per_second": 12143.836
    },
    {
      "epoch": 1.1052383517371398,
      "grad_norm": 0.904014876714174,
      "learning_rate": 4.1777408311955785e-06,
      "loss": 0.30089755058288575,
      "num_input_tokens_seen": 523190776,
      "step": 16415,
      "train_runtime": 43083.5498,
      "train_tokens_per_second": 12143.632
    },
    {
      "epoch": 1.1055750067330998,
      "grad_norm": 0.837501695323459,
      "learning_rate": 4.175132863008395e-06,
      "loss": 0.34442529678344724,
      "num_input_tokens_seen": 523350912,
      "step": 16420,
      "train_runtime": 43096.4372,
      "train_tokens_per_second": 12143.716
    },
    {
      "epoch": 1.10591166172906,
      "grad_norm": 0.8964836524563399,
      "learning_rate": 4.17252512549268e-06,
      "loss": 0.3033717632293701,
      "num_input_tokens_seen": 523498544,
      "step": 16425,
      "train_runtime": 43109.3181,
      "train_tokens_per_second": 12143.513
    },
    {
      "epoch": 1.10624831672502,
      "grad_norm": 0.8016503796216948,
      "learning_rate": 4.169917619377671e-06,
      "loss": 0.3006324768066406,
      "num_input_tokens_seen": 523660816,
      "step": 16430,
      "train_runtime": 43123.4433,
      "train_tokens_per_second": 12143.298
    },
    {
      "epoch": 1.1065849717209804,
      "grad_norm": 0.9109348937175265,
      "learning_rate": 4.167310345392557e-06,
      "loss": 0.34682116508483884,
      "num_input_tokens_seen": 523815320,
      "step": 16435,
      "train_runtime": 43136.5392,
      "train_tokens_per_second": 12143.193
    },
    {
      "epoch": 1.1069216267169404,
      "grad_norm": 0.7323176495317748,
      "learning_rate": 4.1647033042664455e-06,
      "loss": 0.3289669990539551,
      "num_input_tokens_seen": 523972552,
      "step": 16440,
      "train_runtime": 43149.5086,
      "train_tokens_per_second": 12143.187
    },
    {
      "epoch": 1.1072582817129006,
      "grad_norm": 0.7290678296039488,
      "learning_rate": 4.162096496728394e-06,
      "loss": 0.29602556228637694,
      "num_input_tokens_seen": 524131424,
      "step": 16445,
      "train_runtime": 43162.1037,
      "train_tokens_per_second": 12143.324
    },
    {
      "epoch": 1.1075949367088607,
      "grad_norm": 0.7978316542718361,
      "learning_rate": 4.1594899235073835e-06,
      "loss": 0.30616140365600586,
      "num_input_tokens_seen": 524295264,
      "step": 16450,
      "train_runtime": 43174.536,
      "train_tokens_per_second": 12143.622
    },
    {
      "epoch": 1.107931591704821,
      "grad_norm": 0.781256768571631,
      "learning_rate": 4.156883585332338e-06,
      "loss": 0.334748649597168,
      "num_input_tokens_seen": 524457800,
      "step": 16455,
      "train_runtime": 43187.209,
      "train_tokens_per_second": 12143.823
    },
    {
      "epoch": 1.108268246700781,
      "grad_norm": 0.7021335378116609,
      "learning_rate": 4.154277482932107e-06,
      "loss": 0.32042970657348635,
      "num_input_tokens_seen": 524612320,
      "step": 16460,
      "train_runtime": 43200.701,
      "train_tokens_per_second": 12143.607
    },
    {
      "epoch": 1.1086049016967412,
      "grad_norm": 0.7534201172928595,
      "learning_rate": 4.151671617035482e-06,
      "loss": 0.3039975881576538,
      "num_input_tokens_seen": 524767480,
      "step": 16465,
      "train_runtime": 43214.0198,
      "train_tokens_per_second": 12143.454
    },
    {
      "epoch": 1.1089415566927012,
      "grad_norm": 0.7479141175427881,
      "learning_rate": 4.149065988371184e-06,
      "loss": 0.2894607067108154,
      "num_input_tokens_seen": 524926032,
      "step": 16470,
      "train_runtime": 43227.2765,
      "train_tokens_per_second": 12143.398
    },
    {
      "epoch": 1.1092782116886615,
      "grad_norm": 0.8108807130173762,
      "learning_rate": 4.14646059766787e-06,
      "loss": 0.30585517883300783,
      "num_input_tokens_seen": 525082032,
      "step": 16475,
      "train_runtime": 43240.5102,
      "train_tokens_per_second": 12143.289
    },
    {
      "epoch": 1.1096148666846215,
      "grad_norm": 0.7130893484730619,
      "learning_rate": 4.143855445654127e-06,
      "loss": 0.29894223213195803,
      "num_input_tokens_seen": 525245440,
      "step": 16480,
      "train_runtime": 43253.514,
      "train_tokens_per_second": 12143.417
    },
    {
      "epoch": 1.1099515216805818,
      "grad_norm": 0.79610241847635,
      "learning_rate": 4.14125053305848e-06,
      "loss": 0.28519444465637206,
      "num_input_tokens_seen": 525398920,
      "step": 16485,
      "train_runtime": 43266.79,
      "train_tokens_per_second": 12143.238
    },
    {
      "epoch": 1.1102881766765418,
      "grad_norm": 0.7824865959683505,
      "learning_rate": 4.1386458606093805e-06,
      "loss": 0.29976956844329833,
      "num_input_tokens_seen": 525556680,
      "step": 16490,
      "train_runtime": 43279.5548,
      "train_tokens_per_second": 12143.301
    },
    {
      "epoch": 1.110624831672502,
      "grad_norm": 0.8259920341179898,
      "learning_rate": 4.1360414290352214e-06,
      "loss": 0.3209113121032715,
      "num_input_tokens_seen": 525719296,
      "step": 16495,
      "train_runtime": 43292.9762,
      "train_tokens_per_second": 12143.293
    },
    {
      "epoch": 1.110961486668462,
      "grad_norm": 0.859160619417763,
      "learning_rate": 4.1334372390643205e-06,
      "loss": 0.2984657287597656,
      "num_input_tokens_seen": 525883136,
      "step": 16500,
      "train_runtime": 43305.4094,
      "train_tokens_per_second": 12143.59
    },
    {
      "epoch": 1.1112981416644223,
      "grad_norm": 0.6938627820686201,
      "learning_rate": 4.130833291424934e-06,
      "loss": 0.31500074863433836,
      "num_input_tokens_seen": 526042736,
      "step": 16505,
      "train_runtime": 43317.8314,
      "train_tokens_per_second": 12143.792
    },
    {
      "epoch": 1.1116347966603823,
      "grad_norm": 0.835906323663683,
      "learning_rate": 4.128229586845244e-06,
      "loss": 0.28493990898132326,
      "num_input_tokens_seen": 526203664,
      "step": 16510,
      "train_runtime": 43331.7279,
      "train_tokens_per_second": 12143.611
    },
    {
      "epoch": 1.1119714516563426,
      "grad_norm": 0.6310815118440094,
      "learning_rate": 4.125626126053372e-06,
      "loss": 0.2798593521118164,
      "num_input_tokens_seen": 526362624,
      "step": 16515,
      "train_runtime": 43344.7326,
      "train_tokens_per_second": 12143.635
    },
    {
      "epoch": 1.1123081066523026,
      "grad_norm": 0.9261497840544736,
      "learning_rate": 4.123022909777364e-06,
      "loss": 0.3052553176879883,
      "num_input_tokens_seen": 526521336,
      "step": 16520,
      "train_runtime": 43357.8937,
      "train_tokens_per_second": 12143.61
    },
    {
      "epoch": 1.1126447616482629,
      "grad_norm": 0.7816107238166144,
      "learning_rate": 4.120419938745205e-06,
      "loss": 0.3066908359527588,
      "num_input_tokens_seen": 526684096,
      "step": 16525,
      "train_runtime": 43370.3495,
      "train_tokens_per_second": 12143.875
    },
    {
      "epoch": 1.112981416644223,
      "grad_norm": 0.8504062852433165,
      "learning_rate": 4.1178172136848035e-06,
      "loss": 0.28658390045166016,
      "num_input_tokens_seen": 526838464,
      "step": 16530,
      "train_runtime": 43383.4231,
      "train_tokens_per_second": 12143.774
    },
    {
      "epoch": 1.1133180716401831,
      "grad_norm": 0.8675535768661058,
      "learning_rate": 4.115214735324007e-06,
      "loss": 0.2815955638885498,
      "num_input_tokens_seen": 527000888,
      "step": 16535,
      "train_runtime": 43396.074,
      "train_tokens_per_second": 12143.976
    },
    {
      "epoch": 1.1136547266361432,
      "grad_norm": 0.6944688251897749,
      "learning_rate": 4.112612504390586e-06,
      "loss": 0.27120413780212405,
      "num_input_tokens_seen": 527162456,
      "step": 16540,
      "train_runtime": 43409.5836,
      "train_tokens_per_second": 12143.919
    },
    {
      "epoch": 1.1139913816321034,
      "grad_norm": 0.805256165583911,
      "learning_rate": 4.11001052161225e-06,
      "loss": 0.3365485668182373,
      "num_input_tokens_seen": 527324944,
      "step": 16545,
      "train_runtime": 43422.2685,
      "train_tokens_per_second": 12144.113
    },
    {
      "epoch": 1.1143280366280637,
      "grad_norm": 0.8204477946750075,
      "learning_rate": 4.107408787716632e-06,
      "loss": 0.31297597885131834,
      "num_input_tokens_seen": 527486544,
      "step": 16550,
      "train_runtime": 43435.3376,
      "train_tokens_per_second": 12144.18
    },
    {
      "epoch": 1.1146646916240237,
      "grad_norm": 0.7456119447474984,
      "learning_rate": 4.104807303431303e-06,
      "loss": 0.306234073638916,
      "num_input_tokens_seen": 527647064,
      "step": 16555,
      "train_runtime": 43448.8515,
      "train_tokens_per_second": 12144.097
    },
    {
      "epoch": 1.1150013466199837,
      "grad_norm": 0.9037303363720901,
      "learning_rate": 4.102206069483754e-06,
      "loss": 0.32191832065582277,
      "num_input_tokens_seen": 527805888,
      "step": 16560,
      "train_runtime": 43462.3067,
      "train_tokens_per_second": 12143.992
    },
    {
      "epoch": 1.115338001615944,
      "grad_norm": 0.757956349276041,
      "learning_rate": 4.099605086601417e-06,
      "loss": 0.27638850212097166,
      "num_input_tokens_seen": 527959464,
      "step": 16565,
      "train_runtime": 43475.5428,
      "train_tokens_per_second": 12143.827
    },
    {
      "epoch": 1.1156746566119042,
      "grad_norm": 0.8258892845370119,
      "learning_rate": 4.097004355511645e-06,
      "loss": 0.29239535331726074,
      "num_input_tokens_seen": 528120168,
      "step": 16570,
      "train_runtime": 43488.3866,
      "train_tokens_per_second": 12143.936
    },
    {
      "epoch": 1.1160113116078643,
      "grad_norm": 0.8407072578587406,
      "learning_rate": 4.094403876941728e-06,
      "loss": 0.2621783971786499,
      "num_input_tokens_seen": 528275312,
      "step": 16575,
      "train_runtime": 43500.8833,
      "train_tokens_per_second": 12144.013
    },
    {
      "epoch": 1.1163479666038243,
      "grad_norm": 0.658608673412688,
      "learning_rate": 4.0918036516188755e-06,
      "loss": 0.30341520309448244,
      "num_input_tokens_seen": 528432456,
      "step": 16580,
      "train_runtime": 43514.5116,
      "train_tokens_per_second": 12143.821
    },
    {
      "epoch": 1.1166846215997845,
      "grad_norm": 0.8418078009048707,
      "learning_rate": 4.0892036802702395e-06,
      "loss": 0.30974364280700684,
      "num_input_tokens_seen": 528595328,
      "step": 16585,
      "train_runtime": 43528.0076,
      "train_tokens_per_second": 12143.798
    },
    {
      "epoch": 1.1170212765957448,
      "grad_norm": 0.7506779725085156,
      "learning_rate": 4.086603963622887e-06,
      "loss": 0.28029513359069824,
      "num_input_tokens_seen": 528758272,
      "step": 16590,
      "train_runtime": 43541.365,
      "train_tokens_per_second": 12143.815
    },
    {
      "epoch": 1.1173579315917048,
      "grad_norm": 0.7899748690850563,
      "learning_rate": 4.084004502403825e-06,
      "loss": 0.3236206293106079,
      "num_input_tokens_seen": 528915520,
      "step": 16595,
      "train_runtime": 43553.83,
      "train_tokens_per_second": 12143.95
    },
    {
      "epoch": 1.1176945865876649,
      "grad_norm": 0.8160660957938676,
      "learning_rate": 4.081405297339982e-06,
      "loss": 0.29514460563659667,
      "num_input_tokens_seen": 529077376,
      "step": 16600,
      "train_runtime": 43566.3436,
      "train_tokens_per_second": 12144.177
    },
    {
      "epoch": 1.118031241583625,
      "grad_norm": 0.8004454304729841,
      "learning_rate": 4.0788063491582205e-06,
      "loss": 0.30223424434661866,
      "num_input_tokens_seen": 529234712,
      "step": 16605,
      "train_runtime": 43579.8,
      "train_tokens_per_second": 12144.037
    },
    {
      "epoch": 1.1183678965795854,
      "grad_norm": 0.9197485130715908,
      "learning_rate": 4.076207658585323e-06,
      "loss": 0.2935061454772949,
      "num_input_tokens_seen": 529395640,
      "step": 16610,
      "train_runtime": 43592.2692,
      "train_tokens_per_second": 12144.255
    },
    {
      "epoch": 1.1187045515755454,
      "grad_norm": 0.7625676317862476,
      "learning_rate": 4.07360922634801e-06,
      "loss": 0.290375804901123,
      "num_input_tokens_seen": 529556856,
      "step": 16615,
      "train_runtime": 43604.7781,
      "train_tokens_per_second": 12144.469
    },
    {
      "epoch": 1.1190412065715054,
      "grad_norm": 0.8281950940102131,
      "learning_rate": 4.071011053172922e-06,
      "loss": 0.30091009140014646,
      "num_input_tokens_seen": 529713856,
      "step": 16620,
      "train_runtime": 43618.8239,
      "train_tokens_per_second": 12144.157
    },
    {
      "epoch": 1.1193778615674657,
      "grad_norm": 0.7652072948984868,
      "learning_rate": 4.068413139786631e-06,
      "loss": 0.29552178382873534,
      "num_input_tokens_seen": 529873080,
      "step": 16625,
      "train_runtime": 43632.1521,
      "train_tokens_per_second": 12144.097
    },
    {
      "epoch": 1.119714516563426,
      "grad_norm": 0.8067537723081993,
      "learning_rate": 4.065815486915635e-06,
      "loss": 0.29994003772735595,
      "num_input_tokens_seen": 530036304,
      "step": 16630,
      "train_runtime": 43645.0514,
      "train_tokens_per_second": 12144.247
    },
    {
      "epoch": 1.120051171559386,
      "grad_norm": 0.7635841509137418,
      "learning_rate": 4.063218095286359e-06,
      "loss": 0.3320046901702881,
      "num_input_tokens_seen": 530199312,
      "step": 16635,
      "train_runtime": 43658.4209,
      "train_tokens_per_second": 12144.262
    },
    {
      "epoch": 1.120387826555346,
      "grad_norm": 0.8083874794780219,
      "learning_rate": 4.0606209656251555e-06,
      "loss": 0.28829739093780515,
      "num_input_tokens_seen": 530360624,
      "step": 16640,
      "train_runtime": 43671.5548,
      "train_tokens_per_second": 12144.304
    },
    {
      "epoch": 1.1207244815513062,
      "grad_norm": 0.7545106323823049,
      "learning_rate": 4.058024098658304e-06,
      "loss": 0.3185067415237427,
      "num_input_tokens_seen": 530522584,
      "step": 16645,
      "train_runtime": 43684.555,
      "train_tokens_per_second": 12144.397
    },
    {
      "epoch": 1.1210611365472665,
      "grad_norm": 0.7614003109930435,
      "learning_rate": 4.055427495112008e-06,
      "loss": 0.33021135330200196,
      "num_input_tokens_seen": 530683832,
      "step": 16650,
      "train_runtime": 43698.2113,
      "train_tokens_per_second": 12144.292
    },
    {
      "epoch": 1.1213977915432265,
      "grad_norm": 0.8262777941969183,
      "learning_rate": 4.052831155712403e-06,
      "loss": 0.2895543098449707,
      "num_input_tokens_seen": 530843408,
      "step": 16655,
      "train_runtime": 43710.8212,
      "train_tokens_per_second": 12144.439
    },
    {
      "epoch": 1.1217344465391865,
      "grad_norm": 0.681354448943446,
      "learning_rate": 4.050235081185544e-06,
      "loss": 0.26581830978393556,
      "num_input_tokens_seen": 531001680,
      "step": 16660,
      "train_runtime": 43723.8715,
      "train_tokens_per_second": 12144.434
    },
    {
      "epoch": 1.1220711015351468,
      "grad_norm": 0.8463176991030703,
      "learning_rate": 4.047639272257415e-06,
      "loss": 0.30289411544799805,
      "num_input_tokens_seen": 531164384,
      "step": 16665,
      "train_runtime": 43737.2279,
      "train_tokens_per_second": 12144.446
    },
    {
      "epoch": 1.122407756531107,
      "grad_norm": 0.7271047054456528,
      "learning_rate": 4.045043729653927e-06,
      "loss": 0.27604360580444337,
      "num_input_tokens_seen": 531327568,
      "step": 16670,
      "train_runtime": 43750.3117,
      "train_tokens_per_second": 12144.544
    },
    {
      "epoch": 1.122744411527067,
      "grad_norm": 0.7827703865542933,
      "learning_rate": 4.0424484541009175e-06,
      "loss": 0.3208983659744263,
      "num_input_tokens_seen": 531490688,
      "step": 16675,
      "train_runtime": 43762.7727,
      "train_tokens_per_second": 12144.813
    },
    {
      "epoch": 1.1230810665230273,
      "grad_norm": 0.8113315064570628,
      "learning_rate": 4.039853446324141e-06,
      "loss": 0.2941084146499634,
      "num_input_tokens_seen": 531649392,
      "step": 16680,
      "train_runtime": 43776.1322,
      "train_tokens_per_second": 12144.732
    },
    {
      "epoch": 1.1234177215189873,
      "grad_norm": 0.8475446734045625,
      "learning_rate": 4.037258707049289e-06,
      "loss": 0.2989093780517578,
      "num_input_tokens_seen": 531805968,
      "step": 16685,
      "train_runtime": 43789.2538,
      "train_tokens_per_second": 12144.668
    },
    {
      "epoch": 1.1237543765149476,
      "grad_norm": 0.6924909422030741,
      "learning_rate": 4.034664237001969e-06,
      "loss": 0.2961087703704834,
      "num_input_tokens_seen": 531969024,
      "step": 16690,
      "train_runtime": 43802.1133,
      "train_tokens_per_second": 12144.826
    },
    {
      "epoch": 1.1240910315109076,
      "grad_norm": 0.8351516199395201,
      "learning_rate": 4.032070036907716e-06,
      "loss": 0.3192195177078247,
      "num_input_tokens_seen": 532125688,
      "step": 16695,
      "train_runtime": 43814.7815,
      "train_tokens_per_second": 12144.89
    },
    {
      "epoch": 1.1244276865068679,
      "grad_norm": 0.8427751125560105,
      "learning_rate": 4.029476107491991e-06,
      "loss": 0.32015242576599123,
      "num_input_tokens_seen": 532283456,
      "step": 16700,
      "train_runtime": 43827.3178,
      "train_tokens_per_second": 12145.016
    },
    {
      "epoch": 1.124764341502828,
      "grad_norm": 0.872841252197788,
      "learning_rate": 4.0268824494801784e-06,
      "loss": 0.30301501750946047,
      "num_input_tokens_seen": 532442440,
      "step": 16705,
      "train_runtime": 43839.9885,
      "train_tokens_per_second": 12145.132
    },
    {
      "epoch": 1.1251009964987881,
      "grad_norm": 0.7604015208835635,
      "learning_rate": 4.024289063597585e-06,
      "loss": 0.3071014165878296,
      "num_input_tokens_seen": 532599904,
      "step": 16710,
      "train_runtime": 43853.5971,
      "train_tokens_per_second": 12144.954
    },
    {
      "epoch": 1.1254376514947482,
      "grad_norm": 0.8372602312254506,
      "learning_rate": 4.021695950569443e-06,
      "loss": 0.30000388622283936,
      "num_input_tokens_seen": 532757808,
      "step": 16715,
      "train_runtime": 43866.3862,
      "train_tokens_per_second": 12145.012
    },
    {
      "epoch": 1.1257743064907082,
      "grad_norm": 0.848013159357868,
      "learning_rate": 4.01910311112091e-06,
      "loss": 0.31327269077301023,
      "num_input_tokens_seen": 532919472,
      "step": 16720,
      "train_runtime": 43880.4437,
      "train_tokens_per_second": 12144.806
    },
    {
      "epoch": 1.1261109614866684,
      "grad_norm": 0.8372382746973116,
      "learning_rate": 4.016510545977064e-06,
      "loss": 0.3042671918869019,
      "num_input_tokens_seen": 533083312,
      "step": 16725,
      "train_runtime": 43892.8645,
      "train_tokens_per_second": 12145.102
    },
    {
      "epoch": 1.1264476164826287,
      "grad_norm": 1.002446156871759,
      "learning_rate": 4.013918255862908e-06,
      "loss": 0.31206183433532714,
      "num_input_tokens_seen": 533238416,
      "step": 16730,
      "train_runtime": 43906.0774,
      "train_tokens_per_second": 12144.98
    },
    {
      "epoch": 1.1267842714785887,
      "grad_norm": 0.7782055563668424,
      "learning_rate": 4.011326241503369e-06,
      "loss": 0.2938835144042969,
      "num_input_tokens_seen": 533396536,
      "step": 16735,
      "train_runtime": 43918.7653,
      "train_tokens_per_second": 12145.071
    },
    {
      "epoch": 1.127120926474549,
      "grad_norm": 0.764581520508395,
      "learning_rate": 4.00873450362329e-06,
      "loss": 0.314333176612854,
      "num_input_tokens_seen": 533558696,
      "step": 16740,
      "train_runtime": 43931.6648,
      "train_tokens_per_second": 12145.196
    },
    {
      "epoch": 1.127457581470509,
      "grad_norm": 0.755947972872487,
      "learning_rate": 4.0061430429474476e-06,
      "loss": 0.31098499298095705,
      "num_input_tokens_seen": 533712368,
      "step": 16745,
      "train_runtime": 43944.7345,
      "train_tokens_per_second": 12145.081
    },
    {
      "epoch": 1.1277942364664693,
      "grad_norm": 0.7593443960403506,
      "learning_rate": 4.003551860200534e-06,
      "loss": 0.3035487174987793,
      "num_input_tokens_seen": 533867016,
      "step": 16750,
      "train_runtime": 43957.4589,
      "train_tokens_per_second": 12145.084
    },
    {
      "epoch": 1.1281308914624293,
      "grad_norm": 0.7711903899630963,
      "learning_rate": 4.000960956107167e-06,
      "loss": 0.33294000625610354,
      "num_input_tokens_seen": 534022360,
      "step": 16755,
      "train_runtime": 43970.6044,
      "train_tokens_per_second": 12144.986
    },
    {
      "epoch": 1.1284675464583895,
      "grad_norm": 0.8281540525610708,
      "learning_rate": 3.998370331391881e-06,
      "loss": 0.29976511001586914,
      "num_input_tokens_seen": 534175392,
      "step": 16760,
      "train_runtime": 43983.9755,
      "train_tokens_per_second": 12144.773
    },
    {
      "epoch": 1.1288042014543496,
      "grad_norm": 0.7783058500319283,
      "learning_rate": 3.995779986779139e-06,
      "loss": 0.299426007270813,
      "num_input_tokens_seen": 534329608,
      "step": 16765,
      "train_runtime": 43997.7702,
      "train_tokens_per_second": 12144.47
    },
    {
      "epoch": 1.1291408564503098,
      "grad_norm": 0.7614041058927454,
      "learning_rate": 3.9931899229933204e-06,
      "loss": 0.3206852674484253,
      "num_input_tokens_seen": 534488904,
      "step": 16770,
      "train_runtime": 44010.9183,
      "train_tokens_per_second": 12144.462
    },
    {
      "epoch": 1.1294775114462698,
      "grad_norm": 0.7699015784928196,
      "learning_rate": 3.990600140758731e-06,
      "loss": 0.29125022888183594,
      "num_input_tokens_seen": 534651376,
      "step": 16775,
      "train_runtime": 44023.5763,
      "train_tokens_per_second": 12144.66
    },
    {
      "epoch": 1.12981416644223,
      "grad_norm": 0.7163997109366104,
      "learning_rate": 3.988010640799591e-06,
      "loss": 0.2984292507171631,
      "num_input_tokens_seen": 534813448,
      "step": 16780,
      "train_runtime": 44036.4786,
      "train_tokens_per_second": 12144.782
    },
    {
      "epoch": 1.1301508214381901,
      "grad_norm": 0.8695339779715218,
      "learning_rate": 3.985421423840051e-06,
      "loss": 0.33612785339355467,
      "num_input_tokens_seen": 534967624,
      "step": 16785,
      "train_runtime": 44049.3083,
      "train_tokens_per_second": 12144.745
    },
    {
      "epoch": 1.1304874764341504,
      "grad_norm": 0.8500388113489217,
      "learning_rate": 3.982832490604173e-06,
      "loss": 0.28299052715301515,
      "num_input_tokens_seen": 535129816,
      "step": 16790,
      "train_runtime": 44061.9063,
      "train_tokens_per_second": 12144.954
    },
    {
      "epoch": 1.1308241314301104,
      "grad_norm": 0.8741870304525861,
      "learning_rate": 3.9802438418159465e-06,
      "loss": 0.30555727481842043,
      "num_input_tokens_seen": 535293656,
      "step": 16795,
      "train_runtime": 44074.3279,
      "train_tokens_per_second": 12145.248
    },
    {
      "epoch": 1.1311607864260707,
      "grad_norm": 0.8275118782174252,
      "learning_rate": 3.977655478199275e-06,
      "loss": 0.3156377077102661,
      "num_input_tokens_seen": 535455832,
      "step": 16800,
      "train_runtime": 44086.8875,
      "train_tokens_per_second": 12145.467
    },
    {
      "epoch": 1.1314974414220307,
      "grad_norm": 0.8268152983326666,
      "learning_rate": 3.975067400477993e-06,
      "loss": 0.3336094856262207,
      "num_input_tokens_seen": 535604992,
      "step": 16805,
      "train_runtime": 44099.5493,
      "train_tokens_per_second": 12145.362
    },
    {
      "epoch": 1.131834096417991,
      "grad_norm": 0.8703703138733289,
      "learning_rate": 3.972479609375842e-06,
      "loss": 0.3120142936706543,
      "num_input_tokens_seen": 535762856,
      "step": 16810,
      "train_runtime": 44112.0283,
      "train_tokens_per_second": 12145.505
    },
    {
      "epoch": 1.132170751413951,
      "grad_norm": 0.7819466382200351,
      "learning_rate": 3.969892105616492e-06,
      "loss": 0.3231369495391846,
      "num_input_tokens_seen": 535926696,
      "step": 16815,
      "train_runtime": 44124.4668,
      "train_tokens_per_second": 12145.794
    },
    {
      "epoch": 1.1325074064099112,
      "grad_norm": 0.8270335347693695,
      "learning_rate": 3.967304889923529e-06,
      "loss": 0.29298269748687744,
      "num_input_tokens_seen": 536089032,
      "step": 16820,
      "train_runtime": 44136.9047,
      "train_tokens_per_second": 12146.05
    },
    {
      "epoch": 1.1328440614058712,
      "grad_norm": 0.8376474886061578,
      "learning_rate": 3.964717963020463e-06,
      "loss": 0.29854645729064944,
      "num_input_tokens_seen": 536245584,
      "step": 16825,
      "train_runtime": 44149.904,
      "train_tokens_per_second": 12146.019
    },
    {
      "epoch": 1.1331807164018315,
      "grad_norm": 0.879058724499406,
      "learning_rate": 3.962131325630715e-06,
      "loss": 0.29366650581359866,
      "num_input_tokens_seen": 536408920,
      "step": 16830,
      "train_runtime": 44162.875,
      "train_tokens_per_second": 12146.15
    },
    {
      "epoch": 1.1335173713977915,
      "grad_norm": 0.916843758393496,
      "learning_rate": 3.959544978477634e-06,
      "loss": 0.3356415510177612,
      "num_input_tokens_seen": 536565656,
      "step": 16835,
      "train_runtime": 44175.9399,
      "train_tokens_per_second": 12146.106
    },
    {
      "epoch": 1.1338540263937518,
      "grad_norm": 0.7347836325822532,
      "learning_rate": 3.9569589222844805e-06,
      "loss": 0.3050784349441528,
      "num_input_tokens_seen": 536727544,
      "step": 16840,
      "train_runtime": 44189.4018,
      "train_tokens_per_second": 12146.069
    },
    {
      "epoch": 1.1341906813897118,
      "grad_norm": 0.7579372765853083,
      "learning_rate": 3.954373157774439e-06,
      "loss": 0.31089119911193847,
      "num_input_tokens_seen": 536886440,
      "step": 16845,
      "train_runtime": 44201.9278,
      "train_tokens_per_second": 12146.222
    },
    {
      "epoch": 1.134527336385672,
      "grad_norm": 0.8404684636704761,
      "learning_rate": 3.951787685670609e-06,
      "loss": 0.2953361749649048,
      "num_input_tokens_seen": 537045840,
      "step": 16850,
      "train_runtime": 44214.5537,
      "train_tokens_per_second": 12146.359
    },
    {
      "epoch": 1.134863991381632,
      "grad_norm": 0.8611130948039886,
      "learning_rate": 3.949202506696012e-06,
      "loss": 0.3068122386932373,
      "num_input_tokens_seen": 537202272,
      "step": 16855,
      "train_runtime": 44227.9423,
      "train_tokens_per_second": 12146.219
    },
    {
      "epoch": 1.1352006463775923,
      "grad_norm": 0.7623656836361321,
      "learning_rate": 3.946617621573581e-06,
      "loss": 0.29924960136413575,
      "num_input_tokens_seen": 537365400,
      "step": 16860,
      "train_runtime": 44240.8155,
      "train_tokens_per_second": 12146.372
    },
    {
      "epoch": 1.1355373013735524,
      "grad_norm": 0.7745188963701689,
      "learning_rate": 3.944033031026175e-06,
      "loss": 0.27745418548583983,
      "num_input_tokens_seen": 537518480,
      "step": 16865,
      "train_runtime": 44254.2264,
      "train_tokens_per_second": 12146.15
    },
    {
      "epoch": 1.1358739563695126,
      "grad_norm": 0.905542809681017,
      "learning_rate": 3.941448735776563e-06,
      "loss": 0.298679780960083,
      "num_input_tokens_seen": 537677712,
      "step": 16870,
      "train_runtime": 44267.487,
      "train_tokens_per_second": 12146.109
    },
    {
      "epoch": 1.1362106113654726,
      "grad_norm": 0.7570444410974061,
      "learning_rate": 3.938864736547439e-06,
      "loss": 0.30825071334838866,
      "num_input_tokens_seen": 537839096,
      "step": 16875,
      "train_runtime": 44280.6112,
      "train_tokens_per_second": 12146.153
    },
    {
      "epoch": 1.1365472663614329,
      "grad_norm": 0.7487674134882696,
      "learning_rate": 3.936281034061405e-06,
      "loss": 0.2948976993560791,
      "num_input_tokens_seen": 538002160,
      "step": 16880,
      "train_runtime": 44294.1048,
      "train_tokens_per_second": 12146.135
    },
    {
      "epoch": 1.136883921357393,
      "grad_norm": 0.7575851814034306,
      "learning_rate": 3.93369762904099e-06,
      "loss": 0.31870982646942136,
      "num_input_tokens_seen": 538159096,
      "step": 16885,
      "train_runtime": 44307.4066,
      "train_tokens_per_second": 12146.03
    },
    {
      "epoch": 1.1372205763533532,
      "grad_norm": 0.7193113070772132,
      "learning_rate": 3.931114522208631e-06,
      "loss": 0.30561189651489257,
      "num_input_tokens_seen": 538318832,
      "step": 16890,
      "train_runtime": 44321.0629,
      "train_tokens_per_second": 12145.892
    },
    {
      "epoch": 1.1375572313493132,
      "grad_norm": 0.8138741458729092,
      "learning_rate": 3.928531714286689e-06,
      "loss": 0.3008100509643555,
      "num_input_tokens_seen": 538482672,
      "step": 16895,
      "train_runtime": 44333.4859,
      "train_tokens_per_second": 12146.184
    },
    {
      "epoch": 1.1378938863452734,
      "grad_norm": 0.8011588227732436,
      "learning_rate": 3.925949205997434e-06,
      "loss": 0.3239983081817627,
      "num_input_tokens_seen": 538640440,
      "step": 16900,
      "train_runtime": 44346.359,
      "train_tokens_per_second": 12146.216
    },
    {
      "epoch": 1.1382305413412335,
      "grad_norm": 0.7398460559280121,
      "learning_rate": 3.923366998063062e-06,
      "loss": 0.2898430824279785,
      "num_input_tokens_seen": 538795048,
      "step": 16905,
      "train_runtime": 44359.141,
      "train_tokens_per_second": 12146.201
    },
    {
      "epoch": 1.1385671963371937,
      "grad_norm": 0.7360182338583346,
      "learning_rate": 3.920785091205674e-06,
      "loss": 0.27588777542114257,
      "num_input_tokens_seen": 538956056,
      "step": 16910,
      "train_runtime": 44372.6579,
      "train_tokens_per_second": 12146.13
    },
    {
      "epoch": 1.1389038513331537,
      "grad_norm": 0.90712532484992,
      "learning_rate": 3.918203486147294e-06,
      "loss": 0.3511468172073364,
      "num_input_tokens_seen": 539114568,
      "step": 16915,
      "train_runtime": 44386.3731,
      "train_tokens_per_second": 12145.948
    },
    {
      "epoch": 1.139240506329114,
      "grad_norm": 0.8972206889657058,
      "learning_rate": 3.9156221836098594e-06,
      "loss": 0.33047473430633545,
      "num_input_tokens_seen": 539271528,
      "step": 16920,
      "train_runtime": 44399.2903,
      "train_tokens_per_second": 12145.949
    },
    {
      "epoch": 1.139577161325074,
      "grad_norm": 0.8641607801911978,
      "learning_rate": 3.913041184315224e-06,
      "loss": 0.3061272144317627,
      "num_input_tokens_seen": 539432848,
      "step": 16925,
      "train_runtime": 44412.8532,
      "train_tokens_per_second": 12145.872
    },
    {
      "epoch": 1.1399138163210343,
      "grad_norm": 0.8575801137444691,
      "learning_rate": 3.910460488985154e-06,
      "loss": 0.3044703245162964,
      "num_input_tokens_seen": 539594216,
      "step": 16930,
      "train_runtime": 44425.6885,
      "train_tokens_per_second": 12145.996
    },
    {
      "epoch": 1.1402504713169943,
      "grad_norm": 0.7781826121372492,
      "learning_rate": 3.9078800983413355e-06,
      "loss": 0.3161719799041748,
      "num_input_tokens_seen": 539751984,
      "step": 16935,
      "train_runtime": 44438.22,
      "train_tokens_per_second": 12146.121
    },
    {
      "epoch": 1.1405871263129546,
      "grad_norm": 0.7292895761663459,
      "learning_rate": 3.905300013105365e-06,
      "loss": 0.27150721549987794,
      "num_input_tokens_seen": 539913264,
      "step": 16940,
      "train_runtime": 44451.7458,
      "train_tokens_per_second": 12146.053
    },
    {
      "epoch": 1.1409237813089146,
      "grad_norm": 0.7944189981708061,
      "learning_rate": 3.902720233998754e-06,
      "loss": 0.2900334358215332,
      "num_input_tokens_seen": 540072160,
      "step": 16945,
      "train_runtime": 44465.7403,
      "train_tokens_per_second": 12145.804
    },
    {
      "epoch": 1.1412604363048748,
      "grad_norm": 0.7618614198635505,
      "learning_rate": 3.90014076174293e-06,
      "loss": 0.309930419921875,
      "num_input_tokens_seen": 540232376,
      "step": 16950,
      "train_runtime": 44478.7548,
      "train_tokens_per_second": 12145.852
    },
    {
      "epoch": 1.1415970913008349,
      "grad_norm": 0.7377014664591213,
      "learning_rate": 3.897561597059237e-06,
      "loss": 0.29001574516296386,
      "num_input_tokens_seen": 540393464,
      "step": 16955,
      "train_runtime": 44492.251,
      "train_tokens_per_second": 12145.788
    },
    {
      "epoch": 1.1419337462967951,
      "grad_norm": 0.7989097012956159,
      "learning_rate": 3.894982740668927e-06,
      "loss": 0.29951283931732176,
      "num_input_tokens_seen": 540546384,
      "step": 16960,
      "train_runtime": 44505.7638,
      "train_tokens_per_second": 12145.537
    },
    {
      "epoch": 1.1422704012927551,
      "grad_norm": 0.8737483601389187,
      "learning_rate": 3.89240419329317e-06,
      "loss": 0.3264753341674805,
      "num_input_tokens_seen": 540710224,
      "step": 16965,
      "train_runtime": 44518.1758,
      "train_tokens_per_second": 12145.831
    },
    {
      "epoch": 1.1426070562887154,
      "grad_norm": 0.8557395576573769,
      "learning_rate": 3.889825955653046e-06,
      "loss": 0.3058238744735718,
      "num_input_tokens_seen": 540866496,
      "step": 16970,
      "train_runtime": 44530.9682,
      "train_tokens_per_second": 12145.851
    },
    {
      "epoch": 1.1429437112846754,
      "grad_norm": 0.8093664160139745,
      "learning_rate": 3.887248028469558e-06,
      "loss": 0.2706881046295166,
      "num_input_tokens_seen": 541029160,
      "step": 16975,
      "train_runtime": 44544.2609,
      "train_tokens_per_second": 12145.878
    },
    {
      "epoch": 1.1432803662806357,
      "grad_norm": 0.6925577760919885,
      "learning_rate": 3.884670412463607e-06,
      "loss": 0.2888359546661377,
      "num_input_tokens_seen": 541190160,
      "step": 16980,
      "train_runtime": 44558.0227,
      "train_tokens_per_second": 12145.74
    },
    {
      "epoch": 1.1436170212765957,
      "grad_norm": 3.3042252360763786,
      "learning_rate": 3.8820931083560194e-06,
      "loss": 0.3178676128387451,
      "num_input_tokens_seen": 541348240,
      "step": 16985,
      "train_runtime": 44570.8097,
      "train_tokens_per_second": 12145.802
    },
    {
      "epoch": 1.143953676272556,
      "grad_norm": 0.8098723525468328,
      "learning_rate": 3.879516116867531e-06,
      "loss": 0.2848562717437744,
      "num_input_tokens_seen": 541503760,
      "step": 16990,
      "train_runtime": 44583.7478,
      "train_tokens_per_second": 12145.766
    },
    {
      "epoch": 1.144290331268516,
      "grad_norm": 0.9223004347735128,
      "learning_rate": 3.876939438718786e-06,
      "loss": 0.3077042579650879,
      "num_input_tokens_seen": 541664680,
      "step": 16995,
      "train_runtime": 44597.4478,
      "train_tokens_per_second": 12145.643
    },
    {
      "epoch": 1.1446269862644762,
      "grad_norm": 0.7195326616603016,
      "learning_rate": 3.874363074630345e-06,
      "loss": 0.29122610092163087,
      "num_input_tokens_seen": 541820576,
      "step": 17000,
      "train_runtime": 44611.6145,
      "train_tokens_per_second": 12145.281
    },
    {
      "epoch": 1.1449636412604363,
      "grad_norm": 0.836886158853234,
      "learning_rate": 3.871787025322681e-06,
      "loss": 0.31101183891296386,
      "num_input_tokens_seen": 541979120,
      "step": 17005,
      "train_runtime": 44624.1962,
      "train_tokens_per_second": 12145.409
    },
    {
      "epoch": 1.1453002962563965,
      "grad_norm": 0.7986969648231006,
      "learning_rate": 3.86921129151618e-06,
      "loss": 0.3107777118682861,
      "num_input_tokens_seen": 542132792,
      "step": 17010,
      "train_runtime": 44637.361,
      "train_tokens_per_second": 12145.27
    },
    {
      "epoch": 1.1456369512523565,
      "grad_norm": 0.8334286167348997,
      "learning_rate": 3.866635873931133e-06,
      "loss": 0.3248122692108154,
      "num_input_tokens_seen": 542296320,
      "step": 17015,
      "train_runtime": 44650.4076,
      "train_tokens_per_second": 12145.383
    },
    {
      "epoch": 1.1459736062483168,
      "grad_norm": 0.7927407948835387,
      "learning_rate": 3.864060773287751e-06,
      "loss": 0.3016833305358887,
      "num_input_tokens_seen": 542453528,
      "step": 17020,
      "train_runtime": 44663.0007,
      "train_tokens_per_second": 12145.479
    },
    {
      "epoch": 1.1463102612442768,
      "grad_norm": 0.7548343564407981,
      "learning_rate": 3.861485990306151e-06,
      "loss": 0.3086510181427002,
      "num_input_tokens_seen": 542617112,
      "step": 17025,
      "train_runtime": 44675.9328,
      "train_tokens_per_second": 12145.625
    },
    {
      "epoch": 1.146646916240237,
      "grad_norm": 0.8439268572112969,
      "learning_rate": 3.858911525706365e-06,
      "loss": 0.2968446731567383,
      "num_input_tokens_seen": 542778208,
      "step": 17030,
      "train_runtime": 44688.9178,
      "train_tokens_per_second": 12145.7
    },
    {
      "epoch": 1.146983571236197,
      "grad_norm": 0.9103247109831742,
      "learning_rate": 3.85633738020833e-06,
      "loss": 0.3424351215362549,
      "num_input_tokens_seen": 542941192,
      "step": 17035,
      "train_runtime": 44701.7266,
      "train_tokens_per_second": 12145.866
    },
    {
      "epoch": 1.1473202262321573,
      "grad_norm": 0.814960825615862,
      "learning_rate": 3.853763554531902e-06,
      "loss": 0.31267409324645995,
      "num_input_tokens_seen": 543098856,
      "step": 17040,
      "train_runtime": 44714.7156,
      "train_tokens_per_second": 12145.864
    },
    {
      "epoch": 1.1476568812281174,
      "grad_norm": 0.8179817694662539,
      "learning_rate": 3.851190049396839e-06,
      "loss": 0.2735781669616699,
      "num_input_tokens_seen": 543259408,
      "step": 17045,
      "train_runtime": 44727.8779,
      "train_tokens_per_second": 12145.879
    },
    {
      "epoch": 1.1479935362240776,
      "grad_norm": 0.7591411157060375,
      "learning_rate": 3.848616865522815e-06,
      "loss": 0.29964208602905273,
      "num_input_tokens_seen": 543419736,
      "step": 17050,
      "train_runtime": 44741.3544,
      "train_tokens_per_second": 12145.804
    },
    {
      "epoch": 1.1483301912200377,
      "grad_norm": 0.7729175677960473,
      "learning_rate": 3.846044003629414e-06,
      "loss": 0.31813759803771974,
      "num_input_tokens_seen": 543580640,
      "step": 17055,
      "train_runtime": 44753.9635,
      "train_tokens_per_second": 12145.978
    },
    {
      "epoch": 1.148666846215998,
      "grad_norm": 0.7172630460580574,
      "learning_rate": 3.843471464436128e-06,
      "loss": 0.27354888916015624,
      "num_input_tokens_seen": 543733448,
      "step": 17060,
      "train_runtime": 44767.4589,
      "train_tokens_per_second": 12145.73
    },
    {
      "epoch": 1.149003501211958,
      "grad_norm": 0.8323765798131523,
      "learning_rate": 3.840899248662358e-06,
      "loss": 0.32833089828491213,
      "num_input_tokens_seen": 543885248,
      "step": 17065,
      "train_runtime": 44779.9847,
      "train_tokens_per_second": 12145.722
    },
    {
      "epoch": 1.1493401562079182,
      "grad_norm": 0.8543409205123543,
      "learning_rate": 3.838327357027419e-06,
      "loss": 0.3162813186645508,
      "num_input_tokens_seen": 544041384,
      "step": 17070,
      "train_runtime": 44793.2339,
      "train_tokens_per_second": 12145.615
    },
    {
      "epoch": 1.1496768112038782,
      "grad_norm": 0.7892023718918849,
      "learning_rate": 3.835755790250528e-06,
      "loss": 0.3141947269439697,
      "num_input_tokens_seen": 544204440,
      "step": 17075,
      "train_runtime": 44806.0905,
      "train_tokens_per_second": 12145.769
    },
    {
      "epoch": 1.1500134661998385,
      "grad_norm": 0.8647430014610052,
      "learning_rate": 3.83318454905082e-06,
      "loss": 0.3012948751449585,
      "num_input_tokens_seen": 544367704,
      "step": 17080,
      "train_runtime": 44818.9102,
      "train_tokens_per_second": 12145.938
    },
    {
      "epoch": 1.1503501211957985,
      "grad_norm": 0.8069631672690868,
      "learning_rate": 3.830613634147331e-06,
      "loss": 0.31517179012298585,
      "num_input_tokens_seen": 544525640,
      "step": 17085,
      "train_runtime": 44831.6477,
      "train_tokens_per_second": 12146.01
    },
    {
      "epoch": 1.1506867761917587,
      "grad_norm": 0.7556796830990403,
      "learning_rate": 3.82804304625901e-06,
      "loss": 0.3082268714904785,
      "num_input_tokens_seen": 544688312,
      "step": 17090,
      "train_runtime": 44844.3404,
      "train_tokens_per_second": 12146.2
    },
    {
      "epoch": 1.1510234311877188,
      "grad_norm": 0.8225563710288478,
      "learning_rate": 3.825472786104715e-06,
      "loss": 0.28848557472229003,
      "num_input_tokens_seen": 544851280,
      "step": 17095,
      "train_runtime": 44857.1556,
      "train_tokens_per_second": 12146.363
    },
    {
      "epoch": 1.151360086183679,
      "grad_norm": 0.9089195581934792,
      "learning_rate": 3.8229028544032095e-06,
      "loss": 0.3184187889099121,
      "num_input_tokens_seen": 545003872,
      "step": 17100,
      "train_runtime": 44870.1763,
      "train_tokens_per_second": 12146.239
    },
    {
      "epoch": 1.151696741179639,
      "grad_norm": 0.9603453245357494,
      "learning_rate": 3.8203332518731666e-06,
      "loss": 0.30712997913360596,
      "num_input_tokens_seen": 545161944,
      "step": 17105,
      "train_runtime": 44883.3028,
      "train_tokens_per_second": 12146.208
    },
    {
      "epoch": 1.1520333961755993,
      "grad_norm": 0.786889082664916,
      "learning_rate": 3.8177639792331695e-06,
      "loss": 0.3168989658355713,
      "num_input_tokens_seen": 545321336,
      "step": 17110,
      "train_runtime": 44895.7253,
      "train_tokens_per_second": 12146.398
    },
    {
      "epoch": 1.1523700511715593,
      "grad_norm": 0.9223204520474808,
      "learning_rate": 3.815195037201704e-06,
      "loss": 0.3337859630584717,
      "num_input_tokens_seen": 545476080,
      "step": 17115,
      "train_runtime": 44908.4002,
      "train_tokens_per_second": 12146.415
    },
    {
      "epoch": 1.1527067061675196,
      "grad_norm": 1.284631700549255,
      "learning_rate": 3.8126264264971694e-06,
      "loss": 0.3208210229873657,
      "num_input_tokens_seen": 545637128,
      "step": 17120,
      "train_runtime": 44921.4507,
      "train_tokens_per_second": 12146.472
    },
    {
      "epoch": 1.1530433611634796,
      "grad_norm": 0.7800533053499017,
      "learning_rate": 3.8100581478378673e-06,
      "loss": 0.3071185827255249,
      "num_input_tokens_seen": 545797368,
      "step": 17125,
      "train_runtime": 44933.8867,
      "train_tokens_per_second": 12146.676
    },
    {
      "epoch": 1.1533800161594399,
      "grad_norm": 0.8343636532263582,
      "learning_rate": 3.807490201942012e-06,
      "loss": 0.3163712978363037,
      "num_input_tokens_seen": 545958336,
      "step": 17130,
      "train_runtime": 44946.3671,
      "train_tokens_per_second": 12146.885
    },
    {
      "epoch": 1.1537166711553999,
      "grad_norm": 0.9149341286024075,
      "learning_rate": 3.804922589527717e-06,
      "loss": 0.2998265981674194,
      "num_input_tokens_seen": 546115536,
      "step": 17135,
      "train_runtime": 44958.8334,
      "train_tokens_per_second": 12147.013
    },
    {
      "epoch": 1.1540533261513601,
      "grad_norm": 0.8353778423904503,
      "learning_rate": 3.8023553113130096e-06,
      "loss": 0.32700958251953127,
      "num_input_tokens_seen": 546278944,
      "step": 17140,
      "train_runtime": 44971.8266,
      "train_tokens_per_second": 12147.137
    },
    {
      "epoch": 1.1543899811473202,
      "grad_norm": 0.8531399788377376,
      "learning_rate": 3.799788368015821e-06,
      "loss": 0.2914307117462158,
      "num_input_tokens_seen": 546439072,
      "step": 17145,
      "train_runtime": 44985.0962,
      "train_tokens_per_second": 12147.114
    },
    {
      "epoch": 1.1547266361432804,
      "grad_norm": 0.8236647368645641,
      "learning_rate": 3.7972217603539873e-06,
      "loss": 0.32138705253601074,
      "num_input_tokens_seen": 546595032,
      "step": 17150,
      "train_runtime": 44998.6978,
      "train_tokens_per_second": 12146.908
    },
    {
      "epoch": 1.1550632911392404,
      "grad_norm": 0.8090598008703196,
      "learning_rate": 3.7946554890452524e-06,
      "loss": 0.30844531059265134,
      "num_input_tokens_seen": 546756032,
      "step": 17155,
      "train_runtime": 45011.1976,
      "train_tokens_per_second": 12147.111
    },
    {
      "epoch": 1.1553999461352007,
      "grad_norm": 0.8536989597370267,
      "learning_rate": 3.792089554807269e-06,
      "loss": 0.316030478477478,
      "num_input_tokens_seen": 546916072,
      "step": 17160,
      "train_runtime": 45023.9921,
      "train_tokens_per_second": 12147.214
    },
    {
      "epoch": 1.1557366011311607,
      "grad_norm": 0.8503560434872435,
      "learning_rate": 3.789523958357587e-06,
      "loss": 0.30642380714416506,
      "num_input_tokens_seen": 547073992,
      "step": 17165,
      "train_runtime": 45037.0837,
      "train_tokens_per_second": 12147.19
    },
    {
      "epoch": 1.156073256127121,
      "grad_norm": 0.7794887481482436,
      "learning_rate": 3.7869587004136714e-06,
      "loss": 0.26918396949768064,
      "num_input_tokens_seen": 547236528,
      "step": 17170,
      "train_runtime": 45050.5586,
      "train_tokens_per_second": 12147.164
    },
    {
      "epoch": 1.156409911123081,
      "grad_norm": 0.8250918068175273,
      "learning_rate": 3.7843937816928864e-06,
      "loss": 0.3241715431213379,
      "num_input_tokens_seen": 547397360,
      "step": 17175,
      "train_runtime": 45063.0485,
      "train_tokens_per_second": 12147.366
    },
    {
      "epoch": 1.1567465661190413,
      "grad_norm": 0.84984457848406,
      "learning_rate": 3.781829202912506e-06,
      "loss": 0.3171783447265625,
      "num_input_tokens_seen": 547553520,
      "step": 17180,
      "train_runtime": 45076.5319,
      "train_tokens_per_second": 12147.197
    },
    {
      "epoch": 1.1570832211150013,
      "grad_norm": 0.8519543563528018,
      "learning_rate": 3.7792649647897023e-06,
      "loss": 0.3070185661315918,
      "num_input_tokens_seen": 547714776,
      "step": 17185,
      "train_runtime": 45089.0287,
      "train_tokens_per_second": 12147.407
    },
    {
      "epoch": 1.1574198761109615,
      "grad_norm": 0.7472613660191926,
      "learning_rate": 3.7767010680415595e-06,
      "loss": 0.3007053375244141,
      "num_input_tokens_seen": 547874800,
      "step": 17190,
      "train_runtime": 45102.5038,
      "train_tokens_per_second": 12147.326
    },
    {
      "epoch": 1.1577565311069216,
      "grad_norm": 0.7878331235468579,
      "learning_rate": 3.774137513385061e-06,
      "loss": 0.29999527931213377,
      "num_input_tokens_seen": 548034608,
      "step": 17195,
      "train_runtime": 45115.3997,
      "train_tokens_per_second": 12147.396
    },
    {
      "epoch": 1.1580931861028818,
      "grad_norm": 0.7961222031940645,
      "learning_rate": 3.7715743015371005e-06,
      "loss": 0.3082253456115723,
      "num_input_tokens_seen": 548198304,
      "step": 17200,
      "train_runtime": 45128.6497,
      "train_tokens_per_second": 12147.456
    },
    {
      "epoch": 1.1584298410988418,
      "grad_norm": 0.7044360093781137,
      "learning_rate": 3.769011433214467e-06,
      "loss": 0.2966219425201416,
      "num_input_tokens_seen": 548359584,
      "step": 17205,
      "train_runtime": 45141.6554,
      "train_tokens_per_second": 12147.529
    },
    {
      "epoch": 1.158766496094802,
      "grad_norm": 0.8883857700854241,
      "learning_rate": 3.766448909133863e-06,
      "loss": 0.3264200687408447,
      "num_input_tokens_seen": 548521496,
      "step": 17210,
      "train_runtime": 45154.7949,
      "train_tokens_per_second": 12147.58
    },
    {
      "epoch": 1.1591031510907621,
      "grad_norm": 0.7349002287122213,
      "learning_rate": 3.7638867300118864e-06,
      "loss": 0.28376796245574953,
      "num_input_tokens_seen": 548682576,
      "step": 17215,
      "train_runtime": 45168.7585,
      "train_tokens_per_second": 12147.391
    },
    {
      "epoch": 1.1594398060867224,
      "grad_norm": 0.7947616639053485,
      "learning_rate": 3.7613248965650447e-06,
      "loss": 0.29664950370788573,
      "num_input_tokens_seen": 548844600,
      "step": 17220,
      "train_runtime": 45181.3395,
      "train_tokens_per_second": 12147.595
    },
    {
      "epoch": 1.1597764610826824,
      "grad_norm": 0.7575118280644969,
      "learning_rate": 3.758763409509745e-06,
      "loss": 0.2930509805679321,
      "num_input_tokens_seen": 549006720,
      "step": 17225,
      "train_runtime": 45193.9393,
      "train_tokens_per_second": 12147.795
    },
    {
      "epoch": 1.1601131160786426,
      "grad_norm": 0.8176902974561484,
      "learning_rate": 3.7562022695623013e-06,
      "loss": 0.3191149473190308,
      "num_input_tokens_seen": 549168280,
      "step": 17230,
      "train_runtime": 45207.0922,
      "train_tokens_per_second": 12147.835
    },
    {
      "epoch": 1.1604497710746027,
      "grad_norm": 0.6868655985057994,
      "learning_rate": 3.7536414774389252e-06,
      "loss": 0.29757959842681886,
      "num_input_tokens_seen": 549328064,
      "step": 17235,
      "train_runtime": 45220.3702,
      "train_tokens_per_second": 12147.801
    },
    {
      "epoch": 1.160786426070563,
      "grad_norm": 0.9444973451145825,
      "learning_rate": 3.7510810338557363e-06,
      "loss": 0.2959771633148193,
      "num_input_tokens_seen": 549484432,
      "step": 17240,
      "train_runtime": 45233.491,
      "train_tokens_per_second": 12147.734
    },
    {
      "epoch": 1.161123081066523,
      "grad_norm": 0.7881154356543624,
      "learning_rate": 3.748520939528753e-06,
      "loss": 0.30174553394317627,
      "num_input_tokens_seen": 549645136,
      "step": 17245,
      "train_runtime": 45246.2161,
      "train_tokens_per_second": 12147.87
    },
    {
      "epoch": 1.1614597360624832,
      "grad_norm": 0.7256015734046158,
      "learning_rate": 3.7459611951739007e-06,
      "loss": 0.28409805297851565,
      "num_input_tokens_seen": 549806112,
      "step": 17250,
      "train_runtime": 45258.6883,
      "train_tokens_per_second": 12148.079
    },
    {
      "epoch": 1.1617963910584432,
      "grad_norm": 0.8569281465968295,
      "learning_rate": 3.7434018015069984e-06,
      "loss": 0.31328768730163575,
      "num_input_tokens_seen": 549967136,
      "step": 17255,
      "train_runtime": 45272.355,
      "train_tokens_per_second": 12147.968
    },
    {
      "epoch": 1.1621330460544035,
      "grad_norm": 0.8705983444551582,
      "learning_rate": 3.740842759243778e-06,
      "loss": 0.3136750221252441,
      "num_input_tokens_seen": 550127352,
      "step": 17260,
      "train_runtime": 45285.6071,
      "train_tokens_per_second": 12147.951
    },
    {
      "epoch": 1.1624697010503635,
      "grad_norm": 0.7768674997922687,
      "learning_rate": 3.738284069099862e-06,
      "loss": 0.2892464160919189,
      "num_input_tokens_seen": 550289384,
      "step": 17265,
      "train_runtime": 45298.2328,
      "train_tokens_per_second": 12148.142
    },
    {
      "epoch": 1.1628063560463238,
      "grad_norm": 0.71280265279478,
      "learning_rate": 3.735725731790785e-06,
      "loss": 0.3096620082855225,
      "num_input_tokens_seen": 550449632,
      "step": 17270,
      "train_runtime": 45310.9422,
      "train_tokens_per_second": 12148.272
    },
    {
      "epoch": 1.1631430110422838,
      "grad_norm": 0.8267712910645286,
      "learning_rate": 3.733167748031974e-06,
      "loss": 0.30355277061462405,
      "num_input_tokens_seen": 550612192,
      "step": 17275,
      "train_runtime": 45323.5474,
      "train_tokens_per_second": 12148.48
    },
    {
      "epoch": 1.163479666038244,
      "grad_norm": 0.7787240690009112,
      "learning_rate": 3.7306101185387646e-06,
      "loss": 0.2952958822250366,
      "num_input_tokens_seen": 550772864,
      "step": 17280,
      "train_runtime": 45336.2008,
      "train_tokens_per_second": 12148.633
    },
    {
      "epoch": 1.163816321034204,
      "grad_norm": 0.7544324521829344,
      "learning_rate": 3.728052844026386e-06,
      "loss": 0.306009316444397,
      "num_input_tokens_seen": 550935408,
      "step": 17285,
      "train_runtime": 45349.5335,
      "train_tokens_per_second": 12148.646
    },
    {
      "epoch": 1.1641529760301643,
      "grad_norm": 0.9153730047787004,
      "learning_rate": 3.7254959252099745e-06,
      "loss": 0.3008617401123047,
      "num_input_tokens_seen": 551091984,
      "step": 17290,
      "train_runtime": 45363.0471,
      "train_tokens_per_second": 12148.478
    },
    {
      "epoch": 1.1644896310261243,
      "grad_norm": 0.8164412322678456,
      "learning_rate": 3.7229393628045617e-06,
      "loss": 0.28227987289428713,
      "num_input_tokens_seen": 551253512,
      "step": 17295,
      "train_runtime": 45376.1827,
      "train_tokens_per_second": 12148.521
    },
    {
      "epoch": 1.1648262860220846,
      "grad_norm": 0.7889309165450333,
      "learning_rate": 3.720383157525087e-06,
      "loss": 0.28754358291625975,
      "num_input_tokens_seen": 551413992,
      "step": 17300,
      "train_runtime": 45389.125,
      "train_tokens_per_second": 12148.593
    },
    {
      "epoch": 1.1651629410180446,
      "grad_norm": 0.8450470839644866,
      "learning_rate": 3.717827310086379e-06,
      "loss": 0.29293045997619627,
      "num_input_tokens_seen": 551570816,
      "step": 17305,
      "train_runtime": 45402.3576,
      "train_tokens_per_second": 12148.506
    },
    {
      "epoch": 1.1654995960140049,
      "grad_norm": 0.8194621698592489,
      "learning_rate": 3.7152718212031778e-06,
      "loss": 0.2980987787246704,
      "num_input_tokens_seen": 551729072,
      "step": 17310,
      "train_runtime": 45415.2371,
      "train_tokens_per_second": 12148.545
    },
    {
      "epoch": 1.165836251009965,
      "grad_norm": 0.926619065211319,
      "learning_rate": 3.7127166915901123e-06,
      "loss": 0.33600964546203616,
      "num_input_tokens_seen": 551886032,
      "step": 17315,
      "train_runtime": 45429.4406,
      "train_tokens_per_second": 12148.202
    },
    {
      "epoch": 1.1661729060059252,
      "grad_norm": 0.8186364136172581,
      "learning_rate": 3.710161921961721e-06,
      "loss": 0.31142098903656007,
      "num_input_tokens_seen": 552044352,
      "step": 17320,
      "train_runtime": 45442.2533,
      "train_tokens_per_second": 12148.261
    },
    {
      "epoch": 1.1665095610018852,
      "grad_norm": 0.8150016932706918,
      "learning_rate": 3.707607513032434e-06,
      "loss": 0.3039080142974854,
      "num_input_tokens_seen": 552203304,
      "step": 17325,
      "train_runtime": 45455.6107,
      "train_tokens_per_second": 12148.188
    },
    {
      "epoch": 1.1668462159978454,
      "grad_norm": 0.8147878550134559,
      "learning_rate": 3.705053465516587e-06,
      "loss": 0.29843170642852784,
      "num_input_tokens_seen": 552365976,
      "step": 17330,
      "train_runtime": 45468.307,
      "train_tokens_per_second": 12148.373
    },
    {
      "epoch": 1.1671828709938055,
      "grad_norm": 0.7535037497813986,
      "learning_rate": 3.702499780128407e-06,
      "loss": 0.28657941818237304,
      "num_input_tokens_seen": 552522912,
      "step": 17335,
      "train_runtime": 45481.7875,
      "train_tokens_per_second": 12148.223
    },
    {
      "epoch": 1.1675195259897657,
      "grad_norm": 0.8529492339229267,
      "learning_rate": 3.6999464575820266e-06,
      "loss": 0.30574495792388917,
      "num_input_tokens_seen": 552682264,
      "step": 17340,
      "train_runtime": 45494.8939,
      "train_tokens_per_second": 12148.226
    },
    {
      "epoch": 1.1678561809857257,
      "grad_norm": 0.7935911302861638,
      "learning_rate": 3.697393498591473e-06,
      "loss": 0.3005890130996704,
      "num_input_tokens_seen": 552845440,
      "step": 17345,
      "train_runtime": 45507.815,
      "train_tokens_per_second": 12148.363
    },
    {
      "epoch": 1.168192835981686,
      "grad_norm": 0.7460935214464078,
      "learning_rate": 3.694840903870676e-06,
      "loss": 0.34033129215240476,
      "num_input_tokens_seen": 553003512,
      "step": 17350,
      "train_runtime": 45521.0228,
      "train_tokens_per_second": 12148.31
    },
    {
      "epoch": 1.168529490977646,
      "grad_norm": 0.833748691363083,
      "learning_rate": 3.6922886741334564e-06,
      "loss": 0.30861701965332033,
      "num_input_tokens_seen": 553160888,
      "step": 17355,
      "train_runtime": 45533.8378,
      "train_tokens_per_second": 12148.348
    },
    {
      "epoch": 1.1688661459736063,
      "grad_norm": 0.8607726655391488,
      "learning_rate": 3.689736810093541e-06,
      "loss": 0.29722349643707274,
      "num_input_tokens_seen": 553323848,
      "step": 17360,
      "train_runtime": 45547.3855,
      "train_tokens_per_second": 12148.312
    },
    {
      "epoch": 1.1692028009695663,
      "grad_norm": 0.8419115606722761,
      "learning_rate": 3.6871853124645464e-06,
      "loss": 0.3002346992492676,
      "num_input_tokens_seen": 553485640,
      "step": 17365,
      "train_runtime": 45559.876,
      "train_tokens_per_second": 12148.533
    },
    {
      "epoch": 1.1695394559655266,
      "grad_norm": 0.8149321307743822,
      "learning_rate": 3.684634181959994e-06,
      "loss": 0.3011584758758545,
      "num_input_tokens_seen": 553649384,
      "step": 17370,
      "train_runtime": 45572.2891,
      "train_tokens_per_second": 12148.817
    },
    {
      "epoch": 1.1698761109614866,
      "grad_norm": 0.9688274623708251,
      "learning_rate": 3.6820834192932974e-06,
      "loss": 0.30669479370117186,
      "num_input_tokens_seen": 553807576,
      "step": 17375,
      "train_runtime": 45585.0117,
      "train_tokens_per_second": 12148.896
    },
    {
      "epoch": 1.1702127659574468,
      "grad_norm": 0.8270516866998457,
      "learning_rate": 3.679533025177772e-06,
      "loss": 0.33387815952301025,
      "num_input_tokens_seen": 553971416,
      "step": 17380,
      "train_runtime": 45597.4511,
      "train_tokens_per_second": 12149.175
    },
    {
      "epoch": 1.1705494209534069,
      "grad_norm": 0.8129483118355065,
      "learning_rate": 3.6769830003266234e-06,
      "loss": 0.29645040035247805,
      "num_input_tokens_seen": 554130576,
      "step": 17385,
      "train_runtime": 45610.0423,
      "train_tokens_per_second": 12149.311
    },
    {
      "epoch": 1.1708860759493671,
      "grad_norm": 0.7890786876508313,
      "learning_rate": 3.6744333454529607e-06,
      "loss": 0.3048261165618896,
      "num_input_tokens_seen": 554292024,
      "step": 17390,
      "train_runtime": 45622.5242,
      "train_tokens_per_second": 12149.526
    },
    {
      "epoch": 1.1712227309453271,
      "grad_norm": 0.7797851037266244,
      "learning_rate": 3.6718840612697847e-06,
      "loss": 0.301741623878479,
      "num_input_tokens_seen": 554455864,
      "step": 17395,
      "train_runtime": 45634.9629,
      "train_tokens_per_second": 12149.804
    },
    {
      "epoch": 1.1715593859412874,
      "grad_norm": 0.8049698931944371,
      "learning_rate": 3.6693351484899986e-06,
      "loss": 0.2979268550872803,
      "num_input_tokens_seen": 554609008,
      "step": 17400,
      "train_runtime": 45647.5263,
      "train_tokens_per_second": 12149.815
    },
    {
      "epoch": 1.1718960409372474,
      "grad_norm": 1.0347704623956588,
      "learning_rate": 3.6667866078263937e-06,
      "loss": 0.29062800407409667,
      "num_input_tokens_seen": 554761824,
      "step": 17405,
      "train_runtime": 45660.128,
      "train_tokens_per_second": 12149.809
    },
    {
      "epoch": 1.1722326959332077,
      "grad_norm": 0.7487630343529679,
      "learning_rate": 3.664238439991663e-06,
      "loss": 0.3335871696472168,
      "num_input_tokens_seen": 554920280,
      "step": 17410,
      "train_runtime": 45673.585,
      "train_tokens_per_second": 12149.698
    },
    {
      "epoch": 1.1725693509291677,
      "grad_norm": 0.7446295261724114,
      "learning_rate": 3.6616906456983925e-06,
      "loss": 0.3145869731903076,
      "num_input_tokens_seen": 555082128,
      "step": 17415,
      "train_runtime": 45686.8715,
      "train_tokens_per_second": 12149.708
    },
    {
      "epoch": 1.172906005925128,
      "grad_norm": 0.7820036047613893,
      "learning_rate": 3.659143225659065e-06,
      "loss": 0.3277442455291748,
      "num_input_tokens_seen": 555241808,
      "step": 17420,
      "train_runtime": 45699.7653,
      "train_tokens_per_second": 12149.774
    },
    {
      "epoch": 1.173242660921088,
      "grad_norm": 0.7412034470381428,
      "learning_rate": 3.6565961805860595e-06,
      "loss": 0.30117168426513674,
      "num_input_tokens_seen": 555405648,
      "step": 17425,
      "train_runtime": 45712.1777,
      "train_tokens_per_second": 12150.059
    },
    {
      "epoch": 1.1735793159170482,
      "grad_norm": 0.7550550478779435,
      "learning_rate": 3.654049511191649e-06,
      "loss": 0.303936243057251,
      "num_input_tokens_seen": 555564480,
      "step": 17430,
      "train_runtime": 45724.7169,
      "train_tokens_per_second": 12150.201
    },
    {
      "epoch": 1.1739159709130083,
      "grad_norm": 0.8296758009245497,
      "learning_rate": 3.6515032181879995e-06,
      "loss": 0.31157379150390624,
      "num_input_tokens_seen": 555727656,
      "step": 17435,
      "train_runtime": 45737.5856,
      "train_tokens_per_second": 12150.35
    },
    {
      "epoch": 1.1742526259089685,
      "grad_norm": 0.696196065210496,
      "learning_rate": 3.6489573022871755e-06,
      "loss": 0.2630504846572876,
      "num_input_tokens_seen": 555887232,
      "step": 17440,
      "train_runtime": 45750.1805,
      "train_tokens_per_second": 12150.493
    },
    {
      "epoch": 1.1745892809049285,
      "grad_norm": 0.8091269771789984,
      "learning_rate": 3.6464117642011333e-06,
      "loss": 0.3271594524383545,
      "num_input_tokens_seen": 556041680,
      "step": 17445,
      "train_runtime": 45763.7881,
      "train_tokens_per_second": 12150.255
    },
    {
      "epoch": 1.1749259359008888,
      "grad_norm": 0.832838789748369,
      "learning_rate": 3.643866604641726e-06,
      "loss": 0.31891813278198244,
      "num_input_tokens_seen": 556204784,
      "step": 17450,
      "train_runtime": 45776.6362,
      "train_tokens_per_second": 12150.408
    },
    {
      "epoch": 1.1752625908968488,
      "grad_norm": 0.8227746836200858,
      "learning_rate": 3.641321824320697e-06,
      "loss": 0.2925240993499756,
      "num_input_tokens_seen": 556365240,
      "step": 17455,
      "train_runtime": 45789.6982,
      "train_tokens_per_second": 12150.446
    },
    {
      "epoch": 1.175599245892809,
      "grad_norm": 0.7526150328127748,
      "learning_rate": 3.6387774239496893e-06,
      "loss": 0.288327693939209,
      "num_input_tokens_seen": 556529000,
      "step": 17460,
      "train_runtime": 45802.7863,
      "train_tokens_per_second": 12150.549
    },
    {
      "epoch": 1.175935900888769,
      "grad_norm": 0.8424688072879664,
      "learning_rate": 3.6362334042402326e-06,
      "loss": 0.30585613250732424,
      "num_input_tokens_seen": 556689104,
      "step": 17465,
      "train_runtime": 45815.4212,
      "train_tokens_per_second": 12150.693
    },
    {
      "epoch": 1.1762725558847293,
      "grad_norm": 0.7377834377133992,
      "learning_rate": 3.6336897659037563e-06,
      "loss": 0.27790706157684325,
      "num_input_tokens_seen": 556838752,
      "step": 17470,
      "train_runtime": 45828.1594,
      "train_tokens_per_second": 12150.581
    },
    {
      "epoch": 1.1766092108806894,
      "grad_norm": 0.8412082374445791,
      "learning_rate": 3.6311465096515787e-06,
      "loss": 0.30019321441650393,
      "num_input_tokens_seen": 556998160,
      "step": 17475,
      "train_runtime": 45841.2996,
      "train_tokens_per_second": 12150.575
    },
    {
      "epoch": 1.1769458658766496,
      "grad_norm": 0.7592474295372953,
      "learning_rate": 3.6286036361949175e-06,
      "loss": 0.2577529430389404,
      "num_input_tokens_seen": 557159808,
      "step": 17480,
      "train_runtime": 45854.5165,
      "train_tokens_per_second": 12150.598
    },
    {
      "epoch": 1.1772825208726099,
      "grad_norm": 0.8392425164935884,
      "learning_rate": 3.6260611462448736e-06,
      "loss": 0.263155198097229,
      "num_input_tokens_seen": 557321104,
      "step": 17485,
      "train_runtime": 45867.4537,
      "train_tokens_per_second": 12150.688
    },
    {
      "epoch": 1.17761917586857,
      "grad_norm": 1.0412371782799181,
      "learning_rate": 3.623519040512451e-06,
      "loss": 0.33727331161499025,
      "num_input_tokens_seen": 557475784,
      "step": 17490,
      "train_runtime": 45881.5082,
      "train_tokens_per_second": 12150.337
    },
    {
      "epoch": 1.17795583086453,
      "grad_norm": 0.6429483406065157,
      "learning_rate": 3.6209773197085373e-06,
      "loss": 0.2657941818237305,
      "num_input_tokens_seen": 557637824,
      "step": 17495,
      "train_runtime": 45894.0851,
      "train_tokens_per_second": 12150.538
    },
    {
      "epoch": 1.1782924858604902,
      "grad_norm": 0.8639232594851377,
      "learning_rate": 3.6184359845439214e-06,
      "loss": 0.32529640197753906,
      "num_input_tokens_seen": 557800184,
      "step": 17500,
      "train_runtime": 45906.52,
      "train_tokens_per_second": 12150.783
    },
    {
      "epoch": 1.1786291408564504,
      "grad_norm": 0.837791471621092,
      "learning_rate": 3.615895035729275e-06,
      "loss": 0.32194719314575193,
      "num_input_tokens_seen": 557961128,
      "step": 17505,
      "train_runtime": 45919.4797,
      "train_tokens_per_second": 12150.859
    },
    {
      "epoch": 1.1789657958524105,
      "grad_norm": 0.8322176614875726,
      "learning_rate": 3.6133544739751703e-06,
      "loss": 0.309560489654541,
      "num_input_tokens_seen": 558116296,
      "step": 17510,
      "train_runtime": 45932.0657,
      "train_tokens_per_second": 12150.908
    },
    {
      "epoch": 1.1793024508483705,
      "grad_norm": 0.7747817401228362,
      "learning_rate": 3.6108142999920633e-06,
      "loss": 0.3072824239730835,
      "num_input_tokens_seen": 558274864,
      "step": 17515,
      "train_runtime": 45945.5924,
      "train_tokens_per_second": 12150.782
    },
    {
      "epoch": 1.1796391058443307,
      "grad_norm": 0.7576174588939192,
      "learning_rate": 3.608274514490308e-06,
      "loss": 0.29747934341430665,
      "num_input_tokens_seen": 558435600,
      "step": 17520,
      "train_runtime": 45958.3307,
      "train_tokens_per_second": 12150.911
    },
    {
      "epoch": 1.179975760840291,
      "grad_norm": 0.8583549250960663,
      "learning_rate": 3.605735118180146e-06,
      "loss": 0.2899029731750488,
      "num_input_tokens_seen": 558598120,
      "step": 17525,
      "train_runtime": 45971.6569,
      "train_tokens_per_second": 12150.924
    },
    {
      "epoch": 1.180312415836251,
      "grad_norm": 0.7668356206588756,
      "learning_rate": 3.6031961117717144e-06,
      "loss": 0.2827759265899658,
      "num_input_tokens_seen": 558755432,
      "step": 17530,
      "train_runtime": 45984.3794,
      "train_tokens_per_second": 12150.983
    },
    {
      "epoch": 1.180649070832211,
      "grad_norm": 0.8193913954843525,
      "learning_rate": 3.600657495975034e-06,
      "loss": 0.3110912322998047,
      "num_input_tokens_seen": 558917256,
      "step": 17535,
      "train_runtime": 45997.545,
      "train_tokens_per_second": 12151.024
    },
    {
      "epoch": 1.1809857258281713,
      "grad_norm": 0.8270517541795064,
      "learning_rate": 3.598119271500024e-06,
      "loss": 0.3022993326187134,
      "num_input_tokens_seen": 559076464,
      "step": 17540,
      "train_runtime": 46010.9356,
      "train_tokens_per_second": 12150.948
    },
    {
      "epoch": 1.1813223808241315,
      "grad_norm": 0.7708251471259241,
      "learning_rate": 3.5955814390564882e-06,
      "loss": 0.28093929290771485,
      "num_input_tokens_seen": 559238360,
      "step": 17545,
      "train_runtime": 46024.7442,
      "train_tokens_per_second": 12150.82
    },
    {
      "epoch": 1.1816590358200916,
      "grad_norm": 0.8171228914399099,
      "learning_rate": 3.5930439993541265e-06,
      "loss": 0.3191326141357422,
      "num_input_tokens_seen": 559395560,
      "step": 17550,
      "train_runtime": 46038.2235,
      "train_tokens_per_second": 12150.677
    },
    {
      "epoch": 1.1819956908160516,
      "grad_norm": 0.7404132249124264,
      "learning_rate": 3.590506953102522e-06,
      "loss": 0.3204774856567383,
      "num_input_tokens_seen": 559552784,
      "step": 17555,
      "train_runtime": 46051.4285,
      "train_tokens_per_second": 12150.606
    },
    {
      "epoch": 1.1823323458120119,
      "grad_norm": 0.7322016019221741,
      "learning_rate": 3.5879703010111554e-06,
      "loss": 0.3034974575042725,
      "num_input_tokens_seen": 559711520,
      "step": 17560,
      "train_runtime": 46064.0244,
      "train_tokens_per_second": 12150.73
    },
    {
      "epoch": 1.182669000807972,
      "grad_norm": 0.8278122469525622,
      "learning_rate": 3.5854340437893898e-06,
      "loss": 0.3104384183883667,
      "num_input_tokens_seen": 559875048,
      "step": 17565,
      "train_runtime": 46077.1019,
      "train_tokens_per_second": 12150.83
    },
    {
      "epoch": 1.1830056558039321,
      "grad_norm": 0.8298350414877184,
      "learning_rate": 3.5828981821464838e-06,
      "loss": 0.31849663257598876,
      "num_input_tokens_seen": 560030600,
      "step": 17570,
      "train_runtime": 46090.5857,
      "train_tokens_per_second": 12150.651
    },
    {
      "epoch": 1.1833423107998922,
      "grad_norm": 1.1995431217722516,
      "learning_rate": 3.580362716791582e-06,
      "loss": 0.32473204135894773,
      "num_input_tokens_seen": 560194440,
      "step": 17575,
      "train_runtime": 46103.0245,
      "train_tokens_per_second": 12150.926
    },
    {
      "epoch": 1.1836789657958524,
      "grad_norm": 0.7704162233167612,
      "learning_rate": 3.5778276484337205e-06,
      "loss": 0.32905077934265137,
      "num_input_tokens_seen": 560351848,
      "step": 17580,
      "train_runtime": 46116.3093,
      "train_tokens_per_second": 12150.839
    },
    {
      "epoch": 1.1840156207918127,
      "grad_norm": 0.681880325468424,
      "learning_rate": 3.575292977781821e-06,
      "loss": 0.2930294036865234,
      "num_input_tokens_seen": 560512088,
      "step": 17585,
      "train_runtime": 46129.3064,
      "train_tokens_per_second": 12150.889
    },
    {
      "epoch": 1.1843522757877727,
      "grad_norm": 0.8133042008948044,
      "learning_rate": 3.5727587055446976e-06,
      "loss": 0.30595741271972654,
      "num_input_tokens_seen": 560675928,
      "step": 17590,
      "train_runtime": 46141.7614,
      "train_tokens_per_second": 12151.16
    },
    {
      "epoch": 1.1846889307837327,
      "grad_norm": 0.7922630626149236,
      "learning_rate": 3.570224832431051e-06,
      "loss": 0.3272273540496826,
      "num_input_tokens_seen": 560838064,
      "step": 17595,
      "train_runtime": 46155.1007,
      "train_tokens_per_second": 12151.161
    },
    {
      "epoch": 1.185025585779693,
      "grad_norm": 0.7894285816275776,
      "learning_rate": 3.5676913591494715e-06,
      "loss": 0.2990771770477295,
      "num_input_tokens_seen": 560999904,
      "step": 17600,
      "train_runtime": 46168.1263,
      "train_tokens_per_second": 12151.238
    },
    {
      "epoch": 1.1853622407756532,
      "grad_norm": 0.7587555450446855,
      "learning_rate": 3.5651582864084345e-06,
      "loss": 0.3309017181396484,
      "num_input_tokens_seen": 561157912,
      "step": 17605,
      "train_runtime": 46181.0169,
      "train_tokens_per_second": 12151.268
    },
    {
      "epoch": 1.1856988957716132,
      "grad_norm": 0.8019450105065055,
      "learning_rate": 3.5626256149163096e-06,
      "loss": 0.3106407880783081,
      "num_input_tokens_seen": 561319480,
      "step": 17610,
      "train_runtime": 46194.9856,
      "train_tokens_per_second": 12151.091
    },
    {
      "epoch": 1.1860355507675733,
      "grad_norm": 0.8961913065247755,
      "learning_rate": 3.5600933453813464e-06,
      "loss": 0.31326682567596437,
      "num_input_tokens_seen": 561482128,
      "step": 17615,
      "train_runtime": 46207.7686,
      "train_tokens_per_second": 12151.25
    },
    {
      "epoch": 1.1863722057635335,
      "grad_norm": 0.7589926736910999,
      "learning_rate": 3.5575614785116886e-06,
      "loss": 0.29508700370788576,
      "num_input_tokens_seen": 561640568,
      "step": 17620,
      "train_runtime": 46220.3284,
      "train_tokens_per_second": 12151.376
    },
    {
      "epoch": 1.1867088607594938,
      "grad_norm": 0.86259647075367,
      "learning_rate": 3.5550300150153618e-06,
      "loss": 0.3041362285614014,
      "num_input_tokens_seen": 561798064,
      "step": 17625,
      "train_runtime": 46232.957,
      "train_tokens_per_second": 12151.463
    },
    {
      "epoch": 1.1870455157554538,
      "grad_norm": 0.8113768710035015,
      "learning_rate": 3.5524989556002863e-06,
      "loss": 0.28933746814727784,
      "num_input_tokens_seen": 561956256,
      "step": 17630,
      "train_runtime": 46246.0167,
      "train_tokens_per_second": 12151.452
    },
    {
      "epoch": 1.1873821707514138,
      "grad_norm": 0.6987533825329738,
      "learning_rate": 3.5499683009742604e-06,
      "loss": 0.2985436677932739,
      "num_input_tokens_seen": 562114096,
      "step": 17635,
      "train_runtime": 46259.2277,
      "train_tokens_per_second": 12151.394
    },
    {
      "epoch": 1.187718825747374,
      "grad_norm": 0.803484846503656,
      "learning_rate": 3.5474380518449757e-06,
      "loss": 0.28164746761322024,
      "num_input_tokens_seen": 562274840,
      "step": 17640,
      "train_runtime": 46272.2344,
      "train_tokens_per_second": 12151.452
    },
    {
      "epoch": 1.1880554807433343,
      "grad_norm": 0.8778671610181865,
      "learning_rate": 3.5449082089200072e-06,
      "loss": 0.2974414825439453,
      "num_input_tokens_seen": 562437568,
      "step": 17645,
      "train_runtime": 46285.6793,
      "train_tokens_per_second": 12151.438
    },
    {
      "epoch": 1.1883921357392944,
      "grad_norm": 0.7230273876040618,
      "learning_rate": 3.5423787729068206e-06,
      "loss": 0.3281216621398926,
      "num_input_tokens_seen": 562594104,
      "step": 17650,
      "train_runtime": 46298.7514,
      "train_tokens_per_second": 12151.388
    },
    {
      "epoch": 1.1887287907352544,
      "grad_norm": 0.7316016550924603,
      "learning_rate": 3.539849744512759e-06,
      "loss": 0.2960752248764038,
      "num_input_tokens_seen": 562753608,
      "step": 17655,
      "train_runtime": 46312.3612,
      "train_tokens_per_second": 12151.261
    },
    {
      "epoch": 1.1890654457312146,
      "grad_norm": 0.8044973566681579,
      "learning_rate": 3.5373211244450633e-06,
      "loss": 0.29789443016052247,
      "num_input_tokens_seen": 562914248,
      "step": 17660,
      "train_runtime": 46324.8001,
      "train_tokens_per_second": 12151.466
    },
    {
      "epoch": 1.189402100727175,
      "grad_norm": 0.746561599504987,
      "learning_rate": 3.5347929134108483e-06,
      "loss": 0.3126965045928955,
      "num_input_tokens_seen": 563072200,
      "step": 17665,
      "train_runtime": 46338.2293,
      "train_tokens_per_second": 12151.353
    },
    {
      "epoch": 1.189738755723135,
      "grad_norm": 1.1549879037937163,
      "learning_rate": 3.532265112117125e-06,
      "loss": 0.32831697463989257,
      "num_input_tokens_seen": 563235560,
      "step": 17670,
      "train_runtime": 46351.2429,
      "train_tokens_per_second": 12151.466
    },
    {
      "epoch": 1.190075410719095,
      "grad_norm": 0.7140035776257948,
      "learning_rate": 3.5297377212707805e-06,
      "loss": 0.26281650066375734,
      "num_input_tokens_seen": 563393096,
      "step": 17675,
      "train_runtime": 46364.2798,
      "train_tokens_per_second": 12151.447
    },
    {
      "epoch": 1.1904120657150552,
      "grad_norm": 0.8388881295552748,
      "learning_rate": 3.527210741578596e-06,
      "loss": 0.3121821641921997,
      "num_input_tokens_seen": 563553472,
      "step": 17680,
      "train_runtime": 46377.6951,
      "train_tokens_per_second": 12151.39
    },
    {
      "epoch": 1.1907487207110155,
      "grad_norm": 0.8281633960995729,
      "learning_rate": 3.5246841737472304e-06,
      "loss": 0.30962719917297366,
      "num_input_tokens_seen": 563704312,
      "step": 17685,
      "train_runtime": 46390.3052,
      "train_tokens_per_second": 12151.339
    },
    {
      "epoch": 1.1910853757069755,
      "grad_norm": 0.7820749382031627,
      "learning_rate": 3.522158018483231e-06,
      "loss": 0.28770737648010253,
      "num_input_tokens_seen": 563868128,
      "step": 17690,
      "train_runtime": 46403.4297,
      "train_tokens_per_second": 12151.432
    },
    {
      "epoch": 1.1914220307029355,
      "grad_norm": 0.8315654118797977,
      "learning_rate": 3.519632276493029e-06,
      "loss": 0.31392655372619627,
      "num_input_tokens_seen": 564028856,
      "step": 17695,
      "train_runtime": 46416.6713,
      "train_tokens_per_second": 12151.428
    },
    {
      "epoch": 1.1917586856988958,
      "grad_norm": 0.7614957628153395,
      "learning_rate": 3.5171069484829417e-06,
      "loss": 0.3177612781524658,
      "num_input_tokens_seen": 564187040,
      "step": 17700,
      "train_runtime": 46429.9188,
      "train_tokens_per_second": 12151.368
    },
    {
      "epoch": 1.192095340694856,
      "grad_norm": 0.7938877431521906,
      "learning_rate": 3.5145820351591663e-06,
      "loss": 0.2747988224029541,
      "num_input_tokens_seen": 564348432,
      "step": 17705,
      "train_runtime": 46442.9876,
      "train_tokens_per_second": 12151.424
    },
    {
      "epoch": 1.192431995690816,
      "grad_norm": 0.8341317773867487,
      "learning_rate": 3.51205753722779e-06,
      "loss": 0.3229677200317383,
      "num_input_tokens_seen": 564510192,
      "step": 17710,
      "train_runtime": 46455.546,
      "train_tokens_per_second": 12151.621
    },
    {
      "epoch": 1.1927686506867763,
      "grad_norm": 0.840907055895606,
      "learning_rate": 3.5095334553947757e-06,
      "loss": 0.3147230625152588,
      "num_input_tokens_seen": 564671136,
      "step": 17715,
      "train_runtime": 46468.011,
      "train_tokens_per_second": 12151.825
    },
    {
      "epoch": 1.1931053056827363,
      "grad_norm": 0.730440620747168,
      "learning_rate": 3.50700979036598e-06,
      "loss": 0.2902683258056641,
      "num_input_tokens_seen": 564830632,
      "step": 17720,
      "train_runtime": 46481.1116,
      "train_tokens_per_second": 12151.831
    },
    {
      "epoch": 1.1934419606786966,
      "grad_norm": 0.8027784304679113,
      "learning_rate": 3.5044865428471343e-06,
      "loss": 0.29388995170593263,
      "num_input_tokens_seen": 564986928,
      "step": 17725,
      "train_runtime": 46495.073,
      "train_tokens_per_second": 12151.544
    },
    {
      "epoch": 1.1937786156746566,
      "grad_norm": 0.6798330092423434,
      "learning_rate": 3.50196371354386e-06,
      "loss": 0.3117683410644531,
      "num_input_tokens_seen": 565148960,
      "step": 17730,
      "train_runtime": 46509.3614,
      "train_tokens_per_second": 12151.295
    },
    {
      "epoch": 1.1941152706706168,
      "grad_norm": 0.9004462888111853,
      "learning_rate": 3.499441303161655e-06,
      "loss": 0.3353108882904053,
      "num_input_tokens_seen": 565310088,
      "step": 17735,
      "train_runtime": 46522.2541,
      "train_tokens_per_second": 12151.391
    },
    {
      "epoch": 1.1944519256665769,
      "grad_norm": 0.8666486029909161,
      "learning_rate": 3.4969193124059042e-06,
      "loss": 0.29062471389770506,
      "num_input_tokens_seen": 565470024,
      "step": 17740,
      "train_runtime": 46534.9347,
      "train_tokens_per_second": 12151.516
    },
    {
      "epoch": 1.1947885806625371,
      "grad_norm": 0.7900329130547086,
      "learning_rate": 3.4943977419818748e-06,
      "loss": 0.3008563995361328,
      "num_input_tokens_seen": 565632736,
      "step": 17745,
      "train_runtime": 46547.6746,
      "train_tokens_per_second": 12151.686
    },
    {
      "epoch": 1.1951252356584972,
      "grad_norm": 0.8228020002776071,
      "learning_rate": 3.4918765925947173e-06,
      "loss": 0.3041223526000977,
      "num_input_tokens_seen": 565795136,
      "step": 17750,
      "train_runtime": 46561.0547,
      "train_tokens_per_second": 12151.682
    },
    {
      "epoch": 1.1954618906544574,
      "grad_norm": 0.8968691579526893,
      "learning_rate": 3.4893558649494595e-06,
      "loss": 0.3254541873931885,
      "num_input_tokens_seen": 565958912,
      "step": 17755,
      "train_runtime": 46574.0952,
      "train_tokens_per_second": 12151.796
    },
    {
      "epoch": 1.1957985456504174,
      "grad_norm": 1.1354276920818505,
      "learning_rate": 3.4868355597510194e-06,
      "loss": 0.29193673133850095,
      "num_input_tokens_seen": 566116432,
      "step": 17760,
      "train_runtime": 46586.983,
      "train_tokens_per_second": 12151.816
    },
    {
      "epoch": 1.1961352006463777,
      "grad_norm": 0.814188322586844,
      "learning_rate": 3.4843156777041887e-06,
      "loss": 0.28440284729003906,
      "num_input_tokens_seen": 566277480,
      "step": 17765,
      "train_runtime": 46600.5626,
      "train_tokens_per_second": 12151.731
    },
    {
      "epoch": 1.1964718556423377,
      "grad_norm": 0.8795942294365611,
      "learning_rate": 3.4817962195136457e-06,
      "loss": 0.29191007614135744,
      "num_input_tokens_seen": 566433232,
      "step": 17770,
      "train_runtime": 46613.7306,
      "train_tokens_per_second": 12151.639
    },
    {
      "epoch": 1.196808510638298,
      "grad_norm": 0.913075876064718,
      "learning_rate": 3.479277185883947e-06,
      "loss": 0.31943354606628416,
      "num_input_tokens_seen": 566591584,
      "step": 17775,
      "train_runtime": 46626.9579,
      "train_tokens_per_second": 12151.588
    },
    {
      "epoch": 1.197145165634258,
      "grad_norm": 0.7402657189002572,
      "learning_rate": 3.476758577519537e-06,
      "loss": 0.2841529130935669,
      "num_input_tokens_seen": 566752792,
      "step": 17780,
      "train_runtime": 46640.2686,
      "train_tokens_per_second": 12151.577
    },
    {
      "epoch": 1.1974818206302182,
      "grad_norm": 0.8704387508559097,
      "learning_rate": 3.4742403951247318e-06,
      "loss": 0.30609564781188964,
      "num_input_tokens_seen": 566910664,
      "step": 17785,
      "train_runtime": 46652.8446,
      "train_tokens_per_second": 12151.685
    },
    {
      "epoch": 1.1978184756261783,
      "grad_norm": 0.827243114464678,
      "learning_rate": 3.471722639403735e-06,
      "loss": 0.3014299154281616,
      "num_input_tokens_seen": 567068976,
      "step": 17790,
      "train_runtime": 46665.2553,
      "train_tokens_per_second": 12151.846
    },
    {
      "epoch": 1.1981551306221385,
      "grad_norm": 0.7837802792959181,
      "learning_rate": 3.4692053110606277e-06,
      "loss": 0.3090204238891602,
      "num_input_tokens_seen": 567221808,
      "step": 17795,
      "train_runtime": 46678.2929,
      "train_tokens_per_second": 12151.726
    },
    {
      "epoch": 1.1984917856180985,
      "grad_norm": 0.6950229254256485,
      "learning_rate": 3.4666884107993766e-06,
      "loss": 0.2893324136734009,
      "num_input_tokens_seen": 567380120,
      "step": 17800,
      "train_runtime": 46691.7299,
      "train_tokens_per_second": 12151.619
    },
    {
      "epoch": 1.1988284406140588,
      "grad_norm": 0.8179883904951026,
      "learning_rate": 3.464171939323819e-06,
      "loss": 0.29560964107513427,
      "num_input_tokens_seen": 567543280,
      "step": 17805,
      "train_runtime": 46705.3122,
      "train_tokens_per_second": 12151.579
    },
    {
      "epoch": 1.1991650956100188,
      "grad_norm": 0.7855535497025751,
      "learning_rate": 3.4616558973376848e-06,
      "loss": 0.2991500377655029,
      "num_input_tokens_seen": 567697496,
      "step": 17810,
      "train_runtime": 46718.5271,
      "train_tokens_per_second": 12151.443
    },
    {
      "epoch": 1.199501750605979,
      "grad_norm": 0.8381663835542378,
      "learning_rate": 3.4591402855445697e-06,
      "loss": 0.3060968637466431,
      "num_input_tokens_seen": 567855672,
      "step": 17815,
      "train_runtime": 46731.0905,
      "train_tokens_per_second": 12151.56
    },
    {
      "epoch": 1.199838405601939,
      "grad_norm": 0.7219875989091953,
      "learning_rate": 3.4566251046479625e-06,
      "loss": 0.3001152515411377,
      "num_input_tokens_seen": 568018328,
      "step": 17820,
      "train_runtime": 46743.903,
      "train_tokens_per_second": 12151.709
    },
    {
      "epoch": 1.2001750605978994,
      "grad_norm": 0.8119207478165796,
      "learning_rate": 3.454110355351221e-06,
      "loss": 0.30767116546630857,
      "num_input_tokens_seen": 568177128,
      "step": 17825,
      "train_runtime": 46756.4492,
      "train_tokens_per_second": 12151.845
    },
    {
      "epoch": 1.2005117155938594,
      "grad_norm": 0.8102058933334578,
      "learning_rate": 3.451596038357593e-06,
      "loss": 0.3000184535980225,
      "num_input_tokens_seen": 568339944,
      "step": 17830,
      "train_runtime": 46769.9071,
      "train_tokens_per_second": 12151.83
    },
    {
      "epoch": 1.2008483705898196,
      "grad_norm": 0.8011544968228623,
      "learning_rate": 3.4490821543701924e-06,
      "loss": 0.30973286628723146,
      "num_input_tokens_seen": 568500680,
      "step": 17835,
      "train_runtime": 46782.7736,
      "train_tokens_per_second": 12151.923
    },
    {
      "epoch": 1.2011850255857797,
      "grad_norm": 0.8044756743367137,
      "learning_rate": 3.4465687040920227e-06,
      "loss": 0.29834637641906736,
      "num_input_tokens_seen": 568662248,
      "step": 17840,
      "train_runtime": 46795.9797,
      "train_tokens_per_second": 12151.947
    },
    {
      "epoch": 1.20152168058174,
      "grad_norm": 0.7782002230049677,
      "learning_rate": 3.44405568822596e-06,
      "loss": 0.27195043563842775,
      "num_input_tokens_seen": 568822328,
      "step": 17845,
      "train_runtime": 46809.5137,
      "train_tokens_per_second": 12151.853
    },
    {
      "epoch": 1.2018583355777,
      "grad_norm": 0.944995387308152,
      "learning_rate": 3.441543107474765e-06,
      "loss": 0.30285005569458007,
      "num_input_tokens_seen": 568976912,
      "step": 17850,
      "train_runtime": 46822.9454,
      "train_tokens_per_second": 12151.669
    },
    {
      "epoch": 1.2021949905736602,
      "grad_norm": 0.8316285861707877,
      "learning_rate": 3.439030962541069e-06,
      "loss": 0.29416570663452146,
      "num_input_tokens_seen": 569137424,
      "step": 17855,
      "train_runtime": 46836.7547,
      "train_tokens_per_second": 12151.513
    },
    {
      "epoch": 1.2025316455696202,
      "grad_norm": 0.7485861418886484,
      "learning_rate": 3.4365192541273885e-06,
      "loss": 0.28441247940063474,
      "num_input_tokens_seen": 569294928,
      "step": 17860,
      "train_runtime": 46850.2624,
      "train_tokens_per_second": 12151.371
    },
    {
      "epoch": 1.2028683005655805,
      "grad_norm": 0.7711708929097452,
      "learning_rate": 3.4340079829361108e-06,
      "loss": 0.31755905151367186,
      "num_input_tokens_seen": 569446520,
      "step": 17865,
      "train_runtime": 46863.9584,
      "train_tokens_per_second": 12151.055
    },
    {
      "epoch": 1.2032049555615405,
      "grad_norm": 0.8063765257375277,
      "learning_rate": 3.4314971496695078e-06,
      "loss": 0.32381525039672854,
      "num_input_tokens_seen": 569607400,
      "step": 17870,
      "train_runtime": 46876.5212,
      "train_tokens_per_second": 12151.23
    },
    {
      "epoch": 1.2035416105575008,
      "grad_norm": 0.6929707481029308,
      "learning_rate": 3.428986755029724e-06,
      "loss": 0.2725623369216919,
      "num_input_tokens_seen": 569771240,
      "step": 17875,
      "train_runtime": 46888.9433,
      "train_tokens_per_second": 12151.505
    },
    {
      "epoch": 1.2038782655534608,
      "grad_norm": 0.7831387459888866,
      "learning_rate": 3.4264767997187874e-06,
      "loss": 0.2895507335662842,
      "num_input_tokens_seen": 569934840,
      "step": 17880,
      "train_runtime": 46901.9709,
      "train_tokens_per_second": 12151.618
    },
    {
      "epoch": 1.204214920549421,
      "grad_norm": 0.8027126283485034,
      "learning_rate": 3.423967284438594e-06,
      "loss": 0.3056935787200928,
      "num_input_tokens_seen": 570088256,
      "step": 17885,
      "train_runtime": 46914.6059,
      "train_tokens_per_second": 12151.616
    },
    {
      "epoch": 1.204551575545381,
      "grad_norm": 0.7856246685138956,
      "learning_rate": 3.421458209890925e-06,
      "loss": 0.30243093967437745,
      "num_input_tokens_seen": 570251736,
      "step": 17890,
      "train_runtime": 46928.2977,
      "train_tokens_per_second": 12151.554
    },
    {
      "epoch": 1.2048882305413413,
      "grad_norm": 0.7756182820173754,
      "learning_rate": 3.418949576777433e-06,
      "loss": 0.30045175552368164,
      "num_input_tokens_seen": 570415368,
      "step": 17895,
      "train_runtime": 46940.7567,
      "train_tokens_per_second": 12151.814
    },
    {
      "epoch": 1.2052248855373013,
      "grad_norm": 0.8689168690968989,
      "learning_rate": 3.4164413857996513e-06,
      "loss": 0.2956207752227783,
      "num_input_tokens_seen": 570571680,
      "step": 17900,
      "train_runtime": 46954.1829,
      "train_tokens_per_second": 12151.669
    },
    {
      "epoch": 1.2055615405332616,
      "grad_norm": 0.7624302233640684,
      "learning_rate": 3.4139336376589853e-06,
      "loss": 0.2786736011505127,
      "num_input_tokens_seen": 570734048,
      "step": 17905,
      "train_runtime": 46967.2064,
      "train_tokens_per_second": 12151.756
    },
    {
      "epoch": 1.2058981955292216,
      "grad_norm": 0.9260894254062003,
      "learning_rate": 3.411426333056722e-06,
      "loss": 0.3092201709747314,
      "num_input_tokens_seen": 570884856,
      "step": 17910,
      "train_runtime": 46980.6194,
      "train_tokens_per_second": 12151.497
    },
    {
      "epoch": 1.2062348505251819,
      "grad_norm": 0.784201854685884,
      "learning_rate": 3.408919472694017e-06,
      "loss": 0.3036609172821045,
      "num_input_tokens_seen": 571047024,
      "step": 17915,
      "train_runtime": 46993.1975,
      "train_tokens_per_second": 12151.695
    },
    {
      "epoch": 1.206571505521142,
      "grad_norm": 0.7793430943613993,
      "learning_rate": 3.406413057271909e-06,
      "loss": 0.2953254222869873,
      "num_input_tokens_seen": 571204640,
      "step": 17920,
      "train_runtime": 47006.3163,
      "train_tokens_per_second": 12151.657
    },
    {
      "epoch": 1.2069081605171021,
      "grad_norm": 0.8058779774258039,
      "learning_rate": 3.403907087491306e-06,
      "loss": 0.29497694969177246,
      "num_input_tokens_seen": 571362648,
      "step": 17925,
      "train_runtime": 47019.4006,
      "train_tokens_per_second": 12151.636
    },
    {
      "epoch": 1.2072448155130622,
      "grad_norm": 0.8245814192877534,
      "learning_rate": 3.4014015640529996e-06,
      "loss": 0.31172614097595214,
      "num_input_tokens_seen": 571517856,
      "step": 17930,
      "train_runtime": 47032.5861,
      "train_tokens_per_second": 12151.529
    },
    {
      "epoch": 1.2075814705090224,
      "grad_norm": 0.7959023521273502,
      "learning_rate": 3.398896487657646e-06,
      "loss": 0.3130847930908203,
      "num_input_tokens_seen": 571677672,
      "step": 17935,
      "train_runtime": 47045.6381,
      "train_tokens_per_second": 12151.555
    },
    {
      "epoch": 1.2079181255049825,
      "grad_norm": 0.8189190412137193,
      "learning_rate": 3.396391859005785e-06,
      "loss": 0.2894524812698364,
      "num_input_tokens_seen": 571836872,
      "step": 17940,
      "train_runtime": 47059.0874,
      "train_tokens_per_second": 12151.465
    },
    {
      "epoch": 1.2082547805009427,
      "grad_norm": 0.7009741826852409,
      "learning_rate": 3.393887678797826e-06,
      "loss": 0.2940798044204712,
      "num_input_tokens_seen": 571996376,
      "step": 17945,
      "train_runtime": 47072.7064,
      "train_tokens_per_second": 12151.338
    },
    {
      "epoch": 1.2085914354969027,
      "grad_norm": 0.7596536273492601,
      "learning_rate": 3.3913839477340575e-06,
      "loss": 0.312032413482666,
      "num_input_tokens_seen": 572153688,
      "step": 17950,
      "train_runtime": 47086.3945,
      "train_tokens_per_second": 12151.147
    },
    {
      "epoch": 1.208928090492863,
      "grad_norm": 0.8222355649043206,
      "learning_rate": 3.388880666514637e-06,
      "loss": 0.2860033273696899,
      "num_input_tokens_seen": 572313904,
      "step": 17955,
      "train_runtime": 47099.4371,
      "train_tokens_per_second": 12151.184
    },
    {
      "epoch": 1.209264745488823,
      "grad_norm": 0.7602763105292438,
      "learning_rate": 3.3863778358396033e-06,
      "loss": 0.309263277053833,
      "num_input_tokens_seen": 572471736,
      "step": 17960,
      "train_runtime": 47111.9042,
      "train_tokens_per_second": 12151.318
    },
    {
      "epoch": 1.2096014004847833,
      "grad_norm": 0.7889687596431745,
      "learning_rate": 3.3838754564088592e-06,
      "loss": 0.28440108299255373,
      "num_input_tokens_seen": 572633392,
      "step": 17965,
      "train_runtime": 47124.7791,
      "train_tokens_per_second": 12151.429
    },
    {
      "epoch": 1.2099380554807433,
      "grad_norm": 0.7146790274586586,
      "learning_rate": 3.3813735289221923e-06,
      "loss": 0.2906285285949707,
      "num_input_tokens_seen": 572791496,
      "step": 17970,
      "train_runtime": 47137.6727,
      "train_tokens_per_second": 12151.459
    },
    {
      "epoch": 1.2102747104767035,
      "grad_norm": 0.6689714295059437,
      "learning_rate": 3.378872054079255e-06,
      "loss": 0.27923150062561036,
      "num_input_tokens_seen": 572947776,
      "step": 17975,
      "train_runtime": 47150.0933,
      "train_tokens_per_second": 12151.572
    },
    {
      "epoch": 1.2106113654726636,
      "grad_norm": 0.8010649975368276,
      "learning_rate": 3.3763710325795784e-06,
      "loss": 0.3243656396865845,
      "num_input_tokens_seen": 573104936,
      "step": 17980,
      "train_runtime": 47165.1796,
      "train_tokens_per_second": 12151.018
    },
    {
      "epoch": 1.2109480204686238,
      "grad_norm": 0.7690972727209558,
      "learning_rate": 3.3738704651225673e-06,
      "loss": 0.30284533500671384,
      "num_input_tokens_seen": 573267344,
      "step": 17985,
      "train_runtime": 47177.6198,
      "train_tokens_per_second": 12151.256
    },
    {
      "epoch": 1.2112846754645838,
      "grad_norm": 0.7801078488543157,
      "learning_rate": 3.3713703524074936e-06,
      "loss": 0.2852578639984131,
      "num_input_tokens_seen": 573423648,
      "step": 17990,
      "train_runtime": 47190.8304,
      "train_tokens_per_second": 12151.167
    },
    {
      "epoch": 1.211621330460544,
      "grad_norm": 0.8224166749666002,
      "learning_rate": 3.3688706951335082e-06,
      "loss": 0.31235241889953613,
      "num_input_tokens_seen": 573574432,
      "step": 17995,
      "train_runtime": 47203.8552,
      "train_tokens_per_second": 12151.008
    },
    {
      "epoch": 1.2119579854565041,
      "grad_norm": 0.7680398291413184,
      "learning_rate": 3.3663714939996305e-06,
      "loss": 0.31341843605041503,
      "num_input_tokens_seen": 573738272,
      "step": 18000,
      "train_runtime": 47216.2895,
      "train_tokens_per_second": 12151.278
    },
    {
      "epoch": 1.2122946404524644,
      "grad_norm": 0.8254891894824747,
      "learning_rate": 3.3638727497047575e-06,
      "loss": 0.2909956455230713,
      "num_input_tokens_seen": 573890192,
      "step": 18005,
      "train_runtime": 47229.4859,
      "train_tokens_per_second": 12151.1
    },
    {
      "epoch": 1.2126312954484244,
      "grad_norm": 0.7554840919740514,
      "learning_rate": 3.3613744629476507e-06,
      "loss": 0.2978204250335693,
      "num_input_tokens_seen": 574049896,
      "step": 18010,
      "train_runtime": 47243.3519,
      "train_tokens_per_second": 12150.914
    },
    {
      "epoch": 1.2129679504443847,
      "grad_norm": 0.8463464374283453,
      "learning_rate": 3.358876634426952e-06,
      "loss": 0.27528061866760256,
      "num_input_tokens_seen": 574208696,
      "step": 18015,
      "train_runtime": 47255.8509,
      "train_tokens_per_second": 12151.06
    },
    {
      "epoch": 1.2133046054403447,
      "grad_norm": 0.7982674687411516,
      "learning_rate": 3.3563792648411676e-06,
      "loss": 0.3229090929031372,
      "num_input_tokens_seen": 574363352,
      "step": 18020,
      "train_runtime": 47269.1763,
      "train_tokens_per_second": 12150.907
    },
    {
      "epoch": 1.213641260436305,
      "grad_norm": 0.7678606915989108,
      "learning_rate": 3.353882354888681e-06,
      "loss": 0.2996983528137207,
      "num_input_tokens_seen": 574522296,
      "step": 18025,
      "train_runtime": 47282.4558,
      "train_tokens_per_second": 12150.856
    },
    {
      "epoch": 1.213977915432265,
      "grad_norm": 0.8393848577094165,
      "learning_rate": 3.3513859052677444e-06,
      "loss": 0.2800508975982666,
      "num_input_tokens_seen": 574675704,
      "step": 18030,
      "train_runtime": 47295.7111,
      "train_tokens_per_second": 12150.694
    },
    {
      "epoch": 1.2143145704282252,
      "grad_norm": 0.8070502536245979,
      "learning_rate": 3.3488899166764842e-06,
      "loss": 0.31464133262634275,
      "num_input_tokens_seen": 574838592,
      "step": 18035,
      "train_runtime": 47310.0459,
      "train_tokens_per_second": 12150.455
    },
    {
      "epoch": 1.2146512254241852,
      "grad_norm": 0.8024564283004648,
      "learning_rate": 3.346394389812892e-06,
      "loss": 0.2928361415863037,
      "num_input_tokens_seen": 574994176,
      "step": 18040,
      "train_runtime": 47322.4538,
      "train_tokens_per_second": 12150.557
    },
    {
      "epoch": 1.2149878804201455,
      "grad_norm": 0.7612487429733211,
      "learning_rate": 3.3438993253748365e-06,
      "loss": 0.29835045337677,
      "num_input_tokens_seen": 575152168,
      "step": 18045,
      "train_runtime": 47335.0133,
      "train_tokens_per_second": 12150.671
    },
    {
      "epoch": 1.2153245354161055,
      "grad_norm": 0.7351378372598166,
      "learning_rate": 3.3414047240600526e-06,
      "loss": 0.30321204662323,
      "num_input_tokens_seen": 575310856,
      "step": 18050,
      "train_runtime": 47347.5513,
      "train_tokens_per_second": 12150.805
    },
    {
      "epoch": 1.2156611904120658,
      "grad_norm": 0.8620028583674225,
      "learning_rate": 3.338910586566151e-06,
      "loss": 0.2911694526672363,
      "num_input_tokens_seen": 575462672,
      "step": 18055,
      "train_runtime": 47360.0916,
      "train_tokens_per_second": 12150.793
    },
    {
      "epoch": 1.2159978454080258,
      "grad_norm": 0.7853325078595972,
      "learning_rate": 3.336416913590605e-06,
      "loss": 0.3024179935455322,
      "num_input_tokens_seen": 575620280,
      "step": 18060,
      "train_runtime": 47374.2764,
      "train_tokens_per_second": 12150.482
    },
    {
      "epoch": 1.216334500403986,
      "grad_norm": 0.7450515322026561,
      "learning_rate": 3.333923705830766e-06,
      "loss": 0.2946027755737305,
      "num_input_tokens_seen": 575782472,
      "step": 18065,
      "train_runtime": 47387.4884,
      "train_tokens_per_second": 12150.517
    },
    {
      "epoch": 1.216671155399946,
      "grad_norm": 1.0433960135595954,
      "learning_rate": 3.3314309639838484e-06,
      "loss": 0.3284256935119629,
      "num_input_tokens_seen": 575943088,
      "step": 18070,
      "train_runtime": 47400.5762,
      "train_tokens_per_second": 12150.55
    },
    {
      "epoch": 1.2170078103959063,
      "grad_norm": 0.8095019427286518,
      "learning_rate": 3.328938688746942e-06,
      "loss": 0.3049996614456177,
      "num_input_tokens_seen": 576105296,
      "step": 18075,
      "train_runtime": 47413.1637,
      "train_tokens_per_second": 12150.746
    },
    {
      "epoch": 1.2173444653918664,
      "grad_norm": 0.8464215296908669,
      "learning_rate": 3.3264468808170012e-06,
      "loss": 0.35430088043212893,
      "num_input_tokens_seen": 576267016,
      "step": 18080,
      "train_runtime": 47426.6597,
      "train_tokens_per_second": 12150.698
    },
    {
      "epoch": 1.2176811203878266,
      "grad_norm": 0.8681948685878391,
      "learning_rate": 3.3239555408908562e-06,
      "loss": 0.31813869476318357,
      "num_input_tokens_seen": 576417944,
      "step": 18085,
      "train_runtime": 47440.1175,
      "train_tokens_per_second": 12150.432
    },
    {
      "epoch": 1.2180177753837866,
      "grad_norm": 0.7806745614289626,
      "learning_rate": 3.3214646696651974e-06,
      "loss": 0.28757009506225584,
      "num_input_tokens_seen": 576576392,
      "step": 18090,
      "train_runtime": 47452.9539,
      "train_tokens_per_second": 12150.485
    },
    {
      "epoch": 1.2183544303797469,
      "grad_norm": 0.8081053741056332,
      "learning_rate": 3.3189742678365923e-06,
      "loss": 0.3226039886474609,
      "num_input_tokens_seen": 576732864,
      "step": 18095,
      "train_runtime": 47465.6489,
      "train_tokens_per_second": 12150.532
    },
    {
      "epoch": 1.218691085375707,
      "grad_norm": 0.7750560531041005,
      "learning_rate": 3.316484336101472e-06,
      "loss": 0.27719364166259763,
      "num_input_tokens_seen": 576889880,
      "step": 18100,
      "train_runtime": 47479.7963,
      "train_tokens_per_second": 12150.218
    },
    {
      "epoch": 1.2190277403716672,
      "grad_norm": 0.7588890225825011,
      "learning_rate": 3.313994875156141e-06,
      "loss": 0.3108834743499756,
      "num_input_tokens_seen": 577047536,
      "step": 18105,
      "train_runtime": 47492.6661,
      "train_tokens_per_second": 12150.245
    },
    {
      "epoch": 1.2193643953676272,
      "grad_norm": 0.77547839594807,
      "learning_rate": 3.3115058856967643e-06,
      "loss": 0.31075754165649416,
      "num_input_tokens_seen": 577209800,
      "step": 18110,
      "train_runtime": 47505.9234,
      "train_tokens_per_second": 12150.27
    },
    {
      "epoch": 1.2197010503635874,
      "grad_norm": 0.8586114694947962,
      "learning_rate": 3.3090173684193845e-06,
      "loss": 0.2806204080581665,
      "num_input_tokens_seen": 577370640,
      "step": 18115,
      "train_runtime": 47519.5759,
      "train_tokens_per_second": 12150.164
    },
    {
      "epoch": 1.2200377053595475,
      "grad_norm": 0.7446612933946813,
      "learning_rate": 3.3065293240199037e-06,
      "loss": 0.3068239688873291,
      "num_input_tokens_seen": 577532856,
      "step": 18120,
      "train_runtime": 47532.1594,
      "train_tokens_per_second": 12150.36
    },
    {
      "epoch": 1.2203743603555077,
      "grad_norm": 0.774105711261429,
      "learning_rate": 3.3040417531940982e-06,
      "loss": 0.31845722198486326,
      "num_input_tokens_seen": 577696280,
      "step": 18125,
      "train_runtime": 47545.1348,
      "train_tokens_per_second": 12150.481
    },
    {
      "epoch": 1.2207110153514678,
      "grad_norm": 0.80099919341523,
      "learning_rate": 3.301554656637607e-06,
      "loss": 0.3007077693939209,
      "num_input_tokens_seen": 577853792,
      "step": 18130,
      "train_runtime": 47558.4724,
      "train_tokens_per_second": 12150.386
    },
    {
      "epoch": 1.221047670347428,
      "grad_norm": 0.6989193431532472,
      "learning_rate": 3.2990680350459427e-06,
      "loss": 0.29122214317321776,
      "num_input_tokens_seen": 578009456,
      "step": 18135,
      "train_runtime": 47571.5838,
      "train_tokens_per_second": 12150.309
    },
    {
      "epoch": 1.221384325343388,
      "grad_norm": 0.7796028735434357,
      "learning_rate": 3.2965818891144763e-06,
      "loss": 0.3225094795227051,
      "num_input_tokens_seen": 578169976,
      "step": 18140,
      "train_runtime": 47585.2889,
      "train_tokens_per_second": 12150.183
    },
    {
      "epoch": 1.2217209803393483,
      "grad_norm": 0.782376539132792,
      "learning_rate": 3.294096219538454e-06,
      "loss": 0.3024899482727051,
      "num_input_tokens_seen": 578333632,
      "step": 18145,
      "train_runtime": 47598.3226,
      "train_tokens_per_second": 12150.294
    },
    {
      "epoch": 1.2220576353353083,
      "grad_norm": 0.8105950417204018,
      "learning_rate": 3.291611027012983e-06,
      "loss": 0.3225839138031006,
      "num_input_tokens_seen": 578496168,
      "step": 18150,
      "train_runtime": 47611.5811,
      "train_tokens_per_second": 12150.325
    },
    {
      "epoch": 1.2223942903312686,
      "grad_norm": 0.854074249453237,
      "learning_rate": 3.289126312233043e-06,
      "loss": 0.2955545663833618,
      "num_input_tokens_seen": 578660008,
      "step": 18155,
      "train_runtime": 47624.008,
      "train_tokens_per_second": 12150.594
    },
    {
      "epoch": 1.2227309453272286,
      "grad_norm": 0.7572560018259764,
      "learning_rate": 3.2866420758934714e-06,
      "loss": 0.30736656188964845,
      "num_input_tokens_seen": 578818600,
      "step": 18160,
      "train_runtime": 47636.7322,
      "train_tokens_per_second": 12150.678
    },
    {
      "epoch": 1.2230676003231888,
      "grad_norm": 0.8009414870758533,
      "learning_rate": 3.2841583186889826e-06,
      "loss": 0.298984432220459,
      "num_input_tokens_seen": 578978624,
      "step": 18165,
      "train_runtime": 47649.3165,
      "train_tokens_per_second": 12150.827
    },
    {
      "epoch": 1.2234042553191489,
      "grad_norm": 0.8144106832655196,
      "learning_rate": 3.281675041314146e-06,
      "loss": 0.27940974235534666,
      "num_input_tokens_seen": 579139136,
      "step": 18170,
      "train_runtime": 47662.4381,
      "train_tokens_per_second": 12150.85
    },
    {
      "epoch": 1.2237409103151091,
      "grad_norm": 0.7046220285854843,
      "learning_rate": 3.279192244463406e-06,
      "loss": 0.29690823554992674,
      "num_input_tokens_seen": 579299240,
      "step": 18175,
      "train_runtime": 47675.747,
      "train_tokens_per_second": 12150.816
    },
    {
      "epoch": 1.2240775653110691,
      "grad_norm": 0.800948066528882,
      "learning_rate": 3.2767099288310667e-06,
      "loss": 0.2955416202545166,
      "num_input_tokens_seen": 579460200,
      "step": 18180,
      "train_runtime": 47688.1834,
      "train_tokens_per_second": 12151.023
    },
    {
      "epoch": 1.2244142203070294,
      "grad_norm": 0.7193330744763012,
      "learning_rate": 3.2742280951113025e-06,
      "loss": 0.29018535614013674,
      "num_input_tokens_seen": 579622272,
      "step": 18185,
      "train_runtime": 47701.3488,
      "train_tokens_per_second": 12151.067
    },
    {
      "epoch": 1.2247508753029894,
      "grad_norm": 0.9277726836180876,
      "learning_rate": 3.2717467439981464e-06,
      "loss": 0.31015563011169434,
      "num_input_tokens_seen": 579776096,
      "step": 18190,
      "train_runtime": 47713.7776,
      "train_tokens_per_second": 12151.125
    },
    {
      "epoch": 1.2250875302989497,
      "grad_norm": 0.7707101647023258,
      "learning_rate": 3.2692658761855033e-06,
      "loss": 0.2692745208740234,
      "num_input_tokens_seen": 579929976,
      "step": 18195,
      "train_runtime": 47727.848,
      "train_tokens_per_second": 12150.767
    },
    {
      "epoch": 1.2254241852949097,
      "grad_norm": 0.7848905594539523,
      "learning_rate": 3.266785492367138e-06,
      "loss": 0.27900674343109133,
      "num_input_tokens_seen": 580089776,
      "step": 18200,
      "train_runtime": 47741.2186,
      "train_tokens_per_second": 12150.712
    },
    {
      "epoch": 1.22576084029087,
      "grad_norm": 0.7980871757386302,
      "learning_rate": 3.2643055932366852e-06,
      "loss": 0.2756039142608643,
      "num_input_tokens_seen": 580246832,
      "step": 18205,
      "train_runtime": 47754.4577,
      "train_tokens_per_second": 12150.632
    },
    {
      "epoch": 1.22609749528683,
      "grad_norm": 0.7641353925385463,
      "learning_rate": 3.2618261794876365e-06,
      "loss": 0.28605146408081056,
      "num_input_tokens_seen": 580409496,
      "step": 18210,
      "train_runtime": 47767.7942,
      "train_tokens_per_second": 12150.645
    },
    {
      "epoch": 1.2264341502827902,
      "grad_norm": 0.732523009843604,
      "learning_rate": 3.259347251813356e-06,
      "loss": 0.28086276054382325,
      "num_input_tokens_seen": 580572080,
      "step": 18215,
      "train_runtime": 47780.4812,
      "train_tokens_per_second": 12150.821
    },
    {
      "epoch": 1.2267708052787503,
      "grad_norm": 0.8303094419003498,
      "learning_rate": 3.256868810907064e-06,
      "loss": 0.32007722854614257,
      "num_input_tokens_seen": 580735920,
      "step": 18220,
      "train_runtime": 47793.5265,
      "train_tokens_per_second": 12150.933
    },
    {
      "epoch": 1.2271074602747105,
      "grad_norm": 0.846108700719284,
      "learning_rate": 3.2543908574618514e-06,
      "loss": 0.3041520595550537,
      "num_input_tokens_seen": 580896328,
      "step": 18225,
      "train_runtime": 47807.3159,
      "train_tokens_per_second": 12150.783
    },
    {
      "epoch": 1.2274441152706705,
      "grad_norm": 0.7348248942399288,
      "learning_rate": 3.251913392170668e-06,
      "loss": 0.29837563037872317,
      "num_input_tokens_seen": 581057616,
      "step": 18230,
      "train_runtime": 47821.0298,
      "train_tokens_per_second": 12150.671
    },
    {
      "epoch": 1.2277807702666308,
      "grad_norm": 0.7520647659959094,
      "learning_rate": 3.249436415726333e-06,
      "loss": 0.301265811920166,
      "num_input_tokens_seen": 581219368,
      "step": 18235,
      "train_runtime": 47834.7866,
      "train_tokens_per_second": 12150.558
    },
    {
      "epoch": 1.2281174252625908,
      "grad_norm": 0.8622896121132416,
      "learning_rate": 3.24695992882152e-06,
      "loss": 0.3059053421020508,
      "num_input_tokens_seen": 581377528,
      "step": 18240,
      "train_runtime": 47848.1416,
      "train_tokens_per_second": 12150.472
    },
    {
      "epoch": 1.228454080258551,
      "grad_norm": 0.8274784849189706,
      "learning_rate": 3.244483932148773e-06,
      "loss": 0.3040431499481201,
      "num_input_tokens_seen": 581535376,
      "step": 18245,
      "train_runtime": 47861.7296,
      "train_tokens_per_second": 12150.321
    },
    {
      "epoch": 1.228790735254511,
      "grad_norm": 0.7549838521912406,
      "learning_rate": 3.2420084264004966e-06,
      "loss": 0.29225361347198486,
      "num_input_tokens_seen": 581694280,
      "step": 18250,
      "train_runtime": 47875.7055,
      "train_tokens_per_second": 12150.093
    },
    {
      "epoch": 1.2291273902504714,
      "grad_norm": 0.7890188016839296,
      "learning_rate": 3.2395334122689594e-06,
      "loss": 0.2962925910949707,
      "num_input_tokens_seen": 581854992,
      "step": 18255,
      "train_runtime": 47888.138,
      "train_tokens_per_second": 12150.295
    },
    {
      "epoch": 1.2294640452464314,
      "grad_norm": 0.7629318323355759,
      "learning_rate": 3.2370588904462873e-06,
      "loss": 0.30828404426574707,
      "num_input_tokens_seen": 582017432,
      "step": 18260,
      "train_runtime": 47900.7998,
      "train_tokens_per_second": 12150.474
    },
    {
      "epoch": 1.2298007002423916,
      "grad_norm": 1.1505239459563361,
      "learning_rate": 3.2345848616244775e-06,
      "loss": 0.3083354949951172,
      "num_input_tokens_seen": 582168928,
      "step": 18265,
      "train_runtime": 47914.3968,
      "train_tokens_per_second": 12150.188
    },
    {
      "epoch": 1.2301373552383517,
      "grad_norm": 0.8664640237311136,
      "learning_rate": 3.2321113264953797e-06,
      "loss": 0.29277701377868653,
      "num_input_tokens_seen": 582329480,
      "step": 18270,
      "train_runtime": 47926.8613,
      "train_tokens_per_second": 12150.378
    },
    {
      "epoch": 1.230474010234312,
      "grad_norm": 0.8031158298531158,
      "learning_rate": 3.2296382857507124e-06,
      "loss": 0.28634214401245117,
      "num_input_tokens_seen": 582487728,
      "step": 18275,
      "train_runtime": 47939.9283,
      "train_tokens_per_second": 12150.367
    },
    {
      "epoch": 1.230810665230272,
      "grad_norm": 0.7172943433278278,
      "learning_rate": 3.227165740082053e-06,
      "loss": 0.3119499921798706,
      "num_input_tokens_seen": 582651176,
      "step": 18280,
      "train_runtime": 47953.0745,
      "train_tokens_per_second": 12150.445
    },
    {
      "epoch": 1.2311473202262322,
      "grad_norm": 0.748157491013375,
      "learning_rate": 3.2246936901808424e-06,
      "loss": 0.27671198844909667,
      "num_input_tokens_seen": 582814320,
      "step": 18285,
      "train_runtime": 47966.5303,
      "train_tokens_per_second": 12150.437
    },
    {
      "epoch": 1.2314839752221922,
      "grad_norm": 0.8262162707913374,
      "learning_rate": 3.2222221367383786e-06,
      "loss": 0.3189072132110596,
      "num_input_tokens_seen": 582975992,
      "step": 18290,
      "train_runtime": 47979.0424,
      "train_tokens_per_second": 12150.638
    },
    {
      "epoch": 1.2318206302181525,
      "grad_norm": 0.773105351272229,
      "learning_rate": 3.2197510804458265e-06,
      "loss": 0.28573269844055177,
      "num_input_tokens_seen": 583135064,
      "step": 18295,
      "train_runtime": 47992.7661,
      "train_tokens_per_second": 12150.478
    },
    {
      "epoch": 1.2321572852141125,
      "grad_norm": 0.6836337387790852,
      "learning_rate": 3.217280521994207e-06,
      "loss": 0.295214319229126,
      "num_input_tokens_seen": 583298088,
      "step": 18300,
      "train_runtime": 48006.2532,
      "train_tokens_per_second": 12150.461
    },
    {
      "epoch": 1.2324939402100727,
      "grad_norm": 0.7878509574429907,
      "learning_rate": 3.214810462074407e-06,
      "loss": 0.29805593490600585,
      "num_input_tokens_seen": 583457672,
      "step": 18305,
      "train_runtime": 48018.7028,
      "train_tokens_per_second": 12150.634
    },
    {
      "epoch": 1.2328305952060328,
      "grad_norm": 0.8278604048903027,
      "learning_rate": 3.2123409013771666e-06,
      "loss": 0.2960082530975342,
      "num_input_tokens_seen": 583618288,
      "step": 18310,
      "train_runtime": 48031.1282,
      "train_tokens_per_second": 12150.834
    },
    {
      "epoch": 1.233167250201993,
      "grad_norm": 0.7289641417184982,
      "learning_rate": 3.209871840593095e-06,
      "loss": 0.31537551879882814,
      "num_input_tokens_seen": 583781712,
      "step": 18315,
      "train_runtime": 48044.1301,
      "train_tokens_per_second": 12150.948
    },
    {
      "epoch": 1.233503905197953,
      "grad_norm": 0.8157290511283004,
      "learning_rate": 3.207403280412652e-06,
      "loss": 0.30629892349243165,
      "num_input_tokens_seen": 583942552,
      "step": 18320,
      "train_runtime": 48058.1345,
      "train_tokens_per_second": 12150.754
    },
    {
      "epoch": 1.2338405601939133,
      "grad_norm": 0.7960079110537235,
      "learning_rate": 3.2049352215261677e-06,
      "loss": 0.27916431427001953,
      "num_input_tokens_seen": 584105232,
      "step": 18325,
      "train_runtime": 48071.1704,
      "train_tokens_per_second": 12150.843
    },
    {
      "epoch": 1.2341772151898733,
      "grad_norm": 0.7003397429578239,
      "learning_rate": 3.2024676646238222e-06,
      "loss": 0.2954216480255127,
      "num_input_tokens_seen": 584266440,
      "step": 18330,
      "train_runtime": 48084.7875,
      "train_tokens_per_second": 12150.754
    },
    {
      "epoch": 1.2345138701858336,
      "grad_norm": 1.0413563958948153,
      "learning_rate": 3.2000006103956653e-06,
      "loss": 0.3101287841796875,
      "num_input_tokens_seen": 584424840,
      "step": 18335,
      "train_runtime": 48097.2837,
      "train_tokens_per_second": 12150.891
    },
    {
      "epoch": 1.2348505251817936,
      "grad_norm": 0.8658694552995145,
      "learning_rate": 3.1975340595315956e-06,
      "loss": 0.28044853210449217,
      "num_input_tokens_seen": 584581352,
      "step": 18340,
      "train_runtime": 48110.3356,
      "train_tokens_per_second": 12150.848
    },
    {
      "epoch": 1.2351871801777539,
      "grad_norm": 0.7618542224094096,
      "learning_rate": 3.1950680127213796e-06,
      "loss": 0.2872614860534668,
      "num_input_tokens_seen": 584740536,
      "step": 18345,
      "train_runtime": 48122.9989,
      "train_tokens_per_second": 12150.958
    },
    {
      "epoch": 1.235523835173714,
      "grad_norm": 0.8336117566995166,
      "learning_rate": 3.192602470654638e-06,
      "loss": 0.3127295017242432,
      "num_input_tokens_seen": 584900504,
      "step": 18350,
      "train_runtime": 48136.7686,
      "train_tokens_per_second": 12150.805
    },
    {
      "epoch": 1.2358604901696741,
      "grad_norm": 0.8024403588651453,
      "learning_rate": 3.190137434020853e-06,
      "loss": 0.29363412857055665,
      "num_input_tokens_seen": 585062056,
      "step": 18355,
      "train_runtime": 48149.6921,
      "train_tokens_per_second": 12150.899
    },
    {
      "epoch": 1.2361971451656342,
      "grad_norm": 0.9480512411683943,
      "learning_rate": 3.1876729035093616e-06,
      "loss": 0.29247169494628905,
      "num_input_tokens_seen": 585222080,
      "step": 18360,
      "train_runtime": 48162.9011,
      "train_tokens_per_second": 12150.889
    },
    {
      "epoch": 1.2365338001615944,
      "grad_norm": 0.8280887163803946,
      "learning_rate": 3.185208879809366e-06,
      "loss": 0.28556480407714846,
      "num_input_tokens_seen": 585383144,
      "step": 18365,
      "train_runtime": 48175.91,
      "train_tokens_per_second": 12150.951
    },
    {
      "epoch": 1.2368704551575544,
      "grad_norm": 0.7632898766832383,
      "learning_rate": 3.1827453636099183e-06,
      "loss": 0.29779829978942873,
      "num_input_tokens_seen": 585546984,
      "step": 18370,
      "train_runtime": 48188.3459,
      "train_tokens_per_second": 12151.216
    },
    {
      "epoch": 1.2372071101535147,
      "grad_norm": 0.9136465291321273,
      "learning_rate": 3.1802823555999356e-06,
      "loss": 0.27601118087768556,
      "num_input_tokens_seen": 585708968,
      "step": 18375,
      "train_runtime": 48202.1485,
      "train_tokens_per_second": 12151.097
    },
    {
      "epoch": 1.2375437651494747,
      "grad_norm": 1.0205214542122862,
      "learning_rate": 3.1778198564681885e-06,
      "loss": 0.33374767303466796,
      "num_input_tokens_seen": 585870264,
      "step": 18380,
      "train_runtime": 48214.8389,
      "train_tokens_per_second": 12151.244
    },
    {
      "epoch": 1.237880420145435,
      "grad_norm": 0.823940161056751,
      "learning_rate": 3.17535786690331e-06,
      "loss": 0.31405541896820066,
      "num_input_tokens_seen": 586028728,
      "step": 18385,
      "train_runtime": 48227.6414,
      "train_tokens_per_second": 12151.304
    },
    {
      "epoch": 1.238217075141395,
      "grad_norm": 0.9242764494569721,
      "learning_rate": 3.172896387593784e-06,
      "loss": 0.30832514762878416,
      "num_input_tokens_seen": 586186976,
      "step": 18390,
      "train_runtime": 48240.0564,
      "train_tokens_per_second": 12151.457
    },
    {
      "epoch": 1.2385537301373553,
      "grad_norm": 0.8272352829059965,
      "learning_rate": 3.1704354192279573e-06,
      "loss": 0.29794576168060305,
      "num_input_tokens_seen": 586346040,
      "step": 18395,
      "train_runtime": 48253.1358,
      "train_tokens_per_second": 12151.46
    },
    {
      "epoch": 1.2388903851333153,
      "grad_norm": 0.8711717698745463,
      "learning_rate": 3.1679749624940316e-06,
      "loss": 0.3032227993011475,
      "num_input_tokens_seen": 586505608,
      "step": 18400,
      "train_runtime": 48265.7782,
      "train_tokens_per_second": 12151.583
    },
    {
      "epoch": 1.2392270401292755,
      "grad_norm": 0.8654132947482516,
      "learning_rate": 3.165515018080067e-06,
      "loss": 0.3107591152191162,
      "num_input_tokens_seen": 586664296,
      "step": 18405,
      "train_runtime": 48279.3047,
      "train_tokens_per_second": 12151.465
    },
    {
      "epoch": 1.2395636951252356,
      "grad_norm": 0.9107676182775081,
      "learning_rate": 3.163055586673976e-06,
      "loss": 0.2882097244262695,
      "num_input_tokens_seen": 586821552,
      "step": 18410,
      "train_runtime": 48292.6638,
      "train_tokens_per_second": 12151.36
    },
    {
      "epoch": 1.2399003501211958,
      "grad_norm": 0.8948645268665629,
      "learning_rate": 3.1605966689635348e-06,
      "loss": 0.32485346794128417,
      "num_input_tokens_seen": 586981640,
      "step": 18415,
      "train_runtime": 48305.8877,
      "train_tokens_per_second": 12151.348
    },
    {
      "epoch": 1.2402370051171558,
      "grad_norm": 1.8378352283681567,
      "learning_rate": 3.1581382656363666e-06,
      "loss": 0.30637297630310056,
      "num_input_tokens_seen": 587133256,
      "step": 18420,
      "train_runtime": 48319.0316,
      "train_tokens_per_second": 12151.18
    },
    {
      "epoch": 1.240573660113116,
      "grad_norm": 0.6994908098971349,
      "learning_rate": 3.1556803773799616e-06,
      "loss": 0.30309300422668456,
      "num_input_tokens_seen": 587292264,
      "step": 18425,
      "train_runtime": 48332.1236,
      "train_tokens_per_second": 12151.179
    },
    {
      "epoch": 1.2409103151090761,
      "grad_norm": 0.7680223740591987,
      "learning_rate": 3.1532230048816564e-06,
      "loss": 0.2859400749206543,
      "num_input_tokens_seen": 587452256,
      "step": 18430,
      "train_runtime": 48345.1392,
      "train_tokens_per_second": 12151.217
    },
    {
      "epoch": 1.2412469701050364,
      "grad_norm": 0.755698465757994,
      "learning_rate": 3.150766148828651e-06,
      "loss": 0.3058071851730347,
      "num_input_tokens_seen": 587609720,
      "step": 18435,
      "train_runtime": 48357.6507,
      "train_tokens_per_second": 12151.329
    },
    {
      "epoch": 1.2415836251009964,
      "grad_norm": 0.768736794938594,
      "learning_rate": 3.1483098099079934e-06,
      "loss": 0.2847443580627441,
      "num_input_tokens_seen": 587772536,
      "step": 18440,
      "train_runtime": 48371.2295,
      "train_tokens_per_second": 12151.284
    },
    {
      "epoch": 1.2419202800969567,
      "grad_norm": 0.752636058926188,
      "learning_rate": 3.1458539888065937e-06,
      "loss": 0.3014049530029297,
      "num_input_tokens_seen": 587930216,
      "step": 18445,
      "train_runtime": 48384.6189,
      "train_tokens_per_second": 12151.18
    },
    {
      "epoch": 1.2422569350929167,
      "grad_norm": 0.8716597231988518,
      "learning_rate": 3.143398686211212e-06,
      "loss": 0.2980872631072998,
      "num_input_tokens_seen": 588090624,
      "step": 18450,
      "train_runtime": 48398.5205,
      "train_tokens_per_second": 12151.004
    },
    {
      "epoch": 1.242593590088877,
      "grad_norm": 0.8221083688354442,
      "learning_rate": 3.1409439028084706e-06,
      "loss": 0.3100332260131836,
      "num_input_tokens_seen": 588250072,
      "step": 18455,
      "train_runtime": 48411.4534,
      "train_tokens_per_second": 12151.052
    },
    {
      "epoch": 1.2429302450848372,
      "grad_norm": 0.7941737233275576,
      "learning_rate": 3.1384896392848363e-06,
      "loss": 0.29020214080810547,
      "num_input_tokens_seen": 588410312,
      "step": 18460,
      "train_runtime": 48424.9592,
      "train_tokens_per_second": 12150.972
    },
    {
      "epoch": 1.2432669000807972,
      "grad_norm": 0.7758460075391419,
      "learning_rate": 3.1360358963266404e-06,
      "loss": 0.3344435691833496,
      "num_input_tokens_seen": 588570592,
      "step": 18465,
      "train_runtime": 48437.6883,
      "train_tokens_per_second": 12151.088
    },
    {
      "epoch": 1.2436035550767572,
      "grad_norm": 0.8037128736311963,
      "learning_rate": 3.1335826746200594e-06,
      "loss": 0.2929240226745605,
      "num_input_tokens_seen": 588724760,
      "step": 18470,
      "train_runtime": 48451.0694,
      "train_tokens_per_second": 12150.914
    },
    {
      "epoch": 1.2439402100727175,
      "grad_norm": 0.787467411714124,
      "learning_rate": 3.1311299748511335e-06,
      "loss": 0.306843638420105,
      "num_input_tokens_seen": 588886048,
      "step": 18475,
      "train_runtime": 48464.3406,
      "train_tokens_per_second": 12150.914
    },
    {
      "epoch": 1.2442768650686777,
      "grad_norm": 0.8248330818865219,
      "learning_rate": 3.1286777977057504e-06,
      "loss": 0.3001504421234131,
      "num_input_tokens_seen": 589043904,
      "step": 18480,
      "train_runtime": 48476.8201,
      "train_tokens_per_second": 12151.043
    },
    {
      "epoch": 1.2446135200646378,
      "grad_norm": 0.7725886359498364,
      "learning_rate": 3.126226143869655e-06,
      "loss": 0.3069874048233032,
      "num_input_tokens_seen": 589198776,
      "step": 18485,
      "train_runtime": 48490.2215,
      "train_tokens_per_second": 12150.878
    },
    {
      "epoch": 1.2449501750605978,
      "grad_norm": 0.6971446511754704,
      "learning_rate": 3.1237750140284424e-06,
      "loss": 0.26791791915893554,
      "num_input_tokens_seen": 589357392,
      "step": 18490,
      "train_runtime": 48503.607,
      "train_tokens_per_second": 12150.795
    },
    {
      "epoch": 1.245286830056558,
      "grad_norm": 0.7681998122647371,
      "learning_rate": 3.1213244088675647e-06,
      "loss": 0.31435227394104004,
      "num_input_tokens_seen": 589517760,
      "step": 18495,
      "train_runtime": 48516.4513,
      "train_tokens_per_second": 12150.884
    },
    {
      "epoch": 1.2456234850525183,
      "grad_norm": 0.8802744781549877,
      "learning_rate": 3.1188743290723246e-06,
      "loss": 0.3252289056777954,
      "num_input_tokens_seen": 589676696,
      "step": 18500,
      "train_runtime": 48528.9997,
      "train_tokens_per_second": 12151.017
    },
    {
      "epoch": 1.2459601400484783,
      "grad_norm": 0.8280790221363046,
      "learning_rate": 3.116424775327882e-06,
      "loss": 0.3172433376312256,
      "num_input_tokens_seen": 589835632,
      "step": 18505,
      "train_runtime": 48541.5304,
      "train_tokens_per_second": 12151.154
    },
    {
      "epoch": 1.2462967950444384,
      "grad_norm": 0.7099793541579976,
      "learning_rate": 3.1139757483192423e-06,
      "loss": 0.3101557016372681,
      "num_input_tokens_seen": 589997200,
      "step": 18510,
      "train_runtime": 48554.22,
      "train_tokens_per_second": 12151.306
    },
    {
      "epoch": 1.2466334500403986,
      "grad_norm": 0.8277130000482371,
      "learning_rate": 3.1115272487312727e-06,
      "loss": 0.2895765781402588,
      "num_input_tokens_seen": 590151288,
      "step": 18515,
      "train_runtime": 48567.9546,
      "train_tokens_per_second": 12151.043
    },
    {
      "epoch": 1.2469701050363589,
      "grad_norm": 0.8635354852918634,
      "learning_rate": 3.109079277248684e-06,
      "loss": 0.27374701499938964,
      "num_input_tokens_seen": 590303728,
      "step": 18520,
      "train_runtime": 48580.7261,
      "train_tokens_per_second": 12150.986
    },
    {
      "epoch": 1.2473067600323189,
      "grad_norm": 0.8161375864517258,
      "learning_rate": 3.1066318345560463e-06,
      "loss": 0.31182475090026857,
      "num_input_tokens_seen": 590456352,
      "step": 18525,
      "train_runtime": 48593.2724,
      "train_tokens_per_second": 12150.99
    },
    {
      "epoch": 1.247643415028279,
      "grad_norm": 0.7674389736394365,
      "learning_rate": 3.1041849213377782e-06,
      "loss": 0.3141264200210571,
      "num_input_tokens_seen": 590620192,
      "step": 18530,
      "train_runtime": 48605.7074,
      "train_tokens_per_second": 12151.252
    },
    {
      "epoch": 1.2479800700242392,
      "grad_norm": 0.8094534425257751,
      "learning_rate": 3.101738538278154e-06,
      "loss": 0.29593911170959475,
      "num_input_tokens_seen": 590777336,
      "step": 18535,
      "train_runtime": 48618.5567,
      "train_tokens_per_second": 12151.273
    },
    {
      "epoch": 1.2483167250201994,
      "grad_norm": 0.7450796897526799,
      "learning_rate": 3.0992926860612914e-06,
      "loss": 0.29093637466430666,
      "num_input_tokens_seen": 590937120,
      "step": 18540,
      "train_runtime": 48631.1383,
      "train_tokens_per_second": 12151.415
    },
    {
      "epoch": 1.2486533800161594,
      "grad_norm": 0.7959104325146444,
      "learning_rate": 3.0968473653711693e-06,
      "loss": 0.3038785457611084,
      "num_input_tokens_seen": 591100960,
      "step": 18545,
      "train_runtime": 48643.5864,
      "train_tokens_per_second": 12151.673
    },
    {
      "epoch": 1.2489900350121195,
      "grad_norm": 0.7691748225760048,
      "learning_rate": 3.094402576891613e-06,
      "loss": 0.3200667142868042,
      "num_input_tokens_seen": 591260104,
      "step": 18550,
      "train_runtime": 48656.2174,
      "train_tokens_per_second": 12151.789
    },
    {
      "epoch": 1.2493266900080797,
      "grad_norm": 0.7662235054039968,
      "learning_rate": 3.0919583213063004e-06,
      "loss": 0.3187409400939941,
      "num_input_tokens_seen": 591419000,
      "step": 18555,
      "train_runtime": 48668.6872,
      "train_tokens_per_second": 12151.941
    },
    {
      "epoch": 1.24966334500404,
      "grad_norm": 0.7623305613154947,
      "learning_rate": 3.0895145992987574e-06,
      "loss": 0.29469714164733884,
      "num_input_tokens_seen": 591582840,
      "step": 18560,
      "train_runtime": 48681.1393,
      "train_tokens_per_second": 12152.198
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.7221757026638028,
      "learning_rate": 3.087071411552366e-06,
      "loss": 0.28711910247802735,
      "num_input_tokens_seen": 591742000,
      "step": 18565,
      "train_runtime": 48694.0612,
      "train_tokens_per_second": 12152.242
    },
    {
      "epoch": 1.25033665499596,
      "grad_norm": 0.803799283905667,
      "learning_rate": 3.084628758750353e-06,
      "loss": 0.3040918827056885,
      "num_input_tokens_seen": 591901712,
      "step": 18570,
      "train_runtime": 48707.1391,
      "train_tokens_per_second": 12152.258
    },
    {
      "epoch": 1.2506733099919203,
      "grad_norm": 0.8955465889115405,
      "learning_rate": 3.082186641575801e-06,
      "loss": 0.3238292455673218,
      "num_input_tokens_seen": 592064696,
      "step": 18575,
      "train_runtime": 48719.9888,
      "train_tokens_per_second": 12152.398
    },
    {
      "epoch": 1.2510099649878805,
      "grad_norm": 0.7710750898141562,
      "learning_rate": 3.0797450607116384e-06,
      "loss": 0.2966698169708252,
      "num_input_tokens_seen": 592225616,
      "step": 18580,
      "train_runtime": 48732.9799,
      "train_tokens_per_second": 12152.461
    },
    {
      "epoch": 1.2513466199838406,
      "grad_norm": 0.7768847475404367,
      "learning_rate": 3.0773040168406475e-06,
      "loss": 0.3099688529968262,
      "num_input_tokens_seen": 592389456,
      "step": 18585,
      "train_runtime": 48745.4282,
      "train_tokens_per_second": 12152.718
    },
    {
      "epoch": 1.2516832749798006,
      "grad_norm": 0.88533715743012,
      "learning_rate": 3.074863510645456e-06,
      "loss": 0.3184183597564697,
      "num_input_tokens_seen": 592549896,
      "step": 18590,
      "train_runtime": 48759.1701,
      "train_tokens_per_second": 12152.584
    },
    {
      "epoch": 1.2520199299757608,
      "grad_norm": 0.7081451322686154,
      "learning_rate": 3.072423542808546e-06,
      "loss": 0.30428285598754884,
      "num_input_tokens_seen": 592706712,
      "step": 18595,
      "train_runtime": 48771.625,
      "train_tokens_per_second": 12152.696
    },
    {
      "epoch": 1.252356584971721,
      "grad_norm": 0.8411863854651365,
      "learning_rate": 3.0699841140122443e-06,
      "loss": 0.2989403963088989,
      "num_input_tokens_seen": 592864832,
      "step": 18600,
      "train_runtime": 48784.1101,
      "train_tokens_per_second": 12152.827
    },
    {
      "epoch": 1.2526932399676811,
      "grad_norm": 0.79369129364369,
      "learning_rate": 3.0675452249387327e-06,
      "loss": 0.282259464263916,
      "num_input_tokens_seen": 593027080,
      "step": 18605,
      "train_runtime": 48797.2692,
      "train_tokens_per_second": 12152.874
    },
    {
      "epoch": 1.2530298949636411,
      "grad_norm": 0.7977932583385071,
      "learning_rate": 3.0651068762700354e-06,
      "loss": 0.3334193706512451,
      "num_input_tokens_seen": 593183784,
      "step": 18610,
      "train_runtime": 48810.1179,
      "train_tokens_per_second": 12152.886
    },
    {
      "epoch": 1.2533665499596014,
      "grad_norm": 0.8483578490181285,
      "learning_rate": 3.0626690686880313e-06,
      "loss": 0.3076553583145142,
      "num_input_tokens_seen": 593347512,
      "step": 18615,
      "train_runtime": 48823.2048,
      "train_tokens_per_second": 12152.982
    },
    {
      "epoch": 1.2537032049555616,
      "grad_norm": 0.8476084926325579,
      "learning_rate": 3.060231802874443e-06,
      "loss": 0.31014347076416016,
      "num_input_tokens_seen": 593508880,
      "step": 18620,
      "train_runtime": 48835.7425,
      "train_tokens_per_second": 12153.166
    },
    {
      "epoch": 1.2540398599515217,
      "grad_norm": 0.7972967217870532,
      "learning_rate": 3.0577950795108462e-06,
      "loss": 0.30380656719207766,
      "num_input_tokens_seen": 593672080,
      "step": 18625,
      "train_runtime": 48849.3257,
      "train_tokens_per_second": 12153.127
    },
    {
      "epoch": 1.2543765149474817,
      "grad_norm": 0.814858153426829,
      "learning_rate": 3.055358899278662e-06,
      "loss": 0.30417346954345703,
      "num_input_tokens_seen": 593833280,
      "step": 18630,
      "train_runtime": 48862.2297,
      "train_tokens_per_second": 12153.217
    },
    {
      "epoch": 1.254713169943442,
      "grad_norm": 0.7816142980744644,
      "learning_rate": 3.052923262859162e-06,
      "loss": 0.28918018341064455,
      "num_input_tokens_seen": 593993408,
      "step": 18635,
      "train_runtime": 48875.1578,
      "train_tokens_per_second": 12153.279
    },
    {
      "epoch": 1.2550498249394022,
      "grad_norm": 0.7316289574612365,
      "learning_rate": 3.0504881709334603e-06,
      "loss": 0.2974682331085205,
      "num_input_tokens_seen": 594156864,
      "step": 18640,
      "train_runtime": 48888.1114,
      "train_tokens_per_second": 12153.402
    },
    {
      "epoch": 1.2553864799353622,
      "grad_norm": 0.7389347439232504,
      "learning_rate": 3.0480536241825263e-06,
      "loss": 0.2809246301651001,
      "num_input_tokens_seen": 594320024,
      "step": 18645,
      "train_runtime": 48901.7133,
      "train_tokens_per_second": 12153.358
    },
    {
      "epoch": 1.2557231349313223,
      "grad_norm": 0.7948340319390428,
      "learning_rate": 3.0456196232871715e-06,
      "loss": 0.3050450086593628,
      "num_input_tokens_seen": 594478264,
      "step": 18650,
      "train_runtime": 48915.7119,
      "train_tokens_per_second": 12153.115
    },
    {
      "epoch": 1.2560597899272825,
      "grad_norm": 0.7846576015043074,
      "learning_rate": 3.043186168928059e-06,
      "loss": 0.3019444942474365,
      "num_input_tokens_seen": 594638384,
      "step": 18655,
      "train_runtime": 48929.2866,
      "train_tokens_per_second": 12153.016
    },
    {
      "epoch": 1.2563964449232428,
      "grad_norm": 0.8996658631717475,
      "learning_rate": 3.040753261785692e-06,
      "loss": 0.26363744735717776,
      "num_input_tokens_seen": 594791960,
      "step": 18660,
      "train_runtime": 48942.0032,
      "train_tokens_per_second": 12152.996
    },
    {
      "epoch": 1.2567330999192028,
      "grad_norm": 0.7433787411482523,
      "learning_rate": 3.038320902540429e-06,
      "loss": 0.28797097206115724,
      "num_input_tokens_seen": 594955800,
      "step": 18665,
      "train_runtime": 48954.4507,
      "train_tokens_per_second": 12153.252
    },
    {
      "epoch": 1.2570697549151628,
      "grad_norm": 0.8491208696472583,
      "learning_rate": 3.0358890918724706e-06,
      "loss": 0.30237703323364257,
      "num_input_tokens_seen": 595118152,
      "step": 18670,
      "train_runtime": 48967.5215,
      "train_tokens_per_second": 12153.324
    },
    {
      "epoch": 1.257406409911123,
      "grad_norm": 0.8289160563753655,
      "learning_rate": 3.0334578304618646e-06,
      "loss": 0.2913827419281006,
      "num_input_tokens_seen": 595271112,
      "step": 18675,
      "train_runtime": 48980.6431,
      "train_tokens_per_second": 12153.191
    },
    {
      "epoch": 1.2577430649070833,
      "grad_norm": 0.7832607104535844,
      "learning_rate": 3.0310271189885037e-06,
      "loss": 0.304503345489502,
      "num_input_tokens_seen": 595424368,
      "step": 18680,
      "train_runtime": 48993.9969,
      "train_tokens_per_second": 12153.007
    },
    {
      "epoch": 1.2580797199030433,
      "grad_norm": 0.761528586215838,
      "learning_rate": 3.0285969581321328e-06,
      "loss": 0.2896098613739014,
      "num_input_tokens_seen": 595584936,
      "step": 18685,
      "train_runtime": 49007.1602,
      "train_tokens_per_second": 12153.019
    },
    {
      "epoch": 1.2584163748990034,
      "grad_norm": 0.8598814521591376,
      "learning_rate": 3.0261673485723343e-06,
      "loss": 0.3433079719543457,
      "num_input_tokens_seen": 595742464,
      "step": 18690,
      "train_runtime": 49020.5829,
      "train_tokens_per_second": 12152.905
    },
    {
      "epoch": 1.2587530298949636,
      "grad_norm": 0.7650156769969552,
      "learning_rate": 3.023738290988544e-06,
      "loss": 0.28599891662597654,
      "num_input_tokens_seen": 595902976,
      "step": 18695,
      "train_runtime": 49033.5055,
      "train_tokens_per_second": 12152.975
    },
    {
      "epoch": 1.2590896848909239,
      "grad_norm": 0.7865983196623746,
      "learning_rate": 3.021309786060037e-06,
      "loss": 0.2984224796295166,
      "num_input_tokens_seen": 596059392,
      "step": 18700,
      "train_runtime": 49046.1265,
      "train_tokens_per_second": 12153.037
    },
    {
      "epoch": 1.259426339886884,
      "grad_norm": 0.7340309559741682,
      "learning_rate": 3.018881834465941e-06,
      "loss": 0.2849896430969238,
      "num_input_tokens_seen": 596216648,
      "step": 18705,
      "train_runtime": 49059.4885,
      "train_tokens_per_second": 12152.932
    },
    {
      "epoch": 1.259762994882844,
      "grad_norm": 0.8722883304282024,
      "learning_rate": 3.016454436885221e-06,
      "loss": 0.31316194534301756,
      "num_input_tokens_seen": 596373384,
      "step": 18710,
      "train_runtime": 49071.9163,
      "train_tokens_per_second": 12153.049
    },
    {
      "epoch": 1.2600996498788042,
      "grad_norm": 0.8848448211432962,
      "learning_rate": 3.0140275939966934e-06,
      "loss": 0.29825854301452637,
      "num_input_tokens_seen": 596534464,
      "step": 18715,
      "train_runtime": 49084.8209,
      "train_tokens_per_second": 12153.135
    },
    {
      "epoch": 1.2604363048747644,
      "grad_norm": 0.83738244242207,
      "learning_rate": 3.0116013064790166e-06,
      "loss": 0.3469051122665405,
      "num_input_tokens_seen": 596697272,
      "step": 18720,
      "train_runtime": 49098.2894,
      "train_tokens_per_second": 12153.117
    },
    {
      "epoch": 1.2607729598707245,
      "grad_norm": 0.8478094043957985,
      "learning_rate": 3.0091755750106937e-06,
      "loss": 0.30471096038818357,
      "num_input_tokens_seen": 596856608,
      "step": 18725,
      "train_runtime": 49112.2456,
      "train_tokens_per_second": 12152.908
    },
    {
      "epoch": 1.2611096148666845,
      "grad_norm": 0.7687994753008406,
      "learning_rate": 3.006750400270072e-06,
      "loss": 0.2853196620941162,
      "num_input_tokens_seen": 597015712,
      "step": 18730,
      "train_runtime": 49125.4476,
      "train_tokens_per_second": 12152.881
    },
    {
      "epoch": 1.2614462698626447,
      "grad_norm": 0.7273027775614012,
      "learning_rate": 3.0043257829353466e-06,
      "loss": 0.28669233322143556,
      "num_input_tokens_seen": 597170232,
      "step": 18735,
      "train_runtime": 49137.899,
      "train_tokens_per_second": 12152.946
    },
    {
      "epoch": 1.261782924858605,
      "grad_norm": 0.9139145233962981,
      "learning_rate": 3.0019017236845504e-06,
      "loss": 0.30709290504455566,
      "num_input_tokens_seen": 597330056,
      "step": 18740,
      "train_runtime": 49151.2814,
      "train_tokens_per_second": 12152.889
    },
    {
      "epoch": 1.262119579854565,
      "grad_norm": 0.901721490270064,
      "learning_rate": 2.9994782231955673e-06,
      "loss": 0.30857210159301757,
      "num_input_tokens_seen": 597484664,
      "step": 18745,
      "train_runtime": 49164.3587,
      "train_tokens_per_second": 12152.801
    },
    {
      "epoch": 1.262456234850525,
      "grad_norm": 0.7028076032662609,
      "learning_rate": 2.9970552821461185e-06,
      "loss": 0.29036540985107423,
      "num_input_tokens_seen": 597645744,
      "step": 18750,
      "train_runtime": 49177.9044,
      "train_tokens_per_second": 12152.729
    },
    {
      "epoch": 1.2627928898464853,
      "grad_norm": 0.9106764953190556,
      "learning_rate": 2.994632901213775e-06,
      "loss": 0.30046758651733396,
      "num_input_tokens_seen": 597802072,
      "step": 18755,
      "train_runtime": 49191.4022,
      "train_tokens_per_second": 12152.572
    },
    {
      "epoch": 1.2631295448424456,
      "grad_norm": 0.8737859957723585,
      "learning_rate": 2.9922110810759443e-06,
      "loss": 0.31121273040771485,
      "num_input_tokens_seen": 597959448,
      "step": 18760,
      "train_runtime": 49204.6222,
      "train_tokens_per_second": 12152.506
    },
    {
      "epoch": 1.2634661998384056,
      "grad_norm": 0.8205909059256062,
      "learning_rate": 2.989789822409883e-06,
      "loss": 0.3124363660812378,
      "num_input_tokens_seen": 598120184,
      "step": 18765,
      "train_runtime": 49217.9066,
      "train_tokens_per_second": 12152.491
    },
    {
      "epoch": 1.2638028548343656,
      "grad_norm": 0.8292142974223398,
      "learning_rate": 2.9873691258926864e-06,
      "loss": 0.33669533729553225,
      "num_input_tokens_seen": 598284024,
      "step": 18770,
      "train_runtime": 49230.3356,
      "train_tokens_per_second": 12152.751
    },
    {
      "epoch": 1.2641395098303259,
      "grad_norm": 0.8065415172989155,
      "learning_rate": 2.9849489922012985e-06,
      "loss": 0.29337100982666015,
      "num_input_tokens_seen": 598441672,
      "step": 18775,
      "train_runtime": 49243.0057,
      "train_tokens_per_second": 12152.826
    },
    {
      "epoch": 1.264476164826286,
      "grad_norm": 0.7795059725357247,
      "learning_rate": 2.982529422012498e-06,
      "loss": 0.3284898042678833,
      "num_input_tokens_seen": 598604168,
      "step": 18780,
      "train_runtime": 49255.616,
      "train_tokens_per_second": 12153.014
    },
    {
      "epoch": 1.2648128198222461,
      "grad_norm": 0.8205886554179143,
      "learning_rate": 2.9801104160029127e-06,
      "loss": 0.2727202892303467,
      "num_input_tokens_seen": 598760400,
      "step": 18785,
      "train_runtime": 49268.6174,
      "train_tokens_per_second": 12152.978
    },
    {
      "epoch": 1.2651494748182062,
      "grad_norm": 0.8034382924090836,
      "learning_rate": 2.9776919748490063e-06,
      "loss": 0.30403928756713866,
      "num_input_tokens_seen": 598918128,
      "step": 18790,
      "train_runtime": 49281.4237,
      "train_tokens_per_second": 12153.02
    },
    {
      "epoch": 1.2654861298141664,
      "grad_norm": 0.8442680210073279,
      "learning_rate": 2.975274099227092e-06,
      "loss": 0.322404670715332,
      "num_input_tokens_seen": 599075696,
      "step": 18795,
      "train_runtime": 49294.8268,
      "train_tokens_per_second": 12152.912
    },
    {
      "epoch": 1.2658227848101267,
      "grad_norm": 0.8855148054974662,
      "learning_rate": 2.9728567898133183e-06,
      "loss": 0.3041447877883911,
      "num_input_tokens_seen": 599233536,
      "step": 18800,
      "train_runtime": 49307.365,
      "train_tokens_per_second": 12153.023
    },
    {
      "epoch": 1.2661594398060867,
      "grad_norm": 0.7390092429278636,
      "learning_rate": 2.9704400472836816e-06,
      "loss": 0.3029345989227295,
      "num_input_tokens_seen": 599395768,
      "step": 18805,
      "train_runtime": 49321.3061,
      "train_tokens_per_second": 12152.877
    },
    {
      "epoch": 1.266496094802047,
      "grad_norm": 0.9788237053394778,
      "learning_rate": 2.9680238723140108e-06,
      "loss": 0.323366641998291,
      "num_input_tokens_seen": 599552864,
      "step": 18810,
      "train_runtime": 49334.8884,
      "train_tokens_per_second": 12152.716
    },
    {
      "epoch": 1.266832749798007,
      "grad_norm": 0.8215753605147385,
      "learning_rate": 2.9656082655799866e-06,
      "loss": 0.3169821262359619,
      "num_input_tokens_seen": 599705328,
      "step": 18815,
      "train_runtime": 49347.5162,
      "train_tokens_per_second": 12152.695
    },
    {
      "epoch": 1.2671694047939672,
      "grad_norm": 0.8977088472527147,
      "learning_rate": 2.9631932277571225e-06,
      "loss": 0.2970638990402222,
      "num_input_tokens_seen": 599868192,
      "step": 18820,
      "train_runtime": 49361.5845,
      "train_tokens_per_second": 12152.531
    },
    {
      "epoch": 1.2675060597899273,
      "grad_norm": 0.8567378069145913,
      "learning_rate": 2.9607787595207794e-06,
      "loss": 0.28716063499450684,
      "num_input_tokens_seen": 600029376,
      "step": 18825,
      "train_runtime": 49374.8159,
      "train_tokens_per_second": 12152.539
    },
    {
      "epoch": 1.2678427147858875,
      "grad_norm": 0.7319754507041214,
      "learning_rate": 2.9583648615461514e-06,
      "loss": 0.29655938148498534,
      "num_input_tokens_seen": 600187792,
      "step": 18830,
      "train_runtime": 49388.6552,
      "train_tokens_per_second": 12152.341
    },
    {
      "epoch": 1.2681793697818475,
      "grad_norm": 0.7576553852842867,
      "learning_rate": 2.955951534508281e-06,
      "loss": 0.34119083881378176,
      "num_input_tokens_seen": 600347360,
      "step": 18835,
      "train_runtime": 49401.9523,
      "train_tokens_per_second": 12152.3
    },
    {
      "epoch": 1.2685160247778078,
      "grad_norm": 0.7662576056342805,
      "learning_rate": 2.9535387790820452e-06,
      "loss": 0.31615149974823,
      "num_input_tokens_seen": 600507912,
      "step": 18840,
      "train_runtime": 49414.3753,
      "train_tokens_per_second": 12152.494
    },
    {
      "epoch": 1.2688526797737678,
      "grad_norm": 0.8285085753924084,
      "learning_rate": 2.951126595942165e-06,
      "loss": 0.2867731094360352,
      "num_input_tokens_seen": 600669520,
      "step": 18845,
      "train_runtime": 49427.426,
      "train_tokens_per_second": 12152.555
    },
    {
      "epoch": 1.269189334769728,
      "grad_norm": 0.7500496699263153,
      "learning_rate": 2.948714985763197e-06,
      "loss": 0.2837223052978516,
      "num_input_tokens_seen": 600827376,
      "step": 18850,
      "train_runtime": 49440.6377,
      "train_tokens_per_second": 12152.501
    },
    {
      "epoch": 1.269525989765688,
      "grad_norm": 0.8559095942215431,
      "learning_rate": 2.9463039492195445e-06,
      "loss": 0.303429913520813,
      "num_input_tokens_seen": 600991216,
      "step": 18855,
      "train_runtime": 49453.0737,
      "train_tokens_per_second": 12152.758
    },
    {
      "epoch": 1.2698626447616483,
      "grad_norm": 0.8215644464949324,
      "learning_rate": 2.9438934869854417e-06,
      "loss": 0.28866920471191404,
      "num_input_tokens_seen": 601153024,
      "step": 18860,
      "train_runtime": 49466.2547,
      "train_tokens_per_second": 12152.79
    },
    {
      "epoch": 1.2701992997576084,
      "grad_norm": 0.8547411099372176,
      "learning_rate": 2.9414835997349705e-06,
      "loss": 0.2780210733413696,
      "num_input_tokens_seen": 601314752,
      "step": 18865,
      "train_runtime": 49479.3557,
      "train_tokens_per_second": 12152.841
    },
    {
      "epoch": 1.2705359547535686,
      "grad_norm": 0.7405285784926955,
      "learning_rate": 2.939074288142045e-06,
      "loss": 0.31252293586730956,
      "num_input_tokens_seen": 601477112,
      "step": 18870,
      "train_runtime": 49492.6789,
      "train_tokens_per_second": 12152.85
    },
    {
      "epoch": 1.2708726097495286,
      "grad_norm": 0.7863578441451208,
      "learning_rate": 2.9366655528804257e-06,
      "loss": 0.2919419765472412,
      "num_input_tokens_seen": 601639880,
      "step": 18875,
      "train_runtime": 49506.1482,
      "train_tokens_per_second": 12152.832
    },
    {
      "epoch": 1.271209264745489,
      "grad_norm": 0.7166480684490629,
      "learning_rate": 2.934257394623702e-06,
      "loss": 0.2909376621246338,
      "num_input_tokens_seen": 601800304,
      "step": 18880,
      "train_runtime": 49518.9493,
      "train_tokens_per_second": 12152.93
    },
    {
      "epoch": 1.271545919741449,
      "grad_norm": 0.7163817859262532,
      "learning_rate": 2.9318498140453134e-06,
      "loss": 0.29977262020111084,
      "num_input_tokens_seen": 601960496,
      "step": 18885,
      "train_runtime": 49531.9719,
      "train_tokens_per_second": 12152.969
    },
    {
      "epoch": 1.2718825747374092,
      "grad_norm": 0.9082748367982564,
      "learning_rate": 2.929442811818527e-06,
      "loss": 0.30515575408935547,
      "num_input_tokens_seen": 602116736,
      "step": 18890,
      "train_runtime": 49544.7314,
      "train_tokens_per_second": 12152.992
    },
    {
      "epoch": 1.2722192297333692,
      "grad_norm": 0.8393355793129604,
      "learning_rate": 2.927036388616457e-06,
      "loss": 0.30097830295562744,
      "num_input_tokens_seen": 602270648,
      "step": 18895,
      "train_runtime": 49557.7875,
      "train_tokens_per_second": 12152.896
    },
    {
      "epoch": 1.2725558847293295,
      "grad_norm": 0.8991547677386357,
      "learning_rate": 2.9246305451120494e-06,
      "loss": 0.3226932048797607,
      "num_input_tokens_seen": 602424440,
      "step": 18900,
      "train_runtime": 49570.3818,
      "train_tokens_per_second": 12152.911
    },
    {
      "epoch": 1.2728925397252895,
      "grad_norm": 0.7858401051914873,
      "learning_rate": 2.922225281978095e-06,
      "loss": 0.3055391788482666,
      "num_input_tokens_seen": 602586976,
      "step": 18905,
      "train_runtime": 49583.729,
      "train_tokens_per_second": 12152.918
    },
    {
      "epoch": 1.2732291947212497,
      "grad_norm": 0.8484824536541136,
      "learning_rate": 2.919820599887212e-06,
      "loss": 0.29495623111724856,
      "num_input_tokens_seen": 602744896,
      "step": 18910,
      "train_runtime": 49596.211,
      "train_tokens_per_second": 12153.043
    },
    {
      "epoch": 1.2735658497172098,
      "grad_norm": 0.9002199793468915,
      "learning_rate": 2.917416499511866e-06,
      "loss": 0.29449806213378904,
      "num_input_tokens_seen": 602899200,
      "step": 18915,
      "train_runtime": 49609.4428,
      "train_tokens_per_second": 12152.912
    },
    {
      "epoch": 1.27390250471317,
      "grad_norm": 0.7683660618366864,
      "learning_rate": 2.9150129815243553e-06,
      "loss": 0.28325610160827636,
      "num_input_tokens_seen": 603060584,
      "step": 18920,
      "train_runtime": 49622.4751,
      "train_tokens_per_second": 12152.973
    },
    {
      "epoch": 1.27423915970913,
      "grad_norm": 0.884450978488993,
      "learning_rate": 2.912610046596816e-06,
      "loss": 0.3023883581161499,
      "num_input_tokens_seen": 603217736,
      "step": 18925,
      "train_runtime": 49635.8435,
      "train_tokens_per_second": 12152.866
    },
    {
      "epoch": 1.2745758147050903,
      "grad_norm": 0.748728177113573,
      "learning_rate": 2.91020769540122e-06,
      "loss": 0.30546083450317385,
      "num_input_tokens_seen": 603370984,
      "step": 18930,
      "train_runtime": 49649.3167,
      "train_tokens_per_second": 12152.654
    },
    {
      "epoch": 1.2749124697010503,
      "grad_norm": 0.7573276132254194,
      "learning_rate": 2.9078059286093795e-06,
      "loss": 0.2985365867614746,
      "num_input_tokens_seen": 603529784,
      "step": 18935,
      "train_runtime": 49662.5921,
      "train_tokens_per_second": 12152.603
    },
    {
      "epoch": 1.2752491246970106,
      "grad_norm": 0.7997916730115476,
      "learning_rate": 2.905404746892937e-06,
      "loss": 0.28757438659667967,
      "num_input_tokens_seen": 603685464,
      "step": 18940,
      "train_runtime": 49675.1122,
      "train_tokens_per_second": 12152.674
    },
    {
      "epoch": 1.2755857796929706,
      "grad_norm": 0.7542760729385981,
      "learning_rate": 2.903004150923382e-06,
      "loss": 0.2960397958755493,
      "num_input_tokens_seen": 603846928,
      "step": 18945,
      "train_runtime": 49688.1633,
      "train_tokens_per_second": 12152.732
    },
    {
      "epoch": 1.2759224346889309,
      "grad_norm": 0.8174320935214524,
      "learning_rate": 2.9006041413720252e-06,
      "loss": 0.3410667896270752,
      "num_input_tokens_seen": 604006120,
      "step": 18950,
      "train_runtime": 49700.9718,
      "train_tokens_per_second": 12152.803
    },
    {
      "epoch": 1.2762590896848909,
      "grad_norm": 0.8412227898758261,
      "learning_rate": 2.8982047189100276e-06,
      "loss": 0.3006908416748047,
      "num_input_tokens_seen": 604168768,
      "step": 18955,
      "train_runtime": 49713.8767,
      "train_tokens_per_second": 12152.92
    },
    {
      "epoch": 1.2765957446808511,
      "grad_norm": 0.8879122428055022,
      "learning_rate": 2.895805884208378e-06,
      "loss": 0.30857186317443847,
      "num_input_tokens_seen": 604319416,
      "step": 18960,
      "train_runtime": 49726.4133,
      "train_tokens_per_second": 12152.886
    },
    {
      "epoch": 1.2769323996768112,
      "grad_norm": 0.9027865852267539,
      "learning_rate": 2.8934076379379016e-06,
      "loss": 0.3112833261489868,
      "num_input_tokens_seen": 604480848,
      "step": 18965,
      "train_runtime": 49739.3966,
      "train_tokens_per_second": 12152.959
    },
    {
      "epoch": 1.2772690546727714,
      "grad_norm": 0.7911404334024478,
      "learning_rate": 2.891009980769266e-06,
      "loss": 0.2947298526763916,
      "num_input_tokens_seen": 604641928,
      "step": 18970,
      "train_runtime": 49752.5836,
      "train_tokens_per_second": 12152.975
    },
    {
      "epoch": 1.2776057096687314,
      "grad_norm": 0.9564593228272439,
      "learning_rate": 2.88861291337296e-06,
      "loss": 0.3028806924819946,
      "num_input_tokens_seen": 604801224,
      "step": 18975,
      "train_runtime": 49766.4742,
      "train_tokens_per_second": 12152.784
    },
    {
      "epoch": 1.2779423646646917,
      "grad_norm": 0.7669301412939422,
      "learning_rate": 2.8862164364193236e-06,
      "loss": 0.28706295490264894,
      "num_input_tokens_seen": 604963816,
      "step": 18980,
      "train_runtime": 49779.4913,
      "train_tokens_per_second": 12152.873
    },
    {
      "epoch": 1.2782790196606517,
      "grad_norm": 0.7545436164403377,
      "learning_rate": 2.8838205505785168e-06,
      "loss": 0.3003144502639771,
      "num_input_tokens_seen": 605124856,
      "step": 18985,
      "train_runtime": 49792.561,
      "train_tokens_per_second": 12152.917
    },
    {
      "epoch": 1.278615674656612,
      "grad_norm": 0.7150653788656524,
      "learning_rate": 2.881425256520547e-06,
      "loss": 0.2822824716567993,
      "num_input_tokens_seen": 605282232,
      "step": 18990,
      "train_runtime": 49805.9721,
      "train_tokens_per_second": 12152.804
    },
    {
      "epoch": 1.278952329652572,
      "grad_norm": 0.7405072882818217,
      "learning_rate": 2.8790305549152487e-06,
      "loss": 0.2495818853378296,
      "num_input_tokens_seen": 605444896,
      "step": 18995,
      "train_runtime": 49818.8651,
      "train_tokens_per_second": 12152.924
    },
    {
      "epoch": 1.2792889846485322,
      "grad_norm": 0.7654979207467363,
      "learning_rate": 2.876636446432292e-06,
      "loss": 0.30813183784484866,
      "num_input_tokens_seen": 605605248,
      "step": 19000,
      "train_runtime": 49831.7749,
      "train_tokens_per_second": 12152.994
    },
    {
      "epoch": 1.2796256396444923,
      "grad_norm": 0.7730705415760324,
      "learning_rate": 2.8742429317411826e-06,
      "loss": 0.3068549156188965,
      "num_input_tokens_seen": 605766456,
      "step": 19005,
      "train_runtime": 49844.8748,
      "train_tokens_per_second": 12153.034
    },
    {
      "epoch": 1.2799622946404525,
      "grad_norm": 0.8680882410871714,
      "learning_rate": 2.8718500115112585e-06,
      "loss": 0.31152465343475344,
      "num_input_tokens_seen": 605920088,
      "step": 19010,
      "train_runtime": 49857.6089,
      "train_tokens_per_second": 12153.011
    },
    {
      "epoch": 1.2802989496364126,
      "grad_norm": 0.7703535693196939,
      "learning_rate": 2.869457686411691e-06,
      "loss": 0.3294264316558838,
      "num_input_tokens_seen": 606077856,
      "step": 19015,
      "train_runtime": 49870.1402,
      "train_tokens_per_second": 12153.121
    },
    {
      "epoch": 1.2806356046323728,
      "grad_norm": 0.8404786827537626,
      "learning_rate": 2.8670659571114912e-06,
      "loss": 0.3040056943893433,
      "num_input_tokens_seen": 606238224,
      "step": 19020,
      "train_runtime": 49882.6117,
      "train_tokens_per_second": 12153.298
    },
    {
      "epoch": 1.2809722596283328,
      "grad_norm": 0.6960599715562762,
      "learning_rate": 2.864674824279491e-06,
      "loss": 0.2875059127807617,
      "num_input_tokens_seen": 606398864,
      "step": 19025,
      "train_runtime": 49896.4929,
      "train_tokens_per_second": 12153.136
    },
    {
      "epoch": 1.281308914624293,
      "grad_norm": 0.789796602593778,
      "learning_rate": 2.862284288584371e-06,
      "loss": 0.2823715448379517,
      "num_input_tokens_seen": 606559352,
      "step": 19030,
      "train_runtime": 49909.4206,
      "train_tokens_per_second": 12153.204
    },
    {
      "epoch": 1.2816455696202531,
      "grad_norm": 0.8740623798328989,
      "learning_rate": 2.8598943506946297e-06,
      "loss": 0.3081759691238403,
      "num_input_tokens_seen": 606718928,
      "step": 19035,
      "train_runtime": 49921.9419,
      "train_tokens_per_second": 12153.352
    },
    {
      "epoch": 1.2819822246162134,
      "grad_norm": 0.7950820761611049,
      "learning_rate": 2.85750501127861e-06,
      "loss": 0.28339433670043945,
      "num_input_tokens_seen": 606878744,
      "step": 19040,
      "train_runtime": 49935.7061,
      "train_tokens_per_second": 12153.202
    },
    {
      "epoch": 1.2823188796121734,
      "grad_norm": 0.7385173275255058,
      "learning_rate": 2.8551162710044827e-06,
      "loss": 0.31591482162475587,
      "num_input_tokens_seen": 607038704,
      "step": 19045,
      "train_runtime": 49948.8706,
      "train_tokens_per_second": 12153.202
    },
    {
      "epoch": 1.2826555346081336,
      "grad_norm": 0.7876415246681366,
      "learning_rate": 2.8527281305402503e-06,
      "loss": 0.3255245923995972,
      "num_input_tokens_seen": 607200640,
      "step": 19050,
      "train_runtime": 49962.566,
      "train_tokens_per_second": 12153.112
    },
    {
      "epoch": 1.2829921896040937,
      "grad_norm": 0.9060905067376765,
      "learning_rate": 2.8503405905537496e-06,
      "loss": 0.29764814376831056,
      "num_input_tokens_seen": 607357368,
      "step": 19055,
      "train_runtime": 49976.2314,
      "train_tokens_per_second": 12152.925
    },
    {
      "epoch": 1.283328844600054,
      "grad_norm": 0.7441690653990919,
      "learning_rate": 2.847953651712647e-06,
      "loss": 0.301771354675293,
      "num_input_tokens_seen": 607519960,
      "step": 19060,
      "train_runtime": 49988.9337,
      "train_tokens_per_second": 12153.089
    },
    {
      "epoch": 1.283665499596014,
      "grad_norm": 0.7843632782648001,
      "learning_rate": 2.8455673146844426e-06,
      "loss": 0.2886895418167114,
      "num_input_tokens_seen": 607677216,
      "step": 19065,
      "train_runtime": 50002.7539,
      "train_tokens_per_second": 12152.875
    },
    {
      "epoch": 1.2840021545919742,
      "grad_norm": 1.0631864152436574,
      "learning_rate": 2.8431815801364717e-06,
      "loss": 0.32060041427612307,
      "num_input_tokens_seen": 607838120,
      "step": 19070,
      "train_runtime": 50015.6421,
      "train_tokens_per_second": 12152.96
    },
    {
      "epoch": 1.2843388095879342,
      "grad_norm": 0.8823375813971637,
      "learning_rate": 2.8407964487358906e-06,
      "loss": 0.30711042881011963,
      "num_input_tokens_seen": 607999704,
      "step": 19075,
      "train_runtime": 50028.1577,
      "train_tokens_per_second": 12153.15
    },
    {
      "epoch": 1.2846754645838945,
      "grad_norm": 0.7603407365169693,
      "learning_rate": 2.838411921149702e-06,
      "loss": 0.3008668661117554,
      "num_input_tokens_seen": 608153296,
      "step": 19080,
      "train_runtime": 50041.5861,
      "train_tokens_per_second": 12152.958
    },
    {
      "epoch": 1.2850121195798545,
      "grad_norm": 0.7722295259943268,
      "learning_rate": 2.836027998044722e-06,
      "loss": 0.28954713344573973,
      "num_input_tokens_seen": 608309552,
      "step": 19085,
      "train_runtime": 50055.1985,
      "train_tokens_per_second": 12152.775
    },
    {
      "epoch": 1.2853487745758148,
      "grad_norm": 0.8329708544279469,
      "learning_rate": 2.8336446800876143e-06,
      "loss": 0.2998910903930664,
      "num_input_tokens_seen": 608459888,
      "step": 19090,
      "train_runtime": 50068.9058,
      "train_tokens_per_second": 12152.45
    },
    {
      "epoch": 1.2856854295717748,
      "grad_norm": 0.8256074596429657,
      "learning_rate": 2.8312619679448637e-06,
      "loss": 0.29074892997741697,
      "num_input_tokens_seen": 608618928,
      "step": 19095,
      "train_runtime": 50081.9694,
      "train_tokens_per_second": 12152.456
    },
    {
      "epoch": 1.286022084567735,
      "grad_norm": 0.8655096259439734,
      "learning_rate": 2.8288798622827884e-06,
      "loss": 0.309846019744873,
      "num_input_tokens_seen": 608773120,
      "step": 19100,
      "train_runtime": 50095.267,
      "train_tokens_per_second": 12152.308
    },
    {
      "epoch": 1.286358739563695,
      "grad_norm": 0.9141284798833497,
      "learning_rate": 2.826498363767536e-06,
      "loss": 0.3036093235015869,
      "num_input_tokens_seen": 608926904,
      "step": 19105,
      "train_runtime": 50108.9935,
      "train_tokens_per_second": 12152.048
    },
    {
      "epoch": 1.2866953945596553,
      "grad_norm": 2.2249312562043397,
      "learning_rate": 2.8241174730650855e-06,
      "loss": 0.298044228553772,
      "num_input_tokens_seen": 609085864,
      "step": 19110,
      "train_runtime": 50121.9192,
      "train_tokens_per_second": 12152.086
    },
    {
      "epoch": 1.2870320495556153,
      "grad_norm": 0.8069750249784594,
      "learning_rate": 2.821737190841244e-06,
      "loss": 0.2968132019042969,
      "num_input_tokens_seen": 609246552,
      "step": 19115,
      "train_runtime": 50135.7197,
      "train_tokens_per_second": 12151.946
    },
    {
      "epoch": 1.2873687045515756,
      "grad_norm": 0.8220307299753202,
      "learning_rate": 2.8193575177616537e-06,
      "loss": 0.291119647026062,
      "num_input_tokens_seen": 609405408,
      "step": 19120,
      "train_runtime": 50148.48,
      "train_tokens_per_second": 12152.022
    },
    {
      "epoch": 1.2877053595475356,
      "grad_norm": 0.7963017861512874,
      "learning_rate": 2.816978454491777e-06,
      "loss": 0.2780735731124878,
      "num_input_tokens_seen": 609568344,
      "step": 19125,
      "train_runtime": 50161.2532,
      "train_tokens_per_second": 12152.175
    },
    {
      "epoch": 1.2880420145434959,
      "grad_norm": 0.8850018470467759,
      "learning_rate": 2.814600001696919e-06,
      "loss": 0.308332633972168,
      "num_input_tokens_seen": 609723592,
      "step": 19130,
      "train_runtime": 50173.7579,
      "train_tokens_per_second": 12152.241
    },
    {
      "epoch": 1.288378669539456,
      "grad_norm": 0.8427057723974886,
      "learning_rate": 2.812222160042196e-06,
      "loss": 0.3095693588256836,
      "num_input_tokens_seen": 609885136,
      "step": 19135,
      "train_runtime": 50186.725,
      "train_tokens_per_second": 12152.32
    },
    {
      "epoch": 1.2887153245354162,
      "grad_norm": 0.7686568729059485,
      "learning_rate": 2.809844930192573e-06,
      "loss": 0.3177642345428467,
      "num_input_tokens_seen": 610045128,
      "step": 19140,
      "train_runtime": 50199.4356,
      "train_tokens_per_second": 12152.43
    },
    {
      "epoch": 1.2890519795313762,
      "grad_norm": 0.7975989817940082,
      "learning_rate": 2.807468312812831e-06,
      "loss": 0.31854543685913084,
      "num_input_tokens_seen": 610201792,
      "step": 19145,
      "train_runtime": 50212.3643,
      "train_tokens_per_second": 12152.421
    },
    {
      "epoch": 1.2893886345273364,
      "grad_norm": 0.8114246268565122,
      "learning_rate": 2.805092308567584e-06,
      "loss": 0.2737419128417969,
      "num_input_tokens_seen": 610363648,
      "step": 19150,
      "train_runtime": 50225.4263,
      "train_tokens_per_second": 12152.483
    },
    {
      "epoch": 1.2897252895232965,
      "grad_norm": 0.8168499955312384,
      "learning_rate": 2.802716918121273e-06,
      "loss": 0.29413819313049316,
      "num_input_tokens_seen": 610527280,
      "step": 19155,
      "train_runtime": 50238.4574,
      "train_tokens_per_second": 12152.588
    },
    {
      "epoch": 1.2900619445192567,
      "grad_norm": 0.815486456903987,
      "learning_rate": 2.8003421421381704e-06,
      "loss": 0.28854918479919434,
      "num_input_tokens_seen": 610687872,
      "step": 19160,
      "train_runtime": 50251.5409,
      "train_tokens_per_second": 12152.62
    },
    {
      "epoch": 1.2903985995152167,
      "grad_norm": 0.8428913664999467,
      "learning_rate": 2.797967981282371e-06,
      "loss": 0.3136820077896118,
      "num_input_tokens_seen": 610848680,
      "step": 19165,
      "train_runtime": 50264.1563,
      "train_tokens_per_second": 12152.769
    },
    {
      "epoch": 1.290735254511177,
      "grad_norm": 0.762451034482807,
      "learning_rate": 2.795594436217808e-06,
      "loss": 0.29275429248809814,
      "num_input_tokens_seen": 611008352,
      "step": 19170,
      "train_runtime": 50276.6907,
      "train_tokens_per_second": 12152.915
    },
    {
      "epoch": 1.291071909507137,
      "grad_norm": 0.7392659555762797,
      "learning_rate": 2.7932215076082277e-06,
      "loss": 0.2877834320068359,
      "num_input_tokens_seen": 611162384,
      "step": 19175,
      "train_runtime": 50289.8402,
      "train_tokens_per_second": 12152.8
    },
    {
      "epoch": 1.2914085645030973,
      "grad_norm": 0.7950732434389722,
      "learning_rate": 2.790849196117219e-06,
      "loss": 0.29698824882507324,
      "num_input_tokens_seen": 611326224,
      "step": 19180,
      "train_runtime": 50302.265,
      "train_tokens_per_second": 12153.056
    },
    {
      "epoch": 1.2917452194990573,
      "grad_norm": 0.7809062013733077,
      "learning_rate": 2.7884775024081856e-06,
      "loss": 0.2998488426208496,
      "num_input_tokens_seen": 611484848,
      "step": 19185,
      "train_runtime": 50315.1103,
      "train_tokens_per_second": 12153.106
    },
    {
      "epoch": 1.2920818744950175,
      "grad_norm": 0.7726086980819804,
      "learning_rate": 2.786106427144367e-06,
      "loss": 0.29848670959472656,
      "num_input_tokens_seen": 611643720,
      "step": 19190,
      "train_runtime": 50327.9492,
      "train_tokens_per_second": 12153.162
    },
    {
      "epoch": 1.2924185294909776,
      "grad_norm": 0.7702712513183868,
      "learning_rate": 2.783735970988828e-06,
      "loss": 0.27229182720184325,
      "num_input_tokens_seen": 611804424,
      "step": 19195,
      "train_runtime": 50340.9842,
      "train_tokens_per_second": 12153.207
    },
    {
      "epoch": 1.2927551844869378,
      "grad_norm": 0.8588948549995771,
      "learning_rate": 2.7813661346044566e-06,
      "loss": 0.30161190032958984,
      "num_input_tokens_seen": 611964816,
      "step": 19200,
      "train_runtime": 50354.301,
      "train_tokens_per_second": 12153.179
    },
    {
      "epoch": 1.2930918394828979,
      "grad_norm": 0.854366312663946,
      "learning_rate": 2.7789969186539713e-06,
      "loss": 0.30611362457275393,
      "num_input_tokens_seen": 612123464,
      "step": 19205,
      "train_runtime": 50367.6036,
      "train_tokens_per_second": 12153.119
    },
    {
      "epoch": 1.293428494478858,
      "grad_norm": 0.8495031381467388,
      "learning_rate": 2.776628323799915e-06,
      "loss": 0.2994052886962891,
      "num_input_tokens_seen": 612278920,
      "step": 19210,
      "train_runtime": 50380.3975,
      "train_tokens_per_second": 12153.118
    },
    {
      "epoch": 1.2937651494748181,
      "grad_norm": 0.7975051903657048,
      "learning_rate": 2.774260350704657e-06,
      "loss": 0.3006717681884766,
      "num_input_tokens_seen": 612437832,
      "step": 19215,
      "train_runtime": 50393.2999,
      "train_tokens_per_second": 12153.16
    },
    {
      "epoch": 1.2941018044707784,
      "grad_norm": 0.9686123541201149,
      "learning_rate": 2.771893000030398e-06,
      "loss": 0.3138278007507324,
      "num_input_tokens_seen": 612596920,
      "step": 19220,
      "train_runtime": 50406.1435,
      "train_tokens_per_second": 12153.219
    },
    {
      "epoch": 1.2944384594667384,
      "grad_norm": 0.8461336070935334,
      "learning_rate": 2.7695262724391526e-06,
      "loss": 0.3118725299835205,
      "num_input_tokens_seen": 612754280,
      "step": 19225,
      "train_runtime": 50419.7391,
      "train_tokens_per_second": 12153.063
    },
    {
      "epoch": 1.2947751144626987,
      "grad_norm": 0.8898820073506066,
      "learning_rate": 2.767160168592777e-06,
      "loss": 0.35455176830291746,
      "num_input_tokens_seen": 612917944,
      "step": 19230,
      "train_runtime": 50432.7752,
      "train_tokens_per_second": 12153.167
    },
    {
      "epoch": 1.2951117694586587,
      "grad_norm": 0.8431737620252795,
      "learning_rate": 2.7647946891529355e-06,
      "loss": 0.3121018648147583,
      "num_input_tokens_seen": 613071920,
      "step": 19235,
      "train_runtime": 50446.6787,
      "train_tokens_per_second": 12152.87
    },
    {
      "epoch": 1.295448424454619,
      "grad_norm": 0.7761520879756472,
      "learning_rate": 2.7624298347811342e-06,
      "loss": 0.30832886695861816,
      "num_input_tokens_seen": 613230552,
      "step": 19240,
      "train_runtime": 50460.5808,
      "train_tokens_per_second": 12152.665
    },
    {
      "epoch": 1.295785079450579,
      "grad_norm": 0.902232284504158,
      "learning_rate": 2.7600656061386945e-06,
      "loss": 0.32223007678985593,
      "num_input_tokens_seen": 613389944,
      "step": 19245,
      "train_runtime": 50473.2147,
      "train_tokens_per_second": 12152.781
    },
    {
      "epoch": 1.2961217344465392,
      "grad_norm": 0.8074586172047405,
      "learning_rate": 2.757702003886765e-06,
      "loss": 0.299889063835144,
      "num_input_tokens_seen": 613549304,
      "step": 19250,
      "train_runtime": 50487.7717,
      "train_tokens_per_second": 12152.434
    },
    {
      "epoch": 1.2964583894424992,
      "grad_norm": 0.8467134719842135,
      "learning_rate": 2.7553390286863205e-06,
      "loss": 0.29910955429077146,
      "num_input_tokens_seen": 613705872,
      "step": 19255,
      "train_runtime": 50501.9411,
      "train_tokens_per_second": 12152.124
    },
    {
      "epoch": 1.2967950444384595,
      "grad_norm": 0.7975895517979533,
      "learning_rate": 2.752976681198159e-06,
      "loss": 0.30665476322174073,
      "num_input_tokens_seen": 613867360,
      "step": 19260,
      "train_runtime": 50515.0978,
      "train_tokens_per_second": 12152.156
    },
    {
      "epoch": 1.2971316994344195,
      "grad_norm": 0.8644252621435792,
      "learning_rate": 2.750614962082901e-06,
      "loss": 0.29886689186096194,
      "num_input_tokens_seen": 614030536,
      "step": 19265,
      "train_runtime": 50528.1055,
      "train_tokens_per_second": 12152.257
    },
    {
      "epoch": 1.2974683544303798,
      "grad_norm": 0.705007809844978,
      "learning_rate": 2.748253872001e-06,
      "loss": 0.28118033409118653,
      "num_input_tokens_seen": 614187416,
      "step": 19270,
      "train_runtime": 50540.7303,
      "train_tokens_per_second": 12152.326
    },
    {
      "epoch": 1.2978050094263398,
      "grad_norm": 0.8645902514800793,
      "learning_rate": 2.74589341161272e-06,
      "loss": 0.30217299461364744,
      "num_input_tokens_seen": 614342592,
      "step": 19275,
      "train_runtime": 50554.0926,
      "train_tokens_per_second": 12152.183
    },
    {
      "epoch": 1.2981416644223,
      "grad_norm": 0.8171819188549311,
      "learning_rate": 2.743533581578163e-06,
      "loss": 0.30187819004058836,
      "num_input_tokens_seen": 614500200,
      "step": 19280,
      "train_runtime": 50566.9275,
      "train_tokens_per_second": 12152.215
    },
    {
      "epoch": 1.29847831941826,
      "grad_norm": 0.9786296752176804,
      "learning_rate": 2.7411743825572407e-06,
      "loss": 0.3500667572021484,
      "num_input_tokens_seen": 614663568,
      "step": 19285,
      "train_runtime": 50579.9223,
      "train_tokens_per_second": 12152.323
    },
    {
      "epoch": 1.2988149744142203,
      "grad_norm": 0.7927490365153944,
      "learning_rate": 2.7388158152097015e-06,
      "loss": 0.29631524085998534,
      "num_input_tokens_seen": 614826240,
      "step": 19290,
      "train_runtime": 50593.0132,
      "train_tokens_per_second": 12152.394
    },
    {
      "epoch": 1.2991516294101804,
      "grad_norm": 0.7632317250757582,
      "learning_rate": 2.736457880195109e-06,
      "loss": 0.31474192142486573,
      "num_input_tokens_seen": 614985344,
      "step": 19295,
      "train_runtime": 50606.1471,
      "train_tokens_per_second": 12152.384
    },
    {
      "epoch": 1.2994882844061406,
      "grad_norm": 0.8405221259363495,
      "learning_rate": 2.7341005781728523e-06,
      "loss": 0.33559064865112304,
      "num_input_tokens_seen": 615146608,
      "step": 19300,
      "train_runtime": 50619.2257,
      "train_tokens_per_second": 12152.43
    },
    {
      "epoch": 1.2998249394021006,
      "grad_norm": 0.8294104271001096,
      "learning_rate": 2.7317439098021444e-06,
      "loss": 0.30346877574920655,
      "num_input_tokens_seen": 615303640,
      "step": 19305,
      "train_runtime": 50632.0397,
      "train_tokens_per_second": 12152.456
    },
    {
      "epoch": 1.300161594398061,
      "grad_norm": 0.794912913133469,
      "learning_rate": 2.7293878757420185e-06,
      "loss": 0.30107893943786623,
      "num_input_tokens_seen": 615462952,
      "step": 19310,
      "train_runtime": 50644.6471,
      "train_tokens_per_second": 12152.577
    },
    {
      "epoch": 1.3004982493940211,
      "grad_norm": 0.772799828704457,
      "learning_rate": 2.7270324766513306e-06,
      "loss": 0.3015410900115967,
      "num_input_tokens_seen": 615624992,
      "step": 19315,
      "train_runtime": 50657.213,
      "train_tokens_per_second": 12152.761
    },
    {
      "epoch": 1.3008349043899812,
      "grad_norm": 0.78290767832286,
      "learning_rate": 2.724677713188767e-06,
      "loss": 0.3036373138427734,
      "num_input_tokens_seen": 615783792,
      "step": 19320,
      "train_runtime": 50670.2435,
      "train_tokens_per_second": 12152.77
    },
    {
      "epoch": 1.3011715593859412,
      "grad_norm": 0.9032441620865872,
      "learning_rate": 2.722323586012821e-06,
      "loss": 0.3068065643310547,
      "num_input_tokens_seen": 615946736,
      "step": 19325,
      "train_runtime": 50682.9429,
      "train_tokens_per_second": 12152.939
    },
    {
      "epoch": 1.3015082143819015,
      "grad_norm": 0.8049017214690037,
      "learning_rate": 2.7199700957818252e-06,
      "loss": 0.28746345043182375,
      "num_input_tokens_seen": 616108776,
      "step": 19330,
      "train_runtime": 50695.95,
      "train_tokens_per_second": 12153.018
    },
    {
      "epoch": 1.3018448693778617,
      "grad_norm": 0.8606417673000492,
      "learning_rate": 2.717617243153918e-06,
      "loss": 0.3037557601928711,
      "num_input_tokens_seen": 616265184,
      "step": 19335,
      "train_runtime": 50708.3794,
      "train_tokens_per_second": 12153.123
    },
    {
      "epoch": 1.3021815243738217,
      "grad_norm": 0.7820839143729765,
      "learning_rate": 2.715265028787073e-06,
      "loss": 0.3088447093963623,
      "num_input_tokens_seen": 616419448,
      "step": 19340,
      "train_runtime": 50721.2854,
      "train_tokens_per_second": 12153.072
    },
    {
      "epoch": 1.3025181793697818,
      "grad_norm": 0.9279179195378255,
      "learning_rate": 2.7129134533390765e-06,
      "loss": 0.331957745552063,
      "num_input_tokens_seen": 616582184,
      "step": 19345,
      "train_runtime": 50733.7158,
      "train_tokens_per_second": 12153.302
    },
    {
      "epoch": 1.302854834365742,
      "grad_norm": 0.8795953317221421,
      "learning_rate": 2.7105625174675404e-06,
      "loss": 0.2849281311035156,
      "num_input_tokens_seen": 616743768,
      "step": 19350,
      "train_runtime": 50746.2301,
      "train_tokens_per_second": 12153.489
    },
    {
      "epoch": 1.3031914893617023,
      "grad_norm": 0.8024830172412205,
      "learning_rate": 2.7082122218298957e-06,
      "loss": 0.29079256057739256,
      "num_input_tokens_seen": 616899512,
      "step": 19355,
      "train_runtime": 50759.5112,
      "train_tokens_per_second": 12153.378
    },
    {
      "epoch": 1.3035281443576623,
      "grad_norm": 0.8216082117798608,
      "learning_rate": 2.7058625670833947e-06,
      "loss": 0.28796977996826173,
      "num_input_tokens_seen": 617052856,
      "step": 19360,
      "train_runtime": 50772.1524,
      "train_tokens_per_second": 12153.372
    },
    {
      "epoch": 1.3038647993536223,
      "grad_norm": 0.84879494773889,
      "learning_rate": 2.70351355388511e-06,
      "loss": 0.31571574211120607,
      "num_input_tokens_seen": 617208952,
      "step": 19365,
      "train_runtime": 50785.2184,
      "train_tokens_per_second": 12153.319
    },
    {
      "epoch": 1.3042014543495826,
      "grad_norm": 0.8084811549012915,
      "learning_rate": 2.7011651828919407e-06,
      "loss": 0.3083205699920654,
      "num_input_tokens_seen": 617368432,
      "step": 19370,
      "train_runtime": 50798.1183,
      "train_tokens_per_second": 12153.372
    },
    {
      "epoch": 1.3045381093455428,
      "grad_norm": 0.899456326303277,
      "learning_rate": 2.698817454760593e-06,
      "loss": 0.31144065856933595,
      "num_input_tokens_seen": 617522760,
      "step": 19375,
      "train_runtime": 50810.8754,
      "train_tokens_per_second": 12153.358
    },
    {
      "epoch": 1.3048747643415028,
      "grad_norm": 0.7874306599097675,
      "learning_rate": 2.696470370147611e-06,
      "loss": 0.30310258865356443,
      "num_input_tokens_seen": 617682576,
      "step": 19380,
      "train_runtime": 50823.7822,
      "train_tokens_per_second": 12153.416
    },
    {
      "epoch": 1.3052114193374629,
      "grad_norm": 0.817462490309342,
      "learning_rate": 2.69412392970934e-06,
      "loss": 0.33252105712890623,
      "num_input_tokens_seen": 617842024,
      "step": 19385,
      "train_runtime": 50837.0107,
      "train_tokens_per_second": 12153.39
    },
    {
      "epoch": 1.3055480743334231,
      "grad_norm": 0.7180311841570963,
      "learning_rate": 2.6917781341019615e-06,
      "loss": 0.2898344278335571,
      "num_input_tokens_seen": 618001464,
      "step": 19390,
      "train_runtime": 50850.1063,
      "train_tokens_per_second": 12153.396
    },
    {
      "epoch": 1.3058847293293834,
      "grad_norm": 0.7869603056004898,
      "learning_rate": 2.689432983981467e-06,
      "loss": 0.2947995185852051,
      "num_input_tokens_seen": 618163800,
      "step": 19395,
      "train_runtime": 50862.7308,
      "train_tokens_per_second": 12153.571
    },
    {
      "epoch": 1.3062213843253434,
      "grad_norm": 0.7256848575572841,
      "learning_rate": 2.6870884800036713e-06,
      "loss": 0.2963334798812866,
      "num_input_tokens_seen": 618323008,
      "step": 19400,
      "train_runtime": 50875.6808,
      "train_tokens_per_second": 12153.607
    },
    {
      "epoch": 1.3065580393213034,
      "grad_norm": 0.8359624181575982,
      "learning_rate": 2.684744622824207e-06,
      "loss": 0.32390923500061036,
      "num_input_tokens_seen": 618476008,
      "step": 19405,
      "train_runtime": 50888.4903,
      "train_tokens_per_second": 12153.554
    },
    {
      "epoch": 1.3068946943172637,
      "grad_norm": 0.7561296037628713,
      "learning_rate": 2.6824014130985255e-06,
      "loss": 0.30150132179260253,
      "num_input_tokens_seen": 618634536,
      "step": 19410,
      "train_runtime": 50901.3015,
      "train_tokens_per_second": 12153.609
    },
    {
      "epoch": 1.307231349313224,
      "grad_norm": 0.7625068799720057,
      "learning_rate": 2.680058851481896e-06,
      "loss": 0.28853092193603513,
      "num_input_tokens_seen": 618791032,
      "step": 19415,
      "train_runtime": 50914.2188,
      "train_tokens_per_second": 12153.6
    },
    {
      "epoch": 1.307568004309184,
      "grad_norm": 0.9086420844998367,
      "learning_rate": 2.6777169386294154e-06,
      "loss": 0.3090104103088379,
      "num_input_tokens_seen": 618947568,
      "step": 19420,
      "train_runtime": 50927.1061,
      "train_tokens_per_second": 12153.598
    },
    {
      "epoch": 1.307904659305144,
      "grad_norm": 1.1208392053742278,
      "learning_rate": 2.6753756751959846e-06,
      "loss": 0.30269370079040525,
      "num_input_tokens_seen": 619108664,
      "step": 19425,
      "train_runtime": 50939.577,
      "train_tokens_per_second": 12153.785
    },
    {
      "epoch": 1.3082413143011042,
      "grad_norm": 0.7727990872657099,
      "learning_rate": 2.6730350618363367e-06,
      "loss": 0.2951890707015991,
      "num_input_tokens_seen": 619269176,
      "step": 19430,
      "train_runtime": 50953.2194,
      "train_tokens_per_second": 12153.681
    },
    {
      "epoch": 1.3085779692970645,
      "grad_norm": 0.7683100655665543,
      "learning_rate": 2.6706950992050097e-06,
      "loss": 0.3051133155822754,
      "num_input_tokens_seen": 619425720,
      "step": 19435,
      "train_runtime": 50966.1333,
      "train_tokens_per_second": 12153.673
    },
    {
      "epoch": 1.3089146242930245,
      "grad_norm": 0.8365023474882791,
      "learning_rate": 2.6683557879563726e-06,
      "loss": 0.3211726188659668,
      "num_input_tokens_seen": 619585240,
      "step": 19440,
      "train_runtime": 50979.6561,
      "train_tokens_per_second": 12153.578
    },
    {
      "epoch": 1.3092512792889845,
      "grad_norm": 0.797114249011439,
      "learning_rate": 2.6660171287446045e-06,
      "loss": 0.3359297513961792,
      "num_input_tokens_seen": 619746104,
      "step": 19445,
      "train_runtime": 50992.4773,
      "train_tokens_per_second": 12153.677
    },
    {
      "epoch": 1.3095879342849448,
      "grad_norm": 0.8414524313814291,
      "learning_rate": 2.6636791222237034e-06,
      "loss": 0.3211383819580078,
      "num_input_tokens_seen": 619905264,
      "step": 19450,
      "train_runtime": 51005.4939,
      "train_tokens_per_second": 12153.696
    },
    {
      "epoch": 1.309924589280905,
      "grad_norm": 0.8330208898157156,
      "learning_rate": 2.6613417690474864e-06,
      "loss": 0.29979941844940183,
      "num_input_tokens_seen": 620063264,
      "step": 19455,
      "train_runtime": 51018.4005,
      "train_tokens_per_second": 12153.718
    },
    {
      "epoch": 1.310261244276865,
      "grad_norm": 0.8253621326976531,
      "learning_rate": 2.6590050698695856e-06,
      "loss": 0.3024393081665039,
      "num_input_tokens_seen": 620220976,
      "step": 19460,
      "train_runtime": 51030.8188,
      "train_tokens_per_second": 12153.851
    },
    {
      "epoch": 1.310597899272825,
      "grad_norm": 0.7428989443394158,
      "learning_rate": 2.6566690253434508e-06,
      "loss": 0.3112042903900146,
      "num_input_tokens_seen": 620383056,
      "step": 19465,
      "train_runtime": 51043.4074,
      "train_tokens_per_second": 12154.029
    },
    {
      "epoch": 1.3109345542687854,
      "grad_norm": 0.7885880659010636,
      "learning_rate": 2.654333636122354e-06,
      "loss": 0.28822941780090333,
      "num_input_tokens_seen": 620544160,
      "step": 19470,
      "train_runtime": 51056.3785,
      "train_tokens_per_second": 12154.097
    },
    {
      "epoch": 1.3112712092647456,
      "grad_norm": 0.7905743392699995,
      "learning_rate": 2.6519989028593728e-06,
      "loss": 0.3059583902359009,
      "num_input_tokens_seen": 620705704,
      "step": 19475,
      "train_runtime": 51068.8412,
      "train_tokens_per_second": 12154.294
    },
    {
      "epoch": 1.3116078642607056,
      "grad_norm": 0.7745742693211273,
      "learning_rate": 2.649664826207414e-06,
      "loss": 0.3137577295303345,
      "num_input_tokens_seen": 620866024,
      "step": 19480,
      "train_runtime": 51081.7664,
      "train_tokens_per_second": 12154.357
    },
    {
      "epoch": 1.3119445192566657,
      "grad_norm": 0.82953354582996,
      "learning_rate": 2.647331406819189e-06,
      "loss": 0.3231818199157715,
      "num_input_tokens_seen": 621025552,
      "step": 19485,
      "train_runtime": 51095.7897,
      "train_tokens_per_second": 12154.143
    },
    {
      "epoch": 1.312281174252626,
      "grad_norm": 0.707543081543998,
      "learning_rate": 2.644998645347236e-06,
      "loss": 0.28951272964477537,
      "num_input_tokens_seen": 621183464,
      "step": 19490,
      "train_runtime": 51109.0852,
      "train_tokens_per_second": 12154.071
    },
    {
      "epoch": 1.3126178292485862,
      "grad_norm": 0.9121508647342764,
      "learning_rate": 2.6426665424439025e-06,
      "loss": 0.3198955297470093,
      "num_input_tokens_seen": 621346088,
      "step": 19495,
      "train_runtime": 51121.7112,
      "train_tokens_per_second": 12154.251
    },
    {
      "epoch": 1.3129544842445462,
      "grad_norm": 0.8574420271957544,
      "learning_rate": 2.6403350987613526e-06,
      "loss": 0.30613884925842283,
      "num_input_tokens_seen": 621506320,
      "step": 19500,
      "train_runtime": 51134.5718,
      "train_tokens_per_second": 12154.327
    },
    {
      "epoch": 1.3132911392405062,
      "grad_norm": 0.8503916968253756,
      "learning_rate": 2.6380043149515676e-06,
      "loss": 0.3064993381500244,
      "num_input_tokens_seen": 621663760,
      "step": 19505,
      "train_runtime": 51148.0516,
      "train_tokens_per_second": 12154.202
    },
    {
      "epoch": 1.3136277942364665,
      "grad_norm": 0.86049677615533,
      "learning_rate": 2.6356741916663453e-06,
      "loss": 0.30734429359436033,
      "num_input_tokens_seen": 621826168,
      "step": 19510,
      "train_runtime": 51161.0489,
      "train_tokens_per_second": 12154.289
    },
    {
      "epoch": 1.3139644492324267,
      "grad_norm": 0.8533114870241088,
      "learning_rate": 2.6333447295572933e-06,
      "loss": 0.32071857452392577,
      "num_input_tokens_seen": 621988744,
      "step": 19515,
      "train_runtime": 51174.3456,
      "train_tokens_per_second": 12154.308
    },
    {
      "epoch": 1.3143011042283868,
      "grad_norm": 0.7663248817016668,
      "learning_rate": 2.631015929275845e-06,
      "loss": 0.30270018577575686,
      "num_input_tokens_seen": 622145832,
      "step": 19520,
      "train_runtime": 51187.5188,
      "train_tokens_per_second": 12154.249
    },
    {
      "epoch": 1.3146377592243468,
      "grad_norm": 0.896019977610222,
      "learning_rate": 2.628687791473234e-06,
      "loss": 0.28989624977111816,
      "num_input_tokens_seen": 622306016,
      "step": 19525,
      "train_runtime": 51200.4916,
      "train_tokens_per_second": 12154.298
    },
    {
      "epoch": 1.314974414220307,
      "grad_norm": 0.860712890152173,
      "learning_rate": 2.6263603168005224e-06,
      "loss": 0.2762287616729736,
      "num_input_tokens_seen": 622467576,
      "step": 19530,
      "train_runtime": 51213.0121,
      "train_tokens_per_second": 12154.481
    },
    {
      "epoch": 1.3153110692162673,
      "grad_norm": 0.7509866742864634,
      "learning_rate": 2.6240335059085797e-06,
      "loss": 0.30350775718688966,
      "num_input_tokens_seen": 622628464,
      "step": 19535,
      "train_runtime": 51226.1602,
      "train_tokens_per_second": 12154.502
    },
    {
      "epoch": 1.3156477242122273,
      "grad_norm": 0.9272330754229331,
      "learning_rate": 2.62170735944809e-06,
      "loss": 0.3107863664627075,
      "num_input_tokens_seen": 622786016,
      "step": 19540,
      "train_runtime": 51239.5308,
      "train_tokens_per_second": 12154.405
    },
    {
      "epoch": 1.3159843792081873,
      "grad_norm": 0.7967580808965196,
      "learning_rate": 2.6193818780695537e-06,
      "loss": 0.29716806411743163,
      "num_input_tokens_seen": 622945112,
      "step": 19545,
      "train_runtime": 51252.3172,
      "train_tokens_per_second": 12154.477
    },
    {
      "epoch": 1.3163210342041476,
      "grad_norm": 0.9051371337678246,
      "learning_rate": 2.6170570624232837e-06,
      "loss": 0.3097166776657104,
      "num_input_tokens_seen": 623097736,
      "step": 19550,
      "train_runtime": 51265.6402,
      "train_tokens_per_second": 12154.295
    },
    {
      "epoch": 1.3166576892001078,
      "grad_norm": 0.823702453692766,
      "learning_rate": 2.6147329131594078e-06,
      "loss": 0.290910530090332,
      "num_input_tokens_seen": 623260688,
      "step": 19555,
      "train_runtime": 51278.4662,
      "train_tokens_per_second": 12154.433
    },
    {
      "epoch": 1.3169943441960679,
      "grad_norm": 0.9126620153570457,
      "learning_rate": 2.612409430927867e-06,
      "loss": 0.2691087245941162,
      "num_input_tokens_seen": 623421912,
      "step": 19560,
      "train_runtime": 51291.4874,
      "train_tokens_per_second": 12154.491
    },
    {
      "epoch": 1.317330999192028,
      "grad_norm": 1.034136265084071,
      "learning_rate": 2.610086616378413e-06,
      "loss": 0.27335002422332766,
      "num_input_tokens_seen": 623578456,
      "step": 19565,
      "train_runtime": 51305.0401,
      "train_tokens_per_second": 12154.331
    },
    {
      "epoch": 1.3176676541879881,
      "grad_norm": 0.7594328571843869,
      "learning_rate": 2.60776447016062e-06,
      "loss": 0.30710887908935547,
      "num_input_tokens_seen": 623741128,
      "step": 19570,
      "train_runtime": 51318.0715,
      "train_tokens_per_second": 12154.415
    },
    {
      "epoch": 1.3180043091839484,
      "grad_norm": 0.8750229927937976,
      "learning_rate": 2.6054429929238607e-06,
      "loss": 0.31640024185180665,
      "num_input_tokens_seen": 623895912,
      "step": 19575,
      "train_runtime": 51331.1475,
      "train_tokens_per_second": 12154.334
    },
    {
      "epoch": 1.3183409641799084,
      "grad_norm": 0.7169821376224397,
      "learning_rate": 2.603122185317335e-06,
      "loss": 0.28309829235076905,
      "num_input_tokens_seen": 624052808,
      "step": 19580,
      "train_runtime": 51344.8004,
      "train_tokens_per_second": 12154.158
    },
    {
      "epoch": 1.3186776191758685,
      "grad_norm": 0.8084567445019526,
      "learning_rate": 2.6008020479900465e-06,
      "loss": 0.2615352630615234,
      "num_input_tokens_seen": 624213840,
      "step": 19585,
      "train_runtime": 51358.0629,
      "train_tokens_per_second": 12154.155
    },
    {
      "epoch": 1.3190142741718287,
      "grad_norm": 0.7788142530057693,
      "learning_rate": 2.598482581590815e-06,
      "loss": 0.2996666431427002,
      "num_input_tokens_seen": 624375832,
      "step": 19590,
      "train_runtime": 51370.6315,
      "train_tokens_per_second": 12154.334
    },
    {
      "epoch": 1.319350929167789,
      "grad_norm": 0.776199913372685,
      "learning_rate": 2.5961637867682715e-06,
      "loss": 0.27076921463012693,
      "num_input_tokens_seen": 624538968,
      "step": 19595,
      "train_runtime": 51383.4589,
      "train_tokens_per_second": 12154.475
    },
    {
      "epoch": 1.319687584163749,
      "grad_norm": 0.7731871966869023,
      "learning_rate": 2.5938456641708598e-06,
      "loss": 0.27604074478149415,
      "num_input_tokens_seen": 624694792,
      "step": 19600,
      "train_runtime": 51395.917,
      "train_tokens_per_second": 12154.561
    },
    {
      "epoch": 1.320024239159709,
      "grad_norm": 0.8869622761454629,
      "learning_rate": 2.591528214446836e-06,
      "loss": 0.2942762613296509,
      "num_input_tokens_seen": 624857320,
      "step": 19605,
      "train_runtime": 51408.6066,
      "train_tokens_per_second": 12154.722
    },
    {
      "epoch": 1.3203608941556693,
      "grad_norm": 0.8319141741007687,
      "learning_rate": 2.5892114382442664e-06,
      "loss": 0.3119437456130981,
      "num_input_tokens_seen": 625018504,
      "step": 19610,
      "train_runtime": 51422.2564,
      "train_tokens_per_second": 12154.63
    },
    {
      "epoch": 1.3206975491516295,
      "grad_norm": 0.7453286931899984,
      "learning_rate": 2.5868953362110287e-06,
      "loss": 0.3089602947235107,
      "num_input_tokens_seen": 625176808,
      "step": 19615,
      "train_runtime": 51434.7773,
      "train_tokens_per_second": 12154.749
    },
    {
      "epoch": 1.3210342041475895,
      "grad_norm": 0.7695081180328626,
      "learning_rate": 2.5845799089948196e-06,
      "loss": 0.3006046533584595,
      "num_input_tokens_seen": 625334984,
      "step": 19620,
      "train_runtime": 51448.1783,
      "train_tokens_per_second": 12154.657
    },
    {
      "epoch": 1.3213708591435496,
      "grad_norm": 0.8313242401041385,
      "learning_rate": 2.5822651572431323e-06,
      "loss": 0.28742949962615966,
      "num_input_tokens_seen": 625494304,
      "step": 19625,
      "train_runtime": 51460.6615,
      "train_tokens_per_second": 12154.805
    },
    {
      "epoch": 1.3217075141395098,
      "grad_norm": 0.8923232472036869,
      "learning_rate": 2.5799510816032867e-06,
      "loss": 0.3045827388763428,
      "num_input_tokens_seen": 625646248,
      "step": 19630,
      "train_runtime": 51474.6109,
      "train_tokens_per_second": 12154.463
    },
    {
      "epoch": 1.32204416913547,
      "grad_norm": 0.8206765679477448,
      "learning_rate": 2.577637682722404e-06,
      "loss": 0.3064920663833618,
      "num_input_tokens_seen": 625805272,
      "step": 19635,
      "train_runtime": 51488.0176,
      "train_tokens_per_second": 12154.387
    },
    {
      "epoch": 1.32238082413143,
      "grad_norm": 0.9669158006397705,
      "learning_rate": 2.5753249612474195e-06,
      "loss": 0.31276593208312986,
      "num_input_tokens_seen": 625965696,
      "step": 19640,
      "train_runtime": 51501.534,
      "train_tokens_per_second": 12154.312
    },
    {
      "epoch": 1.3227174791273901,
      "grad_norm": 0.7936738656190904,
      "learning_rate": 2.5730129178250774e-06,
      "loss": 0.308038067817688,
      "num_input_tokens_seen": 626127968,
      "step": 19645,
      "train_runtime": 51513.9695,
      "train_tokens_per_second": 12154.528
    },
    {
      "epoch": 1.3230541341233504,
      "grad_norm": 0.7292287721186009,
      "learning_rate": 2.570701553101933e-06,
      "loss": 0.29520750045776367,
      "num_input_tokens_seen": 626287800,
      "step": 19650,
      "train_runtime": 51527.5551,
      "train_tokens_per_second": 12154.425
    },
    {
      "epoch": 1.3233907891193106,
      "grad_norm": 0.810933398662699,
      "learning_rate": 2.568390867724354e-06,
      "loss": 0.3198778867721558,
      "num_input_tokens_seen": 626447696,
      "step": 19655,
      "train_runtime": 51540.1949,
      "train_tokens_per_second": 12154.547
    },
    {
      "epoch": 1.3237274441152707,
      "grad_norm": 0.8893684432004132,
      "learning_rate": 2.5660808623385143e-06,
      "loss": 0.30401718616485596,
      "num_input_tokens_seen": 626604976,
      "step": 19660,
      "train_runtime": 51553.1487,
      "train_tokens_per_second": 12154.543
    },
    {
      "epoch": 1.3240640991112307,
      "grad_norm": 0.737504220497609,
      "learning_rate": 2.5637715375903984e-06,
      "loss": 0.27608156204223633,
      "num_input_tokens_seen": 626763904,
      "step": 19665,
      "train_runtime": 51565.6547,
      "train_tokens_per_second": 12154.678
    },
    {
      "epoch": 1.324400754107191,
      "grad_norm": 0.8679580728059286,
      "learning_rate": 2.5614628941258075e-06,
      "loss": 0.28482856750488283,
      "num_input_tokens_seen": 626915392,
      "step": 19670,
      "train_runtime": 51578.7953,
      "train_tokens_per_second": 12154.518
    },
    {
      "epoch": 1.3247374091031512,
      "grad_norm": 0.8926079342176564,
      "learning_rate": 2.5591549325903388e-06,
      "loss": 0.30926659107208254,
      "num_input_tokens_seen": 627078840,
      "step": 19675,
      "train_runtime": 51591.2308,
      "train_tokens_per_second": 12154.756
    },
    {
      "epoch": 1.3250740640991112,
      "grad_norm": 0.8225348615484551,
      "learning_rate": 2.5568476536294105e-06,
      "loss": 0.3154122829437256,
      "num_input_tokens_seen": 627240232,
      "step": 19680,
      "train_runtime": 51605.0014,
      "train_tokens_per_second": 12154.64
    },
    {
      "epoch": 1.3254107190950712,
      "grad_norm": 0.8328104218677164,
      "learning_rate": 2.5545410578882466e-06,
      "loss": 0.2782647132873535,
      "num_input_tokens_seen": 627400376,
      "step": 19685,
      "train_runtime": 51617.4429,
      "train_tokens_per_second": 12154.813
    },
    {
      "epoch": 1.3257473740910315,
      "grad_norm": 0.9054075095938547,
      "learning_rate": 2.5522351460118776e-06,
      "loss": 0.28180980682373047,
      "num_input_tokens_seen": 627552576,
      "step": 19690,
      "train_runtime": 51630.7248,
      "train_tokens_per_second": 12154.634
    },
    {
      "epoch": 1.3260840290869917,
      "grad_norm": 0.759290983151401,
      "learning_rate": 2.549929918645144e-06,
      "loss": 0.3023005485534668,
      "num_input_tokens_seen": 627713032,
      "step": 19695,
      "train_runtime": 51643.1624,
      "train_tokens_per_second": 12154.814
    },
    {
      "epoch": 1.3264206840829518,
      "grad_norm": 0.7819641715403917,
      "learning_rate": 2.5476253764326963e-06,
      "loss": 0.28793699741363527,
      "num_input_tokens_seen": 627868016,
      "step": 19700,
      "train_runtime": 51656.2834,
      "train_tokens_per_second": 12154.727
    },
    {
      "epoch": 1.3267573390789118,
      "grad_norm": 0.8146269629700942,
      "learning_rate": 2.5453215200189927e-06,
      "loss": 0.29643688201904295,
      "num_input_tokens_seen": 628026096,
      "step": 19705,
      "train_runtime": 51669.4113,
      "train_tokens_per_second": 12154.698
    },
    {
      "epoch": 1.327093994074872,
      "grad_norm": 0.792352710196902,
      "learning_rate": 2.543018350048298e-06,
      "loss": 0.28310070037841795,
      "num_input_tokens_seen": 628185696,
      "step": 19710,
      "train_runtime": 51682.0675,
      "train_tokens_per_second": 12154.81
    },
    {
      "epoch": 1.3274306490708323,
      "grad_norm": 0.8792669158722032,
      "learning_rate": 2.540715867164685e-06,
      "loss": 0.31339011192321775,
      "num_input_tokens_seen": 628347392,
      "step": 19715,
      "train_runtime": 51695.7946,
      "train_tokens_per_second": 12154.71
    },
    {
      "epoch": 1.3277673040667923,
      "grad_norm": 0.7581743801224934,
      "learning_rate": 2.538414072012042e-06,
      "loss": 0.3200429201126099,
      "num_input_tokens_seen": 628505816,
      "step": 19720,
      "train_runtime": 51708.8831,
      "train_tokens_per_second": 12154.697
    },
    {
      "epoch": 1.3281039590627524,
      "grad_norm": 0.9109197326763242,
      "learning_rate": 2.53611296523405e-06,
      "loss": 0.31004900932312013,
      "num_input_tokens_seen": 628662232,
      "step": 19725,
      "train_runtime": 51723.0129,
      "train_tokens_per_second": 12154.401
    },
    {
      "epoch": 1.3284406140587126,
      "grad_norm": 0.8503272190874704,
      "learning_rate": 2.5338125474742127e-06,
      "loss": 0.2877772331237793,
      "num_input_tokens_seen": 628818200,
      "step": 19730,
      "train_runtime": 51735.6435,
      "train_tokens_per_second": 12154.448
    },
    {
      "epoch": 1.3287772690546729,
      "grad_norm": 0.7674807669742923,
      "learning_rate": 2.531512819375832e-06,
      "loss": 0.2906970024108887,
      "num_input_tokens_seen": 628978024,
      "step": 19735,
      "train_runtime": 51748.1296,
      "train_tokens_per_second": 12154.604
    },
    {
      "epoch": 1.3291139240506329,
      "grad_norm": 0.786402406831615,
      "learning_rate": 2.52921378158202e-06,
      "loss": 0.2792159080505371,
      "num_input_tokens_seen": 629132712,
      "step": 19740,
      "train_runtime": 51761.6167,
      "train_tokens_per_second": 12154.425
    },
    {
      "epoch": 1.329450579046593,
      "grad_norm": 0.826128388834396,
      "learning_rate": 2.5269154347356957e-06,
      "loss": 0.27422029972076417,
      "num_input_tokens_seen": 629291504,
      "step": 19745,
      "train_runtime": 51774.3129,
      "train_tokens_per_second": 12154.512
    },
    {
      "epoch": 1.3297872340425532,
      "grad_norm": 0.9624224796152259,
      "learning_rate": 2.5246177794795834e-06,
      "loss": 0.3086543560028076,
      "num_input_tokens_seen": 629445600,
      "step": 19750,
      "train_runtime": 51787.9875,
      "train_tokens_per_second": 12154.278
    },
    {
      "epoch": 1.3301238890385134,
      "grad_norm": 0.7126395621128093,
      "learning_rate": 2.5223208164562163e-06,
      "loss": 0.3087146520614624,
      "num_input_tokens_seen": 629600312,
      "step": 19755,
      "train_runtime": 51800.6331,
      "train_tokens_per_second": 12154.298
    },
    {
      "epoch": 1.3304605440344734,
      "grad_norm": 0.7856457956260945,
      "learning_rate": 2.5200245463079315e-06,
      "loss": 0.3151549816131592,
      "num_input_tokens_seen": 629759920,
      "step": 19760,
      "train_runtime": 51813.8985,
      "train_tokens_per_second": 12154.266
    },
    {
      "epoch": 1.3307971990304335,
      "grad_norm": 0.8079116093773693,
      "learning_rate": 2.5177289696768725e-06,
      "loss": 0.3017305850982666,
      "num_input_tokens_seen": 629917960,
      "step": 19765,
      "train_runtime": 51826.9977,
      "train_tokens_per_second": 12154.244
    },
    {
      "epoch": 1.3311338540263937,
      "grad_norm": 0.837008394646822,
      "learning_rate": 2.515434087204996e-06,
      "loss": 0.29909167289733884,
      "num_input_tokens_seen": 630080144,
      "step": 19770,
      "train_runtime": 51839.5988,
      "train_tokens_per_second": 12154.418
    },
    {
      "epoch": 1.331470509022354,
      "grad_norm": 0.7233785928159718,
      "learning_rate": 2.51313989953405e-06,
      "loss": 0.30140299797058107,
      "num_input_tokens_seen": 630241184,
      "step": 19775,
      "train_runtime": 51852.5332,
      "train_tokens_per_second": 12154.492
    },
    {
      "epoch": 1.331807164018314,
      "grad_norm": 0.7707903583020421,
      "learning_rate": 2.510846407305603e-06,
      "loss": 0.28589801788330077,
      "num_input_tokens_seen": 630400544,
      "step": 19780,
      "train_runtime": 51865.3788,
      "train_tokens_per_second": 12154.554
    },
    {
      "epoch": 1.332143819014274,
      "grad_norm": 0.7641749363354232,
      "learning_rate": 2.508553611161021e-06,
      "loss": 0.30675008296966555,
      "num_input_tokens_seen": 630563344,
      "step": 19785,
      "train_runtime": 51878.3597,
      "train_tokens_per_second": 12154.651
    },
    {
      "epoch": 1.3324804740102343,
      "grad_norm": 0.7893444557412274,
      "learning_rate": 2.5062615117414773e-06,
      "loss": 0.3037196159362793,
      "num_input_tokens_seen": 630727136,
      "step": 19790,
      "train_runtime": 51891.4796,
      "train_tokens_per_second": 12154.734
    },
    {
      "epoch": 1.3328171290061945,
      "grad_norm": 0.7773922753063138,
      "learning_rate": 2.50397010968795e-06,
      "loss": 0.27981841564178467,
      "num_input_tokens_seen": 630882312,
      "step": 19795,
      "train_runtime": 51904.3495,
      "train_tokens_per_second": 12154.71
    },
    {
      "epoch": 1.3331537840021546,
      "grad_norm": 0.869981684318592,
      "learning_rate": 2.501679405641222e-06,
      "loss": 0.3257544279098511,
      "num_input_tokens_seen": 631041224,
      "step": 19800,
      "train_runtime": 51916.8851,
      "train_tokens_per_second": 12154.836
    },
    {
      "epoch": 1.3334904389981148,
      "grad_norm": 0.8316798513378483,
      "learning_rate": 2.499389400241881e-06,
      "loss": 0.29658730030059816,
      "num_input_tokens_seen": 631193760,
      "step": 19805,
      "train_runtime": 51930.268,
      "train_tokens_per_second": 12154.641
    },
    {
      "epoch": 1.3338270939940748,
      "grad_norm": 0.805546278814821,
      "learning_rate": 2.497100094130321e-06,
      "loss": 0.2668275833129883,
      "num_input_tokens_seen": 631353656,
      "step": 19810,
      "train_runtime": 51943.4149,
      "train_tokens_per_second": 12154.643
    },
    {
      "epoch": 1.334163748990035,
      "grad_norm": 0.7900818195890321,
      "learning_rate": 2.494811487946737e-06,
      "loss": 0.2999356746673584,
      "num_input_tokens_seen": 631514072,
      "step": 19815,
      "train_runtime": 51956.4481,
      "train_tokens_per_second": 12154.681
    },
    {
      "epoch": 1.3345004039859951,
      "grad_norm": 0.7916016675742044,
      "learning_rate": 2.492523582331135e-06,
      "loss": 0.3024959325790405,
      "num_input_tokens_seen": 631671160,
      "step": 19820,
      "train_runtime": 51968.8966,
      "train_tokens_per_second": 12154.793
    },
    {
      "epoch": 1.3348370589819554,
      "grad_norm": 0.8189347327381548,
      "learning_rate": 2.4902363779233134e-06,
      "loss": 0.27855596542358396,
      "num_input_tokens_seen": 631830776,
      "step": 19825,
      "train_runtime": 51982.0558,
      "train_tokens_per_second": 12154.786
    },
    {
      "epoch": 1.3351737139779154,
      "grad_norm": 0.7747595173437904,
      "learning_rate": 2.4879498753628885e-06,
      "loss": 0.29611940383911134,
      "num_input_tokens_seen": 631985976,
      "step": 19830,
      "train_runtime": 51995.0139,
      "train_tokens_per_second": 12154.742
    },
    {
      "epoch": 1.3355103689738757,
      "grad_norm": 0.6724048275768274,
      "learning_rate": 2.4856640752892702e-06,
      "loss": 0.2555154085159302,
      "num_input_tokens_seen": 632147616,
      "step": 19835,
      "train_runtime": 52008.1678,
      "train_tokens_per_second": 12154.776
    },
    {
      "epoch": 1.3358470239698357,
      "grad_norm": 0.8092913043168205,
      "learning_rate": 2.4833789783416757e-06,
      "loss": 0.27653489112854,
      "num_input_tokens_seen": 632307536,
      "step": 19840,
      "train_runtime": 52021.689,
      "train_tokens_per_second": 12154.691
    },
    {
      "epoch": 1.336183678965796,
      "grad_norm": 0.8630187769543424,
      "learning_rate": 2.4810945851591245e-06,
      "loss": 0.3180494785308838,
      "num_input_tokens_seen": 632466168,
      "step": 19845,
      "train_runtime": 52036.1571,
      "train_tokens_per_second": 12154.36
    },
    {
      "epoch": 1.336520333961756,
      "grad_norm": 0.829062470257664,
      "learning_rate": 2.4788108963804406e-06,
      "loss": 0.30126290321350097,
      "num_input_tokens_seen": 632627768,
      "step": 19850,
      "train_runtime": 52049.9319,
      "train_tokens_per_second": 12154.248
    },
    {
      "epoch": 1.3368569889577162,
      "grad_norm": 0.7533111845683165,
      "learning_rate": 2.4765279126442502e-06,
      "loss": 0.29823622703552244,
      "num_input_tokens_seen": 632790136,
      "step": 19855,
      "train_runtime": 52063.2331,
      "train_tokens_per_second": 12154.261
    },
    {
      "epoch": 1.3371936439536762,
      "grad_norm": 0.7697610454269382,
      "learning_rate": 2.4742456345889828e-06,
      "loss": 0.30764946937561033,
      "num_input_tokens_seen": 632946720,
      "step": 19860,
      "train_runtime": 52076.6061,
      "train_tokens_per_second": 12154.147
    },
    {
      "epoch": 1.3375302989496365,
      "grad_norm": 0.7979137173361226,
      "learning_rate": 2.471964062852867e-06,
      "loss": 0.3004822015762329,
      "num_input_tokens_seen": 633105872,
      "step": 19865,
      "train_runtime": 52089.1042,
      "train_tokens_per_second": 12154.286
    },
    {
      "epoch": 1.3378669539455965,
      "grad_norm": 0.7211803037205384,
      "learning_rate": 2.469683198073943e-06,
      "loss": 0.2771628379821777,
      "num_input_tokens_seen": 633267904,
      "step": 19870,
      "train_runtime": 52102.6158,
      "train_tokens_per_second": 12154.244
    },
    {
      "epoch": 1.3382036089415568,
      "grad_norm": 0.9844008517763483,
      "learning_rate": 2.4674030408900414e-06,
      "loss": 0.29499120712280275,
      "num_input_tokens_seen": 633420176,
      "step": 19875,
      "train_runtime": 52117.7849,
      "train_tokens_per_second": 12153.628
    },
    {
      "epoch": 1.3385402639375168,
      "grad_norm": 0.8688585282802,
      "learning_rate": 2.465123591938805e-06,
      "loss": 0.30530877113342286,
      "num_input_tokens_seen": 633580640,
      "step": 19880,
      "train_runtime": 52132.0287,
      "train_tokens_per_second": 12153.385
    },
    {
      "epoch": 1.338876918933477,
      "grad_norm": 0.6927180242123807,
      "learning_rate": 2.462844851857673e-06,
      "loss": 0.3009660720825195,
      "num_input_tokens_seen": 633740800,
      "step": 19885,
      "train_runtime": 52145.2335,
      "train_tokens_per_second": 12153.379
    },
    {
      "epoch": 1.339213573929437,
      "grad_norm": 0.9173090432629345,
      "learning_rate": 2.4605668212838878e-06,
      "loss": 0.31364850997924804,
      "num_input_tokens_seen": 633894712,
      "step": 19890,
      "train_runtime": 52158.0247,
      "train_tokens_per_second": 12153.35
    },
    {
      "epoch": 1.3395502289253973,
      "grad_norm": 0.8671865075518089,
      "learning_rate": 2.458289500854493e-06,
      "loss": 0.34432239532470704,
      "num_input_tokens_seen": 634054968,
      "step": 19895,
      "train_runtime": 52170.6499,
      "train_tokens_per_second": 12153.48
    },
    {
      "epoch": 1.3398868839213574,
      "grad_norm": 0.7644951678305448,
      "learning_rate": 2.4560128912063353e-06,
      "loss": 0.3196416854858398,
      "num_input_tokens_seen": 634211088,
      "step": 19900,
      "train_runtime": 52183.176,
      "train_tokens_per_second": 12153.555
    },
    {
      "epoch": 1.3402235389173176,
      "grad_norm": 0.9512354150127618,
      "learning_rate": 2.4537369929760596e-06,
      "loss": 0.33624939918518065,
      "num_input_tokens_seen": 634373920,
      "step": 19905,
      "train_runtime": 52195.9398,
      "train_tokens_per_second": 12153.702
    },
    {
      "epoch": 1.3405601939132776,
      "grad_norm": 0.7645194368857552,
      "learning_rate": 2.4514618068001144e-06,
      "loss": 0.29656047821044923,
      "num_input_tokens_seen": 634532608,
      "step": 19910,
      "train_runtime": 52209.5133,
      "train_tokens_per_second": 12153.582
    },
    {
      "epoch": 1.3408968489092379,
      "grad_norm": 0.8491749965957714,
      "learning_rate": 2.4491873333147463e-06,
      "loss": 0.29364495277404784,
      "num_input_tokens_seen": 634692920,
      "step": 19915,
      "train_runtime": 52223.0599,
      "train_tokens_per_second": 12153.499
    },
    {
      "epoch": 1.341233503905198,
      "grad_norm": 0.7826073957281497,
      "learning_rate": 2.4469135731560106e-06,
      "loss": 0.26431825160980227,
      "num_input_tokens_seen": 634853984,
      "step": 19920,
      "train_runtime": 52236.6501,
      "train_tokens_per_second": 12153.421
    },
    {
      "epoch": 1.3415701589011582,
      "grad_norm": 0.8363216956410162,
      "learning_rate": 2.4446405269597494e-06,
      "loss": 0.3138526439666748,
      "num_input_tokens_seen": 635014616,
      "step": 19925,
      "train_runtime": 52249.1179,
      "train_tokens_per_second": 12153.595
    },
    {
      "epoch": 1.3419068138971182,
      "grad_norm": 0.7565652427878659,
      "learning_rate": 2.442368195361618e-06,
      "loss": 0.26955063343048097,
      "num_input_tokens_seen": 635175840,
      "step": 19930,
      "train_runtime": 52262.2119,
      "train_tokens_per_second": 12153.635
    },
    {
      "epoch": 1.3422434688930784,
      "grad_norm": 0.7927779914012338,
      "learning_rate": 2.4400965789970647e-06,
      "loss": 0.26891498565673827,
      "num_input_tokens_seen": 635332568,
      "step": 19935,
      "train_runtime": 52275.5723,
      "train_tokens_per_second": 12153.527
    },
    {
      "epoch": 1.3425801238890385,
      "grad_norm": 0.810981689028006,
      "learning_rate": 2.437825678501339e-06,
      "loss": 0.30417168140411377,
      "num_input_tokens_seen": 635494384,
      "step": 19940,
      "train_runtime": 52288.68,
      "train_tokens_per_second": 12153.575
    },
    {
      "epoch": 1.3429167788849987,
      "grad_norm": 0.7299683185413863,
      "learning_rate": 2.4355554945094956e-06,
      "loss": 0.3055759906768799,
      "num_input_tokens_seen": 635653272,
      "step": 19945,
      "train_runtime": 52301.4903,
      "train_tokens_per_second": 12153.636
    },
    {
      "epoch": 1.3432534338809587,
      "grad_norm": 0.8175416819947094,
      "learning_rate": 2.4332860276563763e-06,
      "loss": 0.29556589126586913,
      "num_input_tokens_seen": 635814144,
      "step": 19950,
      "train_runtime": 52314.4698,
      "train_tokens_per_second": 12153.696
    },
    {
      "epoch": 1.343590088876919,
      "grad_norm": 0.7521649670790872,
      "learning_rate": 2.4310172785766396e-06,
      "loss": 0.30762975215911864,
      "num_input_tokens_seen": 635973408,
      "step": 19955,
      "train_runtime": 52327.5224,
      "train_tokens_per_second": 12153.708
    },
    {
      "epoch": 1.343926743872879,
      "grad_norm": 0.9515691037989283,
      "learning_rate": 2.428749247904724e-06,
      "loss": 0.3347216844558716,
      "num_input_tokens_seen": 636134256,
      "step": 19960,
      "train_runtime": 52341.0765,
      "train_tokens_per_second": 12153.633
    },
    {
      "epoch": 1.3442633988688393,
      "grad_norm": 0.9032666472354995,
      "learning_rate": 2.4264819362748856e-06,
      "loss": 0.3079484462738037,
      "num_input_tokens_seen": 636294264,
      "step": 19965,
      "train_runtime": 52354.3419,
      "train_tokens_per_second": 12153.61
    },
    {
      "epoch": 1.3446000538647993,
      "grad_norm": 0.8694679246828867,
      "learning_rate": 2.4242153443211663e-06,
      "loss": 0.3199774265289307,
      "num_input_tokens_seen": 636453696,
      "step": 19970,
      "train_runtime": 52366.832,
      "train_tokens_per_second": 12153.756
    },
    {
      "epoch": 1.3449367088607596,
      "grad_norm": 0.6981243209367544,
      "learning_rate": 2.4219494726774124e-06,
      "loss": 0.285122275352478,
      "num_input_tokens_seen": 636615048,
      "step": 19975,
      "train_runtime": 52379.8466,
      "train_tokens_per_second": 12153.817
    },
    {
      "epoch": 1.3452733638567196,
      "grad_norm": 0.8016713620691758,
      "learning_rate": 2.419684321977268e-06,
      "loss": 0.28094542026519775,
      "num_input_tokens_seen": 636773464,
      "step": 19980,
      "train_runtime": 52393.0724,
      "train_tokens_per_second": 12153.772
    },
    {
      "epoch": 1.3456100188526798,
      "grad_norm": 0.7880713494728689,
      "learning_rate": 2.417419892854175e-06,
      "loss": 0.29432547092437744,
      "num_input_tokens_seen": 636936944,
      "step": 19985,
      "train_runtime": 52406.0792,
      "train_tokens_per_second": 12153.875
    },
    {
      "epoch": 1.3459466738486399,
      "grad_norm": 0.8321660809639218,
      "learning_rate": 2.415156185941371e-06,
      "loss": 0.28353219032287597,
      "num_input_tokens_seen": 637093328,
      "step": 19990,
      "train_runtime": 52419.2539,
      "train_tokens_per_second": 12153.804
    },
    {
      "epoch": 1.3462833288446001,
      "grad_norm": 0.9027367087215813,
      "learning_rate": 2.412893201871902e-06,
      "loss": 0.2969489336013794,
      "num_input_tokens_seen": 637249448,
      "step": 19995,
      "train_runtime": 52432.2755,
      "train_tokens_per_second": 12153.763
    },
    {
      "epoch": 1.3466199838405601,
      "grad_norm": 0.8621812867353942,
      "learning_rate": 2.410630941278595e-06,
      "loss": 0.31229348182678224,
      "num_input_tokens_seen": 637409000,
      "step": 20000,
      "train_runtime": 52444.7741,
      "train_tokens_per_second": 12153.909
    },
    {
      "epoch": 1.3469566388365204,
      "grad_norm": 0.8793321350766958,
      "learning_rate": 2.408369404794092e-06,
      "loss": 0.2945550441741943,
      "num_input_tokens_seen": 637566568,
      "step": 20005,
      "train_runtime": 52458.1115,
      "train_tokens_per_second": 12153.822
    },
    {
      "epoch": 1.3472932938324804,
      "grad_norm": 0.8136369098059965,
      "learning_rate": 2.4061085930508176e-06,
      "loss": 0.3121423959732056,
      "num_input_tokens_seen": 637726152,
      "step": 20010,
      "train_runtime": 52471.9089,
      "train_tokens_per_second": 12153.668
    },
    {
      "epoch": 1.3476299488284407,
      "grad_norm": 0.8031971613534401,
      "learning_rate": 2.4038485066810058e-06,
      "loss": 0.30166139602661135,
      "num_input_tokens_seen": 637882736,
      "step": 20015,
      "train_runtime": 52484.5444,
      "train_tokens_per_second": 12153.725
    },
    {
      "epoch": 1.3479666038244007,
      "grad_norm": 0.8433180944775368,
      "learning_rate": 2.40158914631668e-06,
      "loss": 0.28267199993133546,
      "num_input_tokens_seen": 638040056,
      "step": 20020,
      "train_runtime": 52497.8495,
      "train_tokens_per_second": 12153.642
    },
    {
      "epoch": 1.348303258820361,
      "grad_norm": 0.7735758816815589,
      "learning_rate": 2.399330512589664e-06,
      "loss": 0.2838400602340698,
      "num_input_tokens_seen": 638199384,
      "step": 20025,
      "train_runtime": 52510.5884,
      "train_tokens_per_second": 12153.728
    },
    {
      "epoch": 1.348639913816321,
      "grad_norm": 0.7395757615136218,
      "learning_rate": 2.3970726061315775e-06,
      "loss": 0.2910667896270752,
      "num_input_tokens_seen": 638357712,
      "step": 20030,
      "train_runtime": 52523.2855,
      "train_tokens_per_second": 12153.804
    },
    {
      "epoch": 1.3489765688122812,
      "grad_norm": 0.7485677250584262,
      "learning_rate": 2.394815427573836e-06,
      "loss": 0.2860706329345703,
      "num_input_tokens_seen": 638521552,
      "step": 20035,
      "train_runtime": 52535.7247,
      "train_tokens_per_second": 12154.045
    },
    {
      "epoch": 1.3493132238082413,
      "grad_norm": 0.8977453709520217,
      "learning_rate": 2.3925589775476516e-06,
      "loss": 0.29819340705871583,
      "num_input_tokens_seen": 638676840,
      "step": 20040,
      "train_runtime": 52548.2715,
      "train_tokens_per_second": 12154.098
    },
    {
      "epoch": 1.3496498788042015,
      "grad_norm": 0.7675039206175186,
      "learning_rate": 2.3903032566840377e-06,
      "loss": 0.2958210945129395,
      "num_input_tokens_seen": 638838424,
      "step": 20045,
      "train_runtime": 52561.2984,
      "train_tokens_per_second": 12154.16
    },
    {
      "epoch": 1.3499865338001615,
      "grad_norm": 0.8027559792710525,
      "learning_rate": 2.3880482656137926e-06,
      "loss": 0.3185994863510132,
      "num_input_tokens_seen": 639001688,
      "step": 20050,
      "train_runtime": 52574.1419,
      "train_tokens_per_second": 12154.296
    },
    {
      "epoch": 1.3503231887961218,
      "grad_norm": 0.8008503641107605,
      "learning_rate": 2.3857940049675254e-06,
      "loss": 0.2951274156570435,
      "num_input_tokens_seen": 639160168,
      "step": 20055,
      "train_runtime": 52586.6652,
      "train_tokens_per_second": 12154.415
    },
    {
      "epoch": 1.3506598437920818,
      "grad_norm": 0.749398809109107,
      "learning_rate": 2.383540475375625e-06,
      "loss": 0.29441251754760744,
      "num_input_tokens_seen": 639321264,
      "step": 20060,
      "train_runtime": 52599.5699,
      "train_tokens_per_second": 12154.496
    },
    {
      "epoch": 1.350996498788042,
      "grad_norm": 0.8856747280013086,
      "learning_rate": 2.3812876774682886e-06,
      "loss": 0.3095348834991455,
      "num_input_tokens_seen": 639484632,
      "step": 20065,
      "train_runtime": 52612.5574,
      "train_tokens_per_second": 12154.601
    },
    {
      "epoch": 1.351333153784002,
      "grad_norm": 0.8713164599063117,
      "learning_rate": 2.379035611875503e-06,
      "loss": 0.3180835723876953,
      "num_input_tokens_seen": 639648232,
      "step": 20070,
      "train_runtime": 52625.5822,
      "train_tokens_per_second": 12154.701
    },
    {
      "epoch": 1.3516698087799623,
      "grad_norm": 0.7787086722092307,
      "learning_rate": 2.376784279227051e-06,
      "loss": 0.2957012176513672,
      "num_input_tokens_seen": 639810456,
      "step": 20075,
      "train_runtime": 52638.1761,
      "train_tokens_per_second": 12154.875
    },
    {
      "epoch": 1.3520064637759224,
      "grad_norm": 0.7454683425133161,
      "learning_rate": 2.3745336801525105e-06,
      "loss": 0.2989673614501953,
      "num_input_tokens_seen": 639973664,
      "step": 20080,
      "train_runtime": 52651.0791,
      "train_tokens_per_second": 12154.996
    },
    {
      "epoch": 1.3523431187718826,
      "grad_norm": 0.825988768846959,
      "learning_rate": 2.3722838152812543e-06,
      "loss": 0.3139394044876099,
      "num_input_tokens_seen": 640136112,
      "step": 20085,
      "train_runtime": 52663.7533,
      "train_tokens_per_second": 12155.156
    },
    {
      "epoch": 1.3526797737678427,
      "grad_norm": 0.6902843722757951,
      "learning_rate": 2.3700346852424485e-06,
      "loss": 0.29785478115081787,
      "num_input_tokens_seen": 640293120,
      "step": 20090,
      "train_runtime": 52676.675,
      "train_tokens_per_second": 12155.154
    },
    {
      "epoch": 1.353016428763803,
      "grad_norm": 0.769242715723652,
      "learning_rate": 2.3677862906650594e-06,
      "loss": 0.31523053646087645,
      "num_input_tokens_seen": 640451600,
      "step": 20095,
      "train_runtime": 52690.4172,
      "train_tokens_per_second": 12154.992
    },
    {
      "epoch": 1.353353083759763,
      "grad_norm": 0.8916299069224247,
      "learning_rate": 2.365538632177837e-06,
      "loss": 0.3254724025726318,
      "num_input_tokens_seen": 640609000,
      "step": 20100,
      "train_runtime": 52704.209,
      "train_tokens_per_second": 12154.798
    },
    {
      "epoch": 1.3536897387557232,
      "grad_norm": 0.8597770559813573,
      "learning_rate": 2.36329171040934e-06,
      "loss": 0.2984098196029663,
      "num_input_tokens_seen": 640770752,
      "step": 20105,
      "train_runtime": 52717.332,
      "train_tokens_per_second": 12154.84
    },
    {
      "epoch": 1.3540263937516832,
      "grad_norm": 0.832823347777731,
      "learning_rate": 2.3610455259879045e-06,
      "loss": 0.2707779884338379,
      "num_input_tokens_seen": 640928896,
      "step": 20110,
      "train_runtime": 52730.0837,
      "train_tokens_per_second": 12154.9
    },
    {
      "epoch": 1.3543630487476435,
      "grad_norm": 0.8701175587509546,
      "learning_rate": 2.358800079541674e-06,
      "loss": 0.3292681217193604,
      "num_input_tokens_seen": 641085304,
      "step": 20115,
      "train_runtime": 52742.8493,
      "train_tokens_per_second": 12154.924
    },
    {
      "epoch": 1.3546997037436035,
      "grad_norm": 0.8036502264634139,
      "learning_rate": 2.3565553716985794e-06,
      "loss": 0.3127922296524048,
      "num_input_tokens_seen": 641247320,
      "step": 20120,
      "train_runtime": 52756.3204,
      "train_tokens_per_second": 12154.891
    },
    {
      "epoch": 1.3550363587395637,
      "grad_norm": 0.8138511888964206,
      "learning_rate": 2.3543114030863457e-06,
      "loss": 0.2915943622589111,
      "num_input_tokens_seen": 641402344,
      "step": 20125,
      "train_runtime": 52769.5696,
      "train_tokens_per_second": 12154.777
    },
    {
      "epoch": 1.3553730137355238,
      "grad_norm": 0.7866679518898284,
      "learning_rate": 2.352068174332491e-06,
      "loss": 0.29277997016906737,
      "num_input_tokens_seen": 641558016,
      "step": 20130,
      "train_runtime": 52782.4153,
      "train_tokens_per_second": 12154.768
    },
    {
      "epoch": 1.355709668731484,
      "grad_norm": 0.7635350230589175,
      "learning_rate": 2.349825686064329e-06,
      "loss": 0.2615662097930908,
      "num_input_tokens_seen": 641716112,
      "step": 20135,
      "train_runtime": 52795.261,
      "train_tokens_per_second": 12154.805
    },
    {
      "epoch": 1.356046323727444,
      "grad_norm": 0.7900592585406084,
      "learning_rate": 2.3475839389089605e-06,
      "loss": 0.30844969749450685,
      "num_input_tokens_seen": 641877440,
      "step": 20140,
      "train_runtime": 52807.8695,
      "train_tokens_per_second": 12154.958
    },
    {
      "epoch": 1.3563829787234043,
      "grad_norm": 0.8383462166740326,
      "learning_rate": 2.345342933493289e-06,
      "loss": 0.3140614748001099,
      "num_input_tokens_seen": 642037840,
      "step": 20145,
      "train_runtime": 52821.4997,
      "train_tokens_per_second": 12154.858
    },
    {
      "epoch": 1.3567196337193643,
      "grad_norm": 0.7917548783183068,
      "learning_rate": 2.343102670443998e-06,
      "loss": 0.2943148612976074,
      "num_input_tokens_seen": 642197584,
      "step": 20150,
      "train_runtime": 52834.5267,
      "train_tokens_per_second": 12154.885
    },
    {
      "epoch": 1.3570562887153246,
      "grad_norm": 0.8532196620748863,
      "learning_rate": 2.3408631503875763e-06,
      "loss": 0.3162414312362671,
      "num_input_tokens_seen": 642360600,
      "step": 20155,
      "train_runtime": 52847.3705,
      "train_tokens_per_second": 12155.015
    },
    {
      "epoch": 1.3573929437112846,
      "grad_norm": 0.9476095408820352,
      "learning_rate": 2.338624373950291e-06,
      "loss": 0.2608856201171875,
      "num_input_tokens_seen": 642517280,
      "step": 20160,
      "train_runtime": 52860.3661,
      "train_tokens_per_second": 12154.991
    },
    {
      "epoch": 1.3577295987072449,
      "grad_norm": 0.8690424888057344,
      "learning_rate": 2.336386341758216e-06,
      "loss": 0.30047221183776857,
      "num_input_tokens_seen": 642680464,
      "step": 20165,
      "train_runtime": 52873.2499,
      "train_tokens_per_second": 12155.116
    },
    {
      "epoch": 1.3580662537032049,
      "grad_norm": 0.7918968008107469,
      "learning_rate": 2.3341490544372063e-06,
      "loss": 0.30344624519348146,
      "num_input_tokens_seen": 642841328,
      "step": 20170,
      "train_runtime": 52886.3849,
      "train_tokens_per_second": 12155.138
    },
    {
      "epoch": 1.3584029086991651,
      "grad_norm": 0.8283289286328732,
      "learning_rate": 2.3319125126129128e-06,
      "loss": 0.30180726051330564,
      "num_input_tokens_seen": 643001144,
      "step": 20175,
      "train_runtime": 52899.5875,
      "train_tokens_per_second": 12155.126
    },
    {
      "epoch": 1.3587395636951252,
      "grad_norm": 0.8090597312132042,
      "learning_rate": 2.329676716910777e-06,
      "loss": 0.3208785057067871,
      "num_input_tokens_seen": 643160168,
      "step": 20180,
      "train_runtime": 52912.8792,
      "train_tokens_per_second": 12155.078
    },
    {
      "epoch": 1.3590762186910854,
      "grad_norm": 0.8823636834617153,
      "learning_rate": 2.3274416679560325e-06,
      "loss": 0.3125479698181152,
      "num_input_tokens_seen": 643317136,
      "step": 20185,
      "train_runtime": 52925.3834,
      "train_tokens_per_second": 12155.172
    },
    {
      "epoch": 1.3594128736870454,
      "grad_norm": 0.8064380366613495,
      "learning_rate": 2.3252073663737016e-06,
      "loss": 0.3070643901824951,
      "num_input_tokens_seen": 643476488,
      "step": 20190,
      "train_runtime": 52938.1674,
      "train_tokens_per_second": 12155.247
    },
    {
      "epoch": 1.3597495286830057,
      "grad_norm": 0.9077138242757049,
      "learning_rate": 2.3229738127886054e-06,
      "loss": 0.29928030967712405,
      "num_input_tokens_seen": 643632272,
      "step": 20195,
      "train_runtime": 52951.3469,
      "train_tokens_per_second": 12155.163
    },
    {
      "epoch": 1.3600861836789657,
      "grad_norm": 0.764593905570044,
      "learning_rate": 2.3207410078253416e-06,
      "loss": 0.3044393539428711,
      "num_input_tokens_seen": 643794064,
      "step": 20200,
      "train_runtime": 52964.55,
      "train_tokens_per_second": 12155.188
    },
    {
      "epoch": 1.360422838674926,
      "grad_norm": 0.7860160753653275,
      "learning_rate": 2.3185089521083154e-06,
      "loss": 0.2984424591064453,
      "num_input_tokens_seen": 643952064,
      "step": 20205,
      "train_runtime": 52977.7879,
      "train_tokens_per_second": 12155.133
    },
    {
      "epoch": 1.360759493670886,
      "grad_norm": 0.8242439762496754,
      "learning_rate": 2.3162776462617067e-06,
      "loss": 0.29924328327178956,
      "num_input_tokens_seen": 644110720,
      "step": 20210,
      "train_runtime": 52991.2186,
      "train_tokens_per_second": 12155.046
    },
    {
      "epoch": 1.3610961486668463,
      "grad_norm": 0.9064423233978224,
      "learning_rate": 2.314047090909498e-06,
      "loss": 0.2972831726074219,
      "num_input_tokens_seen": 644268552,
      "step": 20215,
      "train_runtime": 53004.8521,
      "train_tokens_per_second": 12154.898
    },
    {
      "epoch": 1.3614328036628063,
      "grad_norm": 0.7953908340523524,
      "learning_rate": 2.3118172866754557e-06,
      "loss": 0.2784623146057129,
      "num_input_tokens_seen": 644428640,
      "step": 20220,
      "train_runtime": 53018.4882,
      "train_tokens_per_second": 12154.791
    },
    {
      "epoch": 1.3617694586587665,
      "grad_norm": 0.8545430247946103,
      "learning_rate": 2.309588234183137e-06,
      "loss": 0.3083958625793457,
      "num_input_tokens_seen": 644590768,
      "step": 20225,
      "train_runtime": 53031.0894,
      "train_tokens_per_second": 12154.96
    },
    {
      "epoch": 1.3621061136547266,
      "grad_norm": 0.8543218659958202,
      "learning_rate": 2.3073599340558896e-06,
      "loss": 0.3121353626251221,
      "num_input_tokens_seen": 644743488,
      "step": 20230,
      "train_runtime": 53043.5142,
      "train_tokens_per_second": 12154.992
    },
    {
      "epoch": 1.3624427686506868,
      "grad_norm": 0.840823539646017,
      "learning_rate": 2.3051323869168506e-06,
      "loss": 0.31676955223083497,
      "num_input_tokens_seen": 644906448,
      "step": 20235,
      "train_runtime": 53056.9507,
      "train_tokens_per_second": 12154.985
    },
    {
      "epoch": 1.3627794236466468,
      "grad_norm": 0.7695727050359703,
      "learning_rate": 2.3029055933889443e-06,
      "loss": 0.29796838760375977,
      "num_input_tokens_seen": 645067216,
      "step": 20240,
      "train_runtime": 53070.5915,
      "train_tokens_per_second": 12154.89
    },
    {
      "epoch": 1.363116078642607,
      "grad_norm": 0.7413222298191311,
      "learning_rate": 2.300679554094892e-06,
      "loss": 0.28080062866210936,
      "num_input_tokens_seen": 645221152,
      "step": 20245,
      "train_runtime": 53084.0085,
      "train_tokens_per_second": 12154.718
    },
    {
      "epoch": 1.3634527336385671,
      "grad_norm": 0.7995496189254991,
      "learning_rate": 2.298454269657191e-06,
      "loss": 0.29342241287231446,
      "num_input_tokens_seen": 645384952,
      "step": 20250,
      "train_runtime": 53097.121,
      "train_tokens_per_second": 12154.801
    },
    {
      "epoch": 1.3637893886345274,
      "grad_norm": 0.8235469360233216,
      "learning_rate": 2.2962297406981432e-06,
      "loss": 0.2905342817306519,
      "num_input_tokens_seen": 645547064,
      "step": 20255,
      "train_runtime": 53110.3531,
      "train_tokens_per_second": 12154.825
    },
    {
      "epoch": 1.3641260436304874,
      "grad_norm": 0.8200034628825106,
      "learning_rate": 2.2940059678398223e-06,
      "loss": 0.29088120460510253,
      "num_input_tokens_seen": 645704912,
      "step": 20260,
      "train_runtime": 53123.9112,
      "train_tokens_per_second": 12154.695
    },
    {
      "epoch": 1.3644626986264476,
      "grad_norm": 0.8196035605889611,
      "learning_rate": 2.291782951704106e-06,
      "loss": 0.3106068134307861,
      "num_input_tokens_seen": 645859536,
      "step": 20265,
      "train_runtime": 53136.4196,
      "train_tokens_per_second": 12154.743
    },
    {
      "epoch": 1.3647993536224077,
      "grad_norm": 0.8369724176849058,
      "learning_rate": 2.289560692912651e-06,
      "loss": 0.31033525466918943,
      "num_input_tokens_seen": 646022440,
      "step": 20270,
      "train_runtime": 53149.1968,
      "train_tokens_per_second": 12154.886
    },
    {
      "epoch": 1.365136008618368,
      "grad_norm": 0.77014717726597,
      "learning_rate": 2.287339192086906e-06,
      "loss": 0.3115870714187622,
      "num_input_tokens_seen": 646185656,
      "step": 20275,
      "train_runtime": 53162.1028,
      "train_tokens_per_second": 12155.006
    },
    {
      "epoch": 1.365472663614328,
      "grad_norm": 0.8339959826906802,
      "learning_rate": 2.2851184498481056e-06,
      "loss": 0.2991509437561035,
      "num_input_tokens_seen": 646342728,
      "step": 20280,
      "train_runtime": 53174.7987,
      "train_tokens_per_second": 12155.057
    },
    {
      "epoch": 1.3658093186102882,
      "grad_norm": 0.9393207244577194,
      "learning_rate": 2.2828984668172732e-06,
      "loss": 0.3031912803649902,
      "num_input_tokens_seen": 646500168,
      "step": 20285,
      "train_runtime": 53187.6798,
      "train_tokens_per_second": 12155.074
    },
    {
      "epoch": 1.3661459736062485,
      "grad_norm": 0.860650306064623,
      "learning_rate": 2.2806792436152185e-06,
      "loss": 0.3075080871582031,
      "num_input_tokens_seen": 646659128,
      "step": 20290,
      "train_runtime": 53201.0408,
      "train_tokens_per_second": 12155.009
    },
    {
      "epoch": 1.3664826286022085,
      "grad_norm": 0.7783009056387936,
      "learning_rate": 2.2784607808625463e-06,
      "loss": 0.30388917922973635,
      "num_input_tokens_seen": 646817080,
      "step": 20295,
      "train_runtime": 53214.1281,
      "train_tokens_per_second": 12154.988
    },
    {
      "epoch": 1.3668192835981685,
      "grad_norm": 0.7917413854384189,
      "learning_rate": 2.276243079179634e-06,
      "loss": 0.3020165920257568,
      "num_input_tokens_seen": 646971976,
      "step": 20300,
      "train_runtime": 53226.7817,
      "train_tokens_per_second": 12155.008
    },
    {
      "epoch": 1.3671559385941288,
      "grad_norm": 0.8149458607077814,
      "learning_rate": 2.2740261391866634e-06,
      "loss": 0.32889571189880373,
      "num_input_tokens_seen": 647132648,
      "step": 20305,
      "train_runtime": 53240.1881,
      "train_tokens_per_second": 12154.965
    },
    {
      "epoch": 1.367492593590089,
      "grad_norm": 0.730122124638197,
      "learning_rate": 2.2718099615035865e-06,
      "loss": 0.25882878303527834,
      "num_input_tokens_seen": 647290056,
      "step": 20310,
      "train_runtime": 53252.9228,
      "train_tokens_per_second": 12155.015
    },
    {
      "epoch": 1.367829248586049,
      "grad_norm": 0.8543694785134872,
      "learning_rate": 2.2695945467501567e-06,
      "loss": 0.30951781272888185,
      "num_input_tokens_seen": 647450296,
      "step": 20315,
      "train_runtime": 53266.3677,
      "train_tokens_per_second": 12154.955
    },
    {
      "epoch": 1.368165903582009,
      "grad_norm": 0.7235852274698236,
      "learning_rate": 2.267379895545905e-06,
      "loss": 0.2906651020050049,
      "num_input_tokens_seen": 647611760,
      "step": 20320,
      "train_runtime": 53279.4908,
      "train_tokens_per_second": 12154.992
    },
    {
      "epoch": 1.3685025585779693,
      "grad_norm": 0.7344515603703453,
      "learning_rate": 2.2651660085101513e-06,
      "loss": 0.3059160470962524,
      "num_input_tokens_seen": 647773504,
      "step": 20325,
      "train_runtime": 53292.8561,
      "train_tokens_per_second": 12154.978
    },
    {
      "epoch": 1.3688392135739296,
      "grad_norm": 0.7776802484841481,
      "learning_rate": 2.262952886262003e-06,
      "loss": 0.31058714389801023,
      "num_input_tokens_seen": 647936232,
      "step": 20330,
      "train_runtime": 53305.2746,
      "train_tokens_per_second": 12155.199
    },
    {
      "epoch": 1.3691758685698896,
      "grad_norm": 0.7566630176376529,
      "learning_rate": 2.260740529420352e-06,
      "loss": 0.29704785346984863,
      "num_input_tokens_seen": 648096152,
      "step": 20335,
      "train_runtime": 53318.5677,
      "train_tokens_per_second": 12155.168
    },
    {
      "epoch": 1.3695125235658496,
      "grad_norm": 0.8749092245796063,
      "learning_rate": 2.2585289386038754e-06,
      "loss": 0.29091415405273435,
      "num_input_tokens_seen": 648259992,
      "step": 20340,
      "train_runtime": 53331.0008,
      "train_tokens_per_second": 12155.406
    },
    {
      "epoch": 1.3698491785618099,
      "grad_norm": 0.8316044451868643,
      "learning_rate": 2.256318114431043e-06,
      "loss": 0.3181851863861084,
      "num_input_tokens_seen": 648423448,
      "step": 20345,
      "train_runtime": 53343.9543,
      "train_tokens_per_second": 12155.519
    },
    {
      "epoch": 1.3701858335577701,
      "grad_norm": 4.219110823247903,
      "learning_rate": 2.2541080575200973e-06,
      "loss": 0.30272440910339354,
      "num_input_tokens_seen": 648577384,
      "step": 20350,
      "train_runtime": 53356.6458,
      "train_tokens_per_second": 12155.513
    },
    {
      "epoch": 1.3705224885537302,
      "grad_norm": 0.9293639816098712,
      "learning_rate": 2.251898768489081e-06,
      "loss": 0.3147286891937256,
      "num_input_tokens_seen": 648739552,
      "step": 20355,
      "train_runtime": 53369.2478,
      "train_tokens_per_second": 12155.681
    },
    {
      "epoch": 1.3708591435496902,
      "grad_norm": 0.8065309017181531,
      "learning_rate": 2.249690247955807e-06,
      "loss": 0.30603728294372556,
      "num_input_tokens_seen": 648901096,
      "step": 20360,
      "train_runtime": 53383.0399,
      "train_tokens_per_second": 12155.567
    },
    {
      "epoch": 1.3711957985456504,
      "grad_norm": 1.1345342644250749,
      "learning_rate": 2.247482496537887e-06,
      "loss": 0.3286911487579346,
      "num_input_tokens_seen": 649061496,
      "step": 20365,
      "train_runtime": 53396.2643,
      "train_tokens_per_second": 12155.56
    },
    {
      "epoch": 1.3715324535416107,
      "grad_norm": 0.7883719025065131,
      "learning_rate": 2.24527551485271e-06,
      "loss": 0.291964054107666,
      "num_input_tokens_seen": 649221616,
      "step": 20370,
      "train_runtime": 53410.646,
      "train_tokens_per_second": 12155.285
    },
    {
      "epoch": 1.3718691085375707,
      "grad_norm": 0.7878282261106022,
      "learning_rate": 2.243069303517451e-06,
      "loss": 0.3231219291687012,
      "num_input_tokens_seen": 649378072,
      "step": 20375,
      "train_runtime": 53423.9051,
      "train_tokens_per_second": 12155.197
    },
    {
      "epoch": 1.3722057635335307,
      "grad_norm": 0.7310456743376753,
      "learning_rate": 2.2408638631490696e-06,
      "loss": 0.2690922260284424,
      "num_input_tokens_seen": 649535328,
      "step": 20380,
      "train_runtime": 53436.9424,
      "train_tokens_per_second": 12155.174
    },
    {
      "epoch": 1.372542418529491,
      "grad_norm": 0.8489539609784894,
      "learning_rate": 2.2386591943643117e-06,
      "loss": 0.2847634792327881,
      "num_input_tokens_seen": 649696992,
      "step": 20385,
      "train_runtime": 53449.4184,
      "train_tokens_per_second": 12155.361
    },
    {
      "epoch": 1.3728790735254512,
      "grad_norm": 0.9903118227967844,
      "learning_rate": 2.2364552977797023e-06,
      "loss": 0.31690526008605957,
      "num_input_tokens_seen": 649858424,
      "step": 20390,
      "train_runtime": 53462.2038,
      "train_tokens_per_second": 12155.474
    },
    {
      "epoch": 1.3732157285214113,
      "grad_norm": 0.9845624989259556,
      "learning_rate": 2.23425217401156e-06,
      "loss": 0.3207815647125244,
      "num_input_tokens_seen": 650016632,
      "step": 20395,
      "train_runtime": 53474.9841,
      "train_tokens_per_second": 12155.527
    },
    {
      "epoch": 1.3735523835173713,
      "grad_norm": 0.7864501432037757,
      "learning_rate": 2.2320498236759746e-06,
      "loss": 0.2835537433624268,
      "num_input_tokens_seen": 650174296,
      "step": 20400,
      "train_runtime": 53488.1165,
      "train_tokens_per_second": 12155.491
    },
    {
      "epoch": 1.3738890385133316,
      "grad_norm": 0.7211067744869608,
      "learning_rate": 2.2298482473888335e-06,
      "loss": 0.2840959787368774,
      "num_input_tokens_seen": 650332048,
      "step": 20405,
      "train_runtime": 53501.3699,
      "train_tokens_per_second": 12155.428
    },
    {
      "epoch": 1.3742256935092918,
      "grad_norm": 0.8430885595797492,
      "learning_rate": 2.227647445765792e-06,
      "loss": 0.30764546394348147,
      "num_input_tokens_seen": 650494928,
      "step": 20410,
      "train_runtime": 53514.0615,
      "train_tokens_per_second": 12155.589
    },
    {
      "epoch": 1.3745623485052518,
      "grad_norm": 0.8836690848301012,
      "learning_rate": 2.225447419422305e-06,
      "loss": 0.26287431716918946,
      "num_input_tokens_seen": 650652568,
      "step": 20415,
      "train_runtime": 53526.8535,
      "train_tokens_per_second": 12155.629
    },
    {
      "epoch": 1.3748990035012119,
      "grad_norm": 0.8416496862738256,
      "learning_rate": 2.223248168973599e-06,
      "loss": 0.28356237411499025,
      "num_input_tokens_seen": 650810656,
      "step": 20420,
      "train_runtime": 53539.5576,
      "train_tokens_per_second": 12155.697
    },
    {
      "epoch": 1.375235658497172,
      "grad_norm": 0.7617847217190507,
      "learning_rate": 2.221049695034688e-06,
      "loss": 0.2823204517364502,
      "num_input_tokens_seen": 650972040,
      "step": 20425,
      "train_runtime": 53553.1804,
      "train_tokens_per_second": 12155.619
    },
    {
      "epoch": 1.3755723134931324,
      "grad_norm": 0.8865559131956195,
      "learning_rate": 2.218851998220369e-06,
      "loss": 0.3196236610412598,
      "num_input_tokens_seen": 651130128,
      "step": 20430,
      "train_runtime": 53566.2387,
      "train_tokens_per_second": 12155.607
    },
    {
      "epoch": 1.3759089684890924,
      "grad_norm": 0.7729107191356331,
      "learning_rate": 2.21665507914522e-06,
      "loss": 0.28019592761993406,
      "num_input_tokens_seen": 651286648,
      "step": 20435,
      "train_runtime": 53579.215,
      "train_tokens_per_second": 12155.584
    },
    {
      "epoch": 1.3762456234850524,
      "grad_norm": 0.752525906536745,
      "learning_rate": 2.214458938423601e-06,
      "loss": 0.29229083061218264,
      "num_input_tokens_seen": 651447448,
      "step": 20440,
      "train_runtime": 53591.8469,
      "train_tokens_per_second": 12155.719
    },
    {
      "epoch": 1.3765822784810127,
      "grad_norm": 0.6961154947465655,
      "learning_rate": 2.212263576669661e-06,
      "loss": 0.2930954933166504,
      "num_input_tokens_seen": 651610472,
      "step": 20445,
      "train_runtime": 53605.2735,
      "train_tokens_per_second": 12155.716
    },
    {
      "epoch": 1.376918933476973,
      "grad_norm": 0.8985677006398,
      "learning_rate": 2.2100689944973182e-06,
      "loss": 0.30328404903411865,
      "num_input_tokens_seen": 651772256,
      "step": 20450,
      "train_runtime": 53617.7376,
      "train_tokens_per_second": 12155.907
    },
    {
      "epoch": 1.377255588472933,
      "grad_norm": 0.7958408325362172,
      "learning_rate": 2.2078751925202886e-06,
      "loss": 0.3348668336868286,
      "num_input_tokens_seen": 651934192,
      "step": 20455,
      "train_runtime": 53630.3207,
      "train_tokens_per_second": 12156.075
    },
    {
      "epoch": 1.377592243468893,
      "grad_norm": 2.077161486397874,
      "learning_rate": 2.205682171352054e-06,
      "loss": 0.3221259355545044,
      "num_input_tokens_seen": 652096248,
      "step": 20460,
      "train_runtime": 53644.219,
      "train_tokens_per_second": 12155.946
    },
    {
      "epoch": 1.3779288984648532,
      "grad_norm": 0.8389439841220695,
      "learning_rate": 2.2034899316058926e-06,
      "loss": 0.27936711311340334,
      "num_input_tokens_seen": 652256688,
      "step": 20465,
      "train_runtime": 53656.646,
      "train_tokens_per_second": 12156.121
    },
    {
      "epoch": 1.3782655534608135,
      "grad_norm": 0.7760689466241427,
      "learning_rate": 2.201298473894854e-06,
      "loss": 0.2952296733856201,
      "num_input_tokens_seen": 652413392,
      "step": 20470,
      "train_runtime": 53671.0839,
      "train_tokens_per_second": 12155.771
    },
    {
      "epoch": 1.3786022084567735,
      "grad_norm": 0.8029590636209288,
      "learning_rate": 2.1991077988317723e-06,
      "loss": 0.2822009801864624,
      "num_input_tokens_seen": 652576280,
      "step": 20475,
      "train_runtime": 53683.8633,
      "train_tokens_per_second": 12155.911
    },
    {
      "epoch": 1.3789388634527335,
      "grad_norm": 0.7636519039850963,
      "learning_rate": 2.1969179070292637e-06,
      "loss": 0.2849242925643921,
      "num_input_tokens_seen": 652736336,
      "step": 20480,
      "train_runtime": 53696.8706,
      "train_tokens_per_second": 12155.947
    },
    {
      "epoch": 1.3792755184486938,
      "grad_norm": 0.8570378389404146,
      "learning_rate": 2.1947287990997236e-06,
      "loss": 0.3027640342712402,
      "num_input_tokens_seen": 652894400,
      "step": 20485,
      "train_runtime": 53710.8954,
      "train_tokens_per_second": 12155.716
    },
    {
      "epoch": 1.379612173444654,
      "grad_norm": 0.8554262849684636,
      "learning_rate": 2.192540475655328e-06,
      "loss": 0.2953397274017334,
      "num_input_tokens_seen": 653049456,
      "step": 20490,
      "train_runtime": 53724.3702,
      "train_tokens_per_second": 12155.554
    },
    {
      "epoch": 1.379948828440614,
      "grad_norm": 0.8026316210684016,
      "learning_rate": 2.1903529373080397e-06,
      "loss": 0.3119431257247925,
      "num_input_tokens_seen": 653210608,
      "step": 20495,
      "train_runtime": 53737.5531,
      "train_tokens_per_second": 12155.57
    },
    {
      "epoch": 1.380285483436574,
      "grad_norm": 0.8228897802583538,
      "learning_rate": 2.1881661846695895e-06,
      "loss": 0.2704279899597168,
      "num_input_tokens_seen": 653357880,
      "step": 20500,
      "train_runtime": 53750.6258,
      "train_tokens_per_second": 12155.354
    },
    {
      "epoch": 1.3806221384325343,
      "grad_norm": 0.8509726085801045,
      "learning_rate": 2.185980218351503e-06,
      "loss": 0.3270567417144775,
      "num_input_tokens_seen": 653517008,
      "step": 20505,
      "train_runtime": 53763.8635,
      "train_tokens_per_second": 12155.321
    },
    {
      "epoch": 1.3809587934284946,
      "grad_norm": 0.9120062736551066,
      "learning_rate": 2.1837950389650714e-06,
      "loss": 0.28235254287719724,
      "num_input_tokens_seen": 653680256,
      "step": 20510,
      "train_runtime": 53776.7675,
      "train_tokens_per_second": 12155.44
    },
    {
      "epoch": 1.3812954484244546,
      "grad_norm": 0.7555083290841399,
      "learning_rate": 2.1816106471213787e-06,
      "loss": 0.2907303810119629,
      "num_input_tokens_seen": 653844096,
      "step": 20515,
      "train_runtime": 53789.2225,
      "train_tokens_per_second": 12155.671
    },
    {
      "epoch": 1.3816321034204146,
      "grad_norm": 0.866327021994922,
      "learning_rate": 2.179427043431282e-06,
      "loss": 0.32350399494171145,
      "num_input_tokens_seen": 654005408,
      "step": 20520,
      "train_runtime": 53802.535,
      "train_tokens_per_second": 12155.662
    },
    {
      "epoch": 1.381968758416375,
      "grad_norm": 0.7528383816585712,
      "learning_rate": 2.177244228505418e-06,
      "loss": 0.3229414939880371,
      "num_input_tokens_seen": 654161520,
      "step": 20525,
      "train_runtime": 53816.0358,
      "train_tokens_per_second": 12155.513
    },
    {
      "epoch": 1.3823054134123351,
      "grad_norm": 0.9989225028687713,
      "learning_rate": 2.1750622029542047e-06,
      "loss": 0.2938276767730713,
      "num_input_tokens_seen": 654324464,
      "step": 20530,
      "train_runtime": 53828.509,
      "train_tokens_per_second": 12155.723
    },
    {
      "epoch": 1.3826420684082952,
      "grad_norm": 0.838121570528202,
      "learning_rate": 2.1728809673878374e-06,
      "loss": 0.2945178270339966,
      "num_input_tokens_seen": 654483440,
      "step": 20535,
      "train_runtime": 53842.0088,
      "train_tokens_per_second": 12155.628
    },
    {
      "epoch": 1.3829787234042552,
      "grad_norm": 0.9043744828770097,
      "learning_rate": 2.1707005224162915e-06,
      "loss": 0.29343178272247317,
      "num_input_tokens_seen": 654645600,
      "step": 20540,
      "train_runtime": 53855.3782,
      "train_tokens_per_second": 12155.622
    },
    {
      "epoch": 1.3833153784002155,
      "grad_norm": 0.8046123406161302,
      "learning_rate": 2.1685208686493257e-06,
      "loss": 0.2951530933380127,
      "num_input_tokens_seen": 654799576,
      "step": 20545,
      "train_runtime": 53868.3171,
      "train_tokens_per_second": 12155.56
    },
    {
      "epoch": 1.3836520333961757,
      "grad_norm": 0.7837290344006682,
      "learning_rate": 2.166342006696467e-06,
      "loss": 0.3003102779388428,
      "num_input_tokens_seen": 654953616,
      "step": 20550,
      "train_runtime": 53881.4343,
      "train_tokens_per_second": 12155.46
    },
    {
      "epoch": 1.3839886883921357,
      "grad_norm": 0.8419208049155158,
      "learning_rate": 2.1641639371670335e-06,
      "loss": 0.3069209098815918,
      "num_input_tokens_seen": 655114608,
      "step": 20555,
      "train_runtime": 53895.609,
      "train_tokens_per_second": 12155.25
    },
    {
      "epoch": 1.3843253433880958,
      "grad_norm": 1.130520302817265,
      "learning_rate": 2.161986660670109e-06,
      "loss": 0.3187168836593628,
      "num_input_tokens_seen": 655277784,
      "step": 20560,
      "train_runtime": 53908.0472,
      "train_tokens_per_second": 12155.472
    },
    {
      "epoch": 1.384661998384056,
      "grad_norm": 0.7371524483213872,
      "learning_rate": 2.159810177814566e-06,
      "loss": 0.29647324085235593,
      "num_input_tokens_seen": 655434416,
      "step": 20565,
      "train_runtime": 53921.427,
      "train_tokens_per_second": 12155.361
    },
    {
      "epoch": 1.3849986533800163,
      "grad_norm": 0.8948364673197701,
      "learning_rate": 2.1576344892090507e-06,
      "loss": 0.3221231460571289,
      "num_input_tokens_seen": 655590320,
      "step": 20570,
      "train_runtime": 53934.6702,
      "train_tokens_per_second": 12155.267
    },
    {
      "epoch": 1.3853353083759763,
      "grad_norm": 0.7702079275267318,
      "learning_rate": 2.1554595954619866e-06,
      "loss": 0.3005852699279785,
      "num_input_tokens_seen": 655749448,
      "step": 20575,
      "train_runtime": 53947.9241,
      "train_tokens_per_second": 12155.23
    },
    {
      "epoch": 1.3856719633719363,
      "grad_norm": 0.8232806743748647,
      "learning_rate": 2.1532854971815763e-06,
      "loss": 0.2936416149139404,
      "num_input_tokens_seen": 655908288,
      "step": 20580,
      "train_runtime": 53961.568,
      "train_tokens_per_second": 12155.101
    },
    {
      "epoch": 1.3860086183678966,
      "grad_norm": 0.6873782863749368,
      "learning_rate": 2.151112194975799e-06,
      "loss": 0.27318587303161623,
      "num_input_tokens_seen": 656071008,
      "step": 20585,
      "train_runtime": 53974.016,
      "train_tokens_per_second": 12155.312
    },
    {
      "epoch": 1.3863452733638568,
      "grad_norm": 0.6943274888202149,
      "learning_rate": 2.1489396894524106e-06,
      "loss": 0.27690088748931885,
      "num_input_tokens_seen": 656234440,
      "step": 20590,
      "train_runtime": 53987.0359,
      "train_tokens_per_second": 12155.408
    },
    {
      "epoch": 1.3866819283598169,
      "grad_norm": 0.7532789543825452,
      "learning_rate": 2.1467679812189497e-06,
      "loss": 0.2981242656707764,
      "num_input_tokens_seen": 656390440,
      "step": 20595,
      "train_runtime": 54000.0767,
      "train_tokens_per_second": 12155.361
    },
    {
      "epoch": 1.3870185833557769,
      "grad_norm": 0.8307455733839373,
      "learning_rate": 2.1445970708827213e-06,
      "loss": 0.3125548124313354,
      "num_input_tokens_seen": 656546320,
      "step": 20600,
      "train_runtime": 54013.7769,
      "train_tokens_per_second": 12155.164
    },
    {
      "epoch": 1.3873552383517371,
      "grad_norm": 0.7249972270232963,
      "learning_rate": 2.14242695905082e-06,
      "loss": 0.2849388122558594,
      "num_input_tokens_seen": 656705688,
      "step": 20605,
      "train_runtime": 54027.2556,
      "train_tokens_per_second": 12155.081
    },
    {
      "epoch": 1.3876918933476974,
      "grad_norm": 0.893174570008363,
      "learning_rate": 2.1402576463301044e-06,
      "loss": 0.3187171459197998,
      "num_input_tokens_seen": 656869528,
      "step": 20610,
      "train_runtime": 54039.6983,
      "train_tokens_per_second": 12155.314
    },
    {
      "epoch": 1.3880285483436574,
      "grad_norm": 0.8678460832552309,
      "learning_rate": 2.13808913332722e-06,
      "loss": 0.2718982458114624,
      "num_input_tokens_seen": 657027912,
      "step": 20615,
      "train_runtime": 54052.7282,
      "train_tokens_per_second": 12155.315
    },
    {
      "epoch": 1.3883652033396174,
      "grad_norm": 0.8579744815924049,
      "learning_rate": 2.1359214206485845e-06,
      "loss": 0.3030441999435425,
      "num_input_tokens_seen": 657185056,
      "step": 20620,
      "train_runtime": 54065.7949,
      "train_tokens_per_second": 12155.283
    },
    {
      "epoch": 1.3887018583355777,
      "grad_norm": 0.7666784497696234,
      "learning_rate": 2.1337545089003905e-06,
      "loss": 0.2615054607391357,
      "num_input_tokens_seen": 657338640,
      "step": 20625,
      "train_runtime": 54079.2954,
      "train_tokens_per_second": 12155.089
    },
    {
      "epoch": 1.389038513331538,
      "grad_norm": 0.8530811699509007,
      "learning_rate": 2.131588398688608e-06,
      "loss": 0.3017681121826172,
      "num_input_tokens_seen": 657498864,
      "step": 20630,
      "train_runtime": 54092.5207,
      "train_tokens_per_second": 12155.079
    },
    {
      "epoch": 1.389375168327498,
      "grad_norm": 0.745741005772621,
      "learning_rate": 2.1294230906189833e-06,
      "loss": 0.2829160213470459,
      "num_input_tokens_seen": 657654664,
      "step": 20635,
      "train_runtime": 54105.1274,
      "train_tokens_per_second": 12155.126
    },
    {
      "epoch": 1.389711823323458,
      "grad_norm": 0.7559887447939505,
      "learning_rate": 2.1272585852970358e-06,
      "loss": 0.28513526916503906,
      "num_input_tokens_seen": 657811880,
      "step": 20640,
      "train_runtime": 54118.4367,
      "train_tokens_per_second": 12155.042
    },
    {
      "epoch": 1.3900484783194182,
      "grad_norm": 0.8485092469868943,
      "learning_rate": 2.125094883328068e-06,
      "loss": 0.2961697578430176,
      "num_input_tokens_seen": 657967248,
      "step": 20645,
      "train_runtime": 54131.878,
      "train_tokens_per_second": 12154.894
    },
    {
      "epoch": 1.3903851333153785,
      "grad_norm": 0.8992611374022871,
      "learning_rate": 2.1229319853171447e-06,
      "loss": 0.27866368293762206,
      "num_input_tokens_seen": 658122768,
      "step": 20650,
      "train_runtime": 54145.4947,
      "train_tokens_per_second": 12154.71
    },
    {
      "epoch": 1.3907217883113385,
      "grad_norm": 0.876485330450587,
      "learning_rate": 2.1207698918691217e-06,
      "loss": 0.31116452217102053,
      "num_input_tokens_seen": 658284032,
      "step": 20655,
      "train_runtime": 54159.2801,
      "train_tokens_per_second": 12154.593
    },
    {
      "epoch": 1.3910584433072986,
      "grad_norm": 0.8895435894322041,
      "learning_rate": 2.1186086035886133e-06,
      "loss": 0.29702067375183105,
      "num_input_tokens_seen": 658445288,
      "step": 20660,
      "train_runtime": 54172.2193,
      "train_tokens_per_second": 12154.667
    },
    {
      "epoch": 1.3913950983032588,
      "grad_norm": 0.8925336299109211,
      "learning_rate": 2.116448121080023e-06,
      "loss": 0.3145890712738037,
      "num_input_tokens_seen": 658609128,
      "step": 20665,
      "train_runtime": 54184.6694,
      "train_tokens_per_second": 12154.898
    },
    {
      "epoch": 1.391731753299219,
      "grad_norm": 0.8412894438536123,
      "learning_rate": 2.1142884449475205e-06,
      "loss": 0.3074498653411865,
      "num_input_tokens_seen": 658771384,
      "step": 20670,
      "train_runtime": 54197.3103,
      "train_tokens_per_second": 12155.057
    },
    {
      "epoch": 1.392068408295179,
      "grad_norm": 0.8060827585149174,
      "learning_rate": 2.112129575795052e-06,
      "loss": 0.29236836433410646,
      "num_input_tokens_seen": 658932776,
      "step": 20675,
      "train_runtime": 54210.3732,
      "train_tokens_per_second": 12155.105
    },
    {
      "epoch": 1.3924050632911391,
      "grad_norm": 0.7592720631038189,
      "learning_rate": 2.1099715142263394e-06,
      "loss": 0.28681292533874514,
      "num_input_tokens_seen": 659090384,
      "step": 20680,
      "train_runtime": 54223.2315,
      "train_tokens_per_second": 12155.129
    },
    {
      "epoch": 1.3927417182870994,
      "grad_norm": 0.7166519857589716,
      "learning_rate": 2.1078142608448773e-06,
      "loss": 0.2725334644317627,
      "num_input_tokens_seen": 659252040,
      "step": 20685,
      "train_runtime": 54236.4617,
      "train_tokens_per_second": 12155.145
    },
    {
      "epoch": 1.3930783732830596,
      "grad_norm": 0.9232562553788313,
      "learning_rate": 2.1056578162539327e-06,
      "loss": 0.31885356903076173,
      "num_input_tokens_seen": 659411592,
      "step": 20690,
      "train_runtime": 54248.9923,
      "train_tokens_per_second": 12155.278
    },
    {
      "epoch": 1.3934150282790196,
      "grad_norm": 0.8297819435919894,
      "learning_rate": 2.103502181056554e-06,
      "loss": 0.30214226245880127,
      "num_input_tokens_seen": 659575088,
      "step": 20695,
      "train_runtime": 54262.0344,
      "train_tokens_per_second": 12155.37
    },
    {
      "epoch": 1.3937516832749797,
      "grad_norm": 0.8241283525710857,
      "learning_rate": 2.1013473558555503e-06,
      "loss": 0.289078426361084,
      "num_input_tokens_seen": 659738928,
      "step": 20700,
      "train_runtime": 54274.479,
      "train_tokens_per_second": 12155.601
    },
    {
      "epoch": 1.39408833827094,
      "grad_norm": 0.7305413380835792,
      "learning_rate": 2.099193341253519e-06,
      "loss": 0.29375219345092773,
      "num_input_tokens_seen": 659897272,
      "step": 20705,
      "train_runtime": 54288.4866,
      "train_tokens_per_second": 12155.382
    },
    {
      "epoch": 1.3944249932669002,
      "grad_norm": 0.7506548867882089,
      "learning_rate": 2.0970401378528167e-06,
      "loss": 0.28227407932281495,
      "num_input_tokens_seen": 660050824,
      "step": 20710,
      "train_runtime": 54301.1669,
      "train_tokens_per_second": 12155.371
    },
    {
      "epoch": 1.3947616482628602,
      "grad_norm": 0.7479026187231906,
      "learning_rate": 2.0948877462555844e-06,
      "loss": 0.2800161838531494,
      "num_input_tokens_seen": 660210016,
      "step": 20715,
      "train_runtime": 54314.4045,
      "train_tokens_per_second": 12155.339
    },
    {
      "epoch": 1.3950983032588202,
      "grad_norm": 0.7679638071694884,
      "learning_rate": 2.09273616706373e-06,
      "loss": 0.30209167003631593,
      "num_input_tokens_seen": 660370848,
      "step": 20720,
      "train_runtime": 54326.8626,
      "train_tokens_per_second": 12155.512
    },
    {
      "epoch": 1.3954349582547805,
      "grad_norm": 0.7955495021385083,
      "learning_rate": 2.090585400878936e-06,
      "loss": 0.31611971855163573,
      "num_input_tokens_seen": 660531464,
      "step": 20725,
      "train_runtime": 54339.698,
      "train_tokens_per_second": 12155.597
    },
    {
      "epoch": 1.3957716132507407,
      "grad_norm": 0.8615853389553981,
      "learning_rate": 2.0884354483026576e-06,
      "loss": 0.3125169038772583,
      "num_input_tokens_seen": 660689512,
      "step": 20730,
      "train_runtime": 54352.5476,
      "train_tokens_per_second": 12155.631
    },
    {
      "epoch": 1.3961082682467008,
      "grad_norm": 0.7579016246920184,
      "learning_rate": 2.0862863099361214e-06,
      "loss": 0.30734028816223147,
      "num_input_tokens_seen": 660853352,
      "step": 20735,
      "train_runtime": 54364.999,
      "train_tokens_per_second": 12155.861
    },
    {
      "epoch": 1.3964449232426608,
      "grad_norm": 0.817064244578435,
      "learning_rate": 2.084137986380326e-06,
      "loss": 0.33032703399658203,
      "num_input_tokens_seen": 661009088,
      "step": 20740,
      "train_runtime": 54377.7708,
      "train_tokens_per_second": 12155.87
    },
    {
      "epoch": 1.396781578238621,
      "grad_norm": 0.7876017236428002,
      "learning_rate": 2.081990478236048e-06,
      "loss": 0.27702276706695556,
      "num_input_tokens_seen": 661163752,
      "step": 20745,
      "train_runtime": 54391.8542,
      "train_tokens_per_second": 12155.566
    },
    {
      "epoch": 1.3971182332345813,
      "grad_norm": 0.8038784262169215,
      "learning_rate": 2.079843786103824e-06,
      "loss": 0.28578896522521974,
      "num_input_tokens_seen": 661324824,
      "step": 20750,
      "train_runtime": 54405.4781,
      "train_tokens_per_second": 12155.482
    },
    {
      "epoch": 1.3974548882305413,
      "grad_norm": 0.8981684674471448,
      "learning_rate": 2.0776979105839774e-06,
      "loss": 0.302164101600647,
      "num_input_tokens_seen": 661486256,
      "step": 20755,
      "train_runtime": 54417.992,
      "train_tokens_per_second": 12155.654
    },
    {
      "epoch": 1.3977915432265013,
      "grad_norm": 0.8968143466418896,
      "learning_rate": 2.075552852276588e-06,
      "loss": 0.3322463035583496,
      "num_input_tokens_seen": 661648400,
      "step": 20760,
      "train_runtime": 54431.0415,
      "train_tokens_per_second": 12155.718
    },
    {
      "epoch": 1.3981281982224616,
      "grad_norm": 0.8893680118872508,
      "learning_rate": 2.07340861178152e-06,
      "loss": 0.29132239818572997,
      "num_input_tokens_seen": 661809656,
      "step": 20765,
      "train_runtime": 54444.8401,
      "train_tokens_per_second": 12155.599
    },
    {
      "epoch": 1.3984648532184218,
      "grad_norm": 0.8298743248245876,
      "learning_rate": 2.0712651896984013e-06,
      "loss": 0.3066413879394531,
      "num_input_tokens_seen": 661973200,
      "step": 20770,
      "train_runtime": 54457.8899,
      "train_tokens_per_second": 12155.689
    },
    {
      "epoch": 1.3988015082143819,
      "grad_norm": 0.8177213771127584,
      "learning_rate": 2.0691225866266335e-06,
      "loss": 0.3036322593688965,
      "num_input_tokens_seen": 662128168,
      "step": 20775,
      "train_runtime": 54470.4473,
      "train_tokens_per_second": 12155.732
    },
    {
      "epoch": 1.3991381632103421,
      "grad_norm": 0.8608732159606955,
      "learning_rate": 2.066980803165388e-06,
      "loss": 0.3045391798019409,
      "num_input_tokens_seen": 662287984,
      "step": 20780,
      "train_runtime": 54483.1962,
      "train_tokens_per_second": 12155.821
    },
    {
      "epoch": 1.3994748182063022,
      "grad_norm": 0.8551144618532229,
      "learning_rate": 2.0648398399136088e-06,
      "loss": 0.30803370475769043,
      "num_input_tokens_seen": 662448088,
      "step": 20785,
      "train_runtime": 54496.2577,
      "train_tokens_per_second": 12155.845
    },
    {
      "epoch": 1.3998114732022624,
      "grad_norm": 0.73451889551276,
      "learning_rate": 2.062699697470007e-06,
      "loss": 0.2977804899215698,
      "num_input_tokens_seen": 662606256,
      "step": 20790,
      "train_runtime": 54509.7638,
      "train_tokens_per_second": 12155.735
    },
    {
      "epoch": 1.4001481281982224,
      "grad_norm": 0.8280462471142737,
      "learning_rate": 2.060560376433072e-06,
      "loss": 0.3162750482559204,
      "num_input_tokens_seen": 662769272,
      "step": 20795,
      "train_runtime": 54523.211,
      "train_tokens_per_second": 12155.727
    },
    {
      "epoch": 1.4004847831941827,
      "grad_norm": 0.7173336569466907,
      "learning_rate": 2.05842187740105e-06,
      "loss": 0.3020288944244385,
      "num_input_tokens_seen": 662928296,
      "step": 20800,
      "train_runtime": 54536.6604,
      "train_tokens_per_second": 12155.645
    },
    {
      "epoch": 1.4008214381901427,
      "grad_norm": 0.777341071381465,
      "learning_rate": 2.0562842009719745e-06,
      "loss": 0.2713037967681885,
      "num_input_tokens_seen": 663075064,
      "step": 20805,
      "train_runtime": 54550.0195,
      "train_tokens_per_second": 12155.359
    },
    {
      "epoch": 1.401158093186103,
      "grad_norm": 0.8247246098014612,
      "learning_rate": 2.0541473477436317e-06,
      "loss": 0.27657513618469237,
      "num_input_tokens_seen": 663235280,
      "step": 20810,
      "train_runtime": 54562.8843,
      "train_tokens_per_second": 12155.429
    },
    {
      "epoch": 1.401494748182063,
      "grad_norm": 0.8072781249572049,
      "learning_rate": 2.052011318313591e-06,
      "loss": 0.27658560276031496,
      "num_input_tokens_seen": 663397752,
      "step": 20815,
      "train_runtime": 54575.9801,
      "train_tokens_per_second": 12155.489
    },
    {
      "epoch": 1.4018314031780232,
      "grad_norm": 0.8187843940548012,
      "learning_rate": 2.049876113279185e-06,
      "loss": 0.3338526010513306,
      "num_input_tokens_seen": 663560984,
      "step": 20820,
      "train_runtime": 54588.8868,
      "train_tokens_per_second": 12155.606
    },
    {
      "epoch": 1.4021680581739833,
      "grad_norm": 0.7982378825561747,
      "learning_rate": 2.0477417332375165e-06,
      "loss": 0.2930063486099243,
      "num_input_tokens_seen": 663722144,
      "step": 20825,
      "train_runtime": 54601.3592,
      "train_tokens_per_second": 12155.781
    },
    {
      "epoch": 1.4025047131699435,
      "grad_norm": 0.7933946412250249,
      "learning_rate": 2.045608178785458e-06,
      "loss": 0.2954334020614624,
      "num_input_tokens_seen": 663884232,
      "step": 20830,
      "train_runtime": 54614.5624,
      "train_tokens_per_second": 12155.81
    },
    {
      "epoch": 1.4028413681659035,
      "grad_norm": 0.8002124374409366,
      "learning_rate": 2.043475450519652e-06,
      "loss": 0.2986750602722168,
      "num_input_tokens_seen": 664043592,
      "step": 20835,
      "train_runtime": 54628.4642,
      "train_tokens_per_second": 12155.634
    },
    {
      "epoch": 1.4031780231618638,
      "grad_norm": 0.8306370883498027,
      "learning_rate": 2.0413435490365064e-06,
      "loss": 0.28760695457458496,
      "num_input_tokens_seen": 664198592,
      "step": 20840,
      "train_runtime": 54641.8279,
      "train_tokens_per_second": 12155.497
    },
    {
      "epoch": 1.4035146781578238,
      "grad_norm": 0.8507119929053484,
      "learning_rate": 2.0392124749322064e-06,
      "loss": 0.3281423330307007,
      "num_input_tokens_seen": 664361728,
      "step": 20845,
      "train_runtime": 54654.6558,
      "train_tokens_per_second": 12155.629
    },
    {
      "epoch": 1.403851333153784,
      "grad_norm": 0.8139173527877805,
      "learning_rate": 2.037082228802693e-06,
      "loss": 0.2736141443252563,
      "num_input_tokens_seen": 664522344,
      "step": 20850,
      "train_runtime": 54668.1191,
      "train_tokens_per_second": 12155.574
    },
    {
      "epoch": 1.404187988149744,
      "grad_norm": 0.7131337798046449,
      "learning_rate": 2.03495281124369e-06,
      "loss": 0.27420663833618164,
      "num_input_tokens_seen": 664680752,
      "step": 20855,
      "train_runtime": 54680.8817,
      "train_tokens_per_second": 12155.633
    },
    {
      "epoch": 1.4045246431457044,
      "grad_norm": 0.7379177529244906,
      "learning_rate": 2.0328242228506746e-06,
      "loss": 0.2869014024734497,
      "num_input_tokens_seen": 664841096,
      "step": 20860,
      "train_runtime": 54694.2912,
      "train_tokens_per_second": 12155.585
    },
    {
      "epoch": 1.4048612981416644,
      "grad_norm": 0.8903796633752868,
      "learning_rate": 2.0306964642189053e-06,
      "loss": 0.2895898580551147,
      "num_input_tokens_seen": 664998856,
      "step": 20865,
      "train_runtime": 54706.9827,
      "train_tokens_per_second": 12155.649
    },
    {
      "epoch": 1.4051979531376246,
      "grad_norm": 0.7547859285642092,
      "learning_rate": 2.028569535943402e-06,
      "loss": 0.28022797107696534,
      "num_input_tokens_seen": 665156200,
      "step": 20870,
      "train_runtime": 54720.264,
      "train_tokens_per_second": 12155.574
    },
    {
      "epoch": 1.4055346081335847,
      "grad_norm": 1.0717408356055804,
      "learning_rate": 2.0264434386189524e-06,
      "loss": 0.2991631507873535,
      "num_input_tokens_seen": 665311160,
      "step": 20875,
      "train_runtime": 54733.3379,
      "train_tokens_per_second": 12155.501
    },
    {
      "epoch": 1.405871263129545,
      "grad_norm": 0.7343491213782462,
      "learning_rate": 2.0243181728401136e-06,
      "loss": 0.2980818510055542,
      "num_input_tokens_seen": 665473592,
      "step": 20880,
      "train_runtime": 54747.6385,
      "train_tokens_per_second": 12155.293
    },
    {
      "epoch": 1.406207918125505,
      "grad_norm": 0.8527889717640886,
      "learning_rate": 2.022193739201208e-06,
      "loss": 0.3042200803756714,
      "num_input_tokens_seen": 665625536,
      "step": 20885,
      "train_runtime": 54760.5554,
      "train_tokens_per_second": 12155.201
    },
    {
      "epoch": 1.4065445731214652,
      "grad_norm": 0.7573194751458604,
      "learning_rate": 2.0200701382963267e-06,
      "loss": 0.2779937744140625,
      "num_input_tokens_seen": 665783392,
      "step": 20890,
      "train_runtime": 54773.3548,
      "train_tokens_per_second": 12155.242
    },
    {
      "epoch": 1.4068812281174252,
      "grad_norm": 0.7664024422424403,
      "learning_rate": 2.0179473707193324e-06,
      "loss": 0.28447072505950927,
      "num_input_tokens_seen": 665946920,
      "step": 20895,
      "train_runtime": 54786.3763,
      "train_tokens_per_second": 12155.338
    },
    {
      "epoch": 1.4072178831133855,
      "grad_norm": 0.7051923882784243,
      "learning_rate": 2.0158254370638435e-06,
      "loss": 0.2863420009613037,
      "num_input_tokens_seen": 666105736,
      "step": 20900,
      "train_runtime": 54799.047,
      "train_tokens_per_second": 12155.426
    },
    {
      "epoch": 1.4075545381093455,
      "grad_norm": 0.7311150450776002,
      "learning_rate": 2.0137043379232585e-06,
      "loss": 0.29042930603027345,
      "num_input_tokens_seen": 666266752,
      "step": 20905,
      "train_runtime": 54812.1899,
      "train_tokens_per_second": 12155.449
    },
    {
      "epoch": 1.4078911931053057,
      "grad_norm": 0.7140574961981422,
      "learning_rate": 2.0115840738907293e-06,
      "loss": 0.2854149341583252,
      "num_input_tokens_seen": 666424320,
      "step": 20910,
      "train_runtime": 54826.0899,
      "train_tokens_per_second": 12155.241
    },
    {
      "epoch": 1.4082278481012658,
      "grad_norm": 0.8091024196731981,
      "learning_rate": 2.0094646455591865e-06,
      "loss": 0.2945026636123657,
      "num_input_tokens_seen": 666586280,
      "step": 20915,
      "train_runtime": 54839.2569,
      "train_tokens_per_second": 12155.276
    },
    {
      "epoch": 1.408564503097226,
      "grad_norm": 0.7489971328317317,
      "learning_rate": 2.007346053521319e-06,
      "loss": 0.31034603118896487,
      "num_input_tokens_seen": 666746312,
      "step": 20920,
      "train_runtime": 54852.016,
      "train_tokens_per_second": 12155.366
    },
    {
      "epoch": 1.408901158093186,
      "grad_norm": 0.8611253904666468,
      "learning_rate": 2.005228298369583e-06,
      "loss": 0.3167555809020996,
      "num_input_tokens_seen": 666897808,
      "step": 20925,
      "train_runtime": 54864.6164,
      "train_tokens_per_second": 12155.335
    },
    {
      "epoch": 1.4092378130891463,
      "grad_norm": 0.8063060042735796,
      "learning_rate": 2.003111380696208e-06,
      "loss": 0.27892022132873534,
      "num_input_tokens_seen": 667060880,
      "step": 20930,
      "train_runtime": 54877.3868,
      "train_tokens_per_second": 12155.478
    },
    {
      "epoch": 1.4095744680851063,
      "grad_norm": 0.906936730667048,
      "learning_rate": 2.0009953010931747e-06,
      "loss": 0.29097919464111327,
      "num_input_tokens_seen": 667223920,
      "step": 20935,
      "train_runtime": 54890.3957,
      "train_tokens_per_second": 12155.568
    },
    {
      "epoch": 1.4099111230810666,
      "grad_norm": 0.85967155792743,
      "learning_rate": 1.998880060152244e-06,
      "loss": 0.3286836862564087,
      "num_input_tokens_seen": 667383176,
      "step": 20940,
      "train_runtime": 54902.9715,
      "train_tokens_per_second": 12155.684
    },
    {
      "epoch": 1.4102477780770266,
      "grad_norm": 0.8202102122420725,
      "learning_rate": 1.9967656584649354e-06,
      "loss": 0.3003001928329468,
      "num_input_tokens_seen": 667542840,
      "step": 20945,
      "train_runtime": 54916.132,
      "train_tokens_per_second": 12155.678
    },
    {
      "epoch": 1.4105844330729869,
      "grad_norm": 0.7859309374999206,
      "learning_rate": 1.994652096622533e-06,
      "loss": 0.31295218467712405,
      "num_input_tokens_seen": 667702752,
      "step": 20950,
      "train_runtime": 54929.2172,
      "train_tokens_per_second": 12155.694
    },
    {
      "epoch": 1.410921088068947,
      "grad_norm": 0.7823562268201744,
      "learning_rate": 1.992539375216089e-06,
      "loss": 0.30660452842712405,
      "num_input_tokens_seen": 667866592,
      "step": 20955,
      "train_runtime": 54941.65,
      "train_tokens_per_second": 12155.925
    },
    {
      "epoch": 1.4112577430649071,
      "grad_norm": 0.7696996231647802,
      "learning_rate": 1.9904274948364184e-06,
      "loss": 0.2958907842636108,
      "num_input_tokens_seen": 668027600,
      "step": 20960,
      "train_runtime": 54955.1517,
      "train_tokens_per_second": 12155.869
    },
    {
      "epoch": 1.4115943980608672,
      "grad_norm": 0.918883900786541,
      "learning_rate": 1.988316456074102e-06,
      "loss": 0.3099415063858032,
      "num_input_tokens_seen": 668191440,
      "step": 20965,
      "train_runtime": 54967.578,
      "train_tokens_per_second": 12156.101
    },
    {
      "epoch": 1.4119310530568274,
      "grad_norm": 0.7918069853503352,
      "learning_rate": 1.9862062595194854e-06,
      "loss": 0.3219482898712158,
      "num_input_tokens_seen": 668348688,
      "step": 20970,
      "train_runtime": 54981.03,
      "train_tokens_per_second": 12155.987
    },
    {
      "epoch": 1.4122677080527875,
      "grad_norm": 0.8746414376875723,
      "learning_rate": 1.984096905762676e-06,
      "loss": 0.3007672786712646,
      "num_input_tokens_seen": 668507072,
      "step": 20975,
      "train_runtime": 54994.023,
      "train_tokens_per_second": 12155.995
    },
    {
      "epoch": 1.4126043630487477,
      "grad_norm": 0.8007824483939163,
      "learning_rate": 1.9819883953935533e-06,
      "loss": 0.3088326930999756,
      "num_input_tokens_seen": 668670016,
      "step": 20980,
      "train_runtime": 55006.7425,
      "train_tokens_per_second": 12156.146
    },
    {
      "epoch": 1.4129410180447077,
      "grad_norm": 0.7940277605983056,
      "learning_rate": 1.9798807290017485e-06,
      "loss": 0.2929655075073242,
      "num_input_tokens_seen": 668826368,
      "step": 20985,
      "train_runtime": 55019.4064,
      "train_tokens_per_second": 12156.19
    },
    {
      "epoch": 1.413277673040668,
      "grad_norm": 0.7713319525670047,
      "learning_rate": 1.9777739071766693e-06,
      "loss": 0.3011338472366333,
      "num_input_tokens_seen": 668984720,
      "step": 20990,
      "train_runtime": 55032.6868,
      "train_tokens_per_second": 12156.134
    },
    {
      "epoch": 1.413614328036628,
      "grad_norm": 0.777433206310864,
      "learning_rate": 1.97566793050748e-06,
      "loss": 0.2870569944381714,
      "num_input_tokens_seen": 669139032,
      "step": 20995,
      "train_runtime": 55045.8846,
      "train_tokens_per_second": 12156.023
    },
    {
      "epoch": 1.4139509830325883,
      "grad_norm": 0.837211950632524,
      "learning_rate": 1.9735627995831096e-06,
      "loss": 0.2744601726531982,
      "num_input_tokens_seen": 669300224,
      "step": 21000,
      "train_runtime": 55058.4522,
      "train_tokens_per_second": 12156.176
    },
    {
      "epoch": 1.4142876380285483,
      "grad_norm": 0.8665244008838211,
      "learning_rate": 1.9714585149922523e-06,
      "loss": 0.31123199462890627,
      "num_input_tokens_seen": 669464040,
      "step": 21005,
      "train_runtime": 55070.8805,
      "train_tokens_per_second": 12156.407
    },
    {
      "epoch": 1.4146242930245085,
      "grad_norm": 0.8056199433128648,
      "learning_rate": 1.9693550773233634e-06,
      "loss": 0.28864521980285646,
      "num_input_tokens_seen": 669623568,
      "step": 21010,
      "train_runtime": 55083.4977,
      "train_tokens_per_second": 12156.519
    },
    {
      "epoch": 1.4149609480204686,
      "grad_norm": 1.0143001505034885,
      "learning_rate": 1.967252487164663e-06,
      "loss": 0.2820691347122192,
      "num_input_tokens_seen": 669784880,
      "step": 21015,
      "train_runtime": 55096.7466,
      "train_tokens_per_second": 12156.523
    },
    {
      "epoch": 1.4152976030164288,
      "grad_norm": 0.8439838637247834,
      "learning_rate": 1.9651507451041344e-06,
      "loss": 0.3045512199401855,
      "num_input_tokens_seen": 669944648,
      "step": 21020,
      "train_runtime": 55110.3362,
      "train_tokens_per_second": 12156.425
    },
    {
      "epoch": 1.4156342580123888,
      "grad_norm": 0.8614437995540749,
      "learning_rate": 1.9630498517295204e-06,
      "loss": 0.29846189022064207,
      "num_input_tokens_seen": 670107408,
      "step": 21025,
      "train_runtime": 55123.435,
      "train_tokens_per_second": 12156.489
    },
    {
      "epoch": 1.415970913008349,
      "grad_norm": 0.8494136532817852,
      "learning_rate": 1.960949807628335e-06,
      "loss": 0.2990385055541992,
      "num_input_tokens_seen": 670268280,
      "step": 21030,
      "train_runtime": 55136.5777,
      "train_tokens_per_second": 12156.509
    },
    {
      "epoch": 1.4163075680043091,
      "grad_norm": 0.810379912538388,
      "learning_rate": 1.9588506133878415e-06,
      "loss": 0.33489117622375486,
      "num_input_tokens_seen": 670427256,
      "step": 21035,
      "train_runtime": 55149.6884,
      "train_tokens_per_second": 12156.501
    },
    {
      "epoch": 1.4166442230002694,
      "grad_norm": 0.7628748636828452,
      "learning_rate": 1.956752269595078e-06,
      "loss": 0.311111307144165,
      "num_input_tokens_seen": 670588248,
      "step": 21040,
      "train_runtime": 55162.9955,
      "train_tokens_per_second": 12156.487
    },
    {
      "epoch": 1.4169808779962294,
      "grad_norm": 0.8270682327454322,
      "learning_rate": 1.9546547768368384e-06,
      "loss": 0.28902468681335447,
      "num_input_tokens_seen": 670750080,
      "step": 21045,
      "train_runtime": 55175.5346,
      "train_tokens_per_second": 12156.658
    },
    {
      "epoch": 1.4173175329921897,
      "grad_norm": 0.9340835893032142,
      "learning_rate": 1.9525581356996793e-06,
      "loss": 0.290776801109314,
      "num_input_tokens_seen": 670911752,
      "step": 21050,
      "train_runtime": 55188.7907,
      "train_tokens_per_second": 12156.667
    },
    {
      "epoch": 1.4176541879881497,
      "grad_norm": 0.8447282940494785,
      "learning_rate": 1.95046234676992e-06,
      "loss": 0.2880901575088501,
      "num_input_tokens_seen": 671075592,
      "step": 21055,
      "train_runtime": 55201.2218,
      "train_tokens_per_second": 12156.897
    },
    {
      "epoch": 1.41799084298411,
      "grad_norm": 0.8254266242570487,
      "learning_rate": 1.9483674106336416e-06,
      "loss": 0.2703708171844482,
      "num_input_tokens_seen": 671228712,
      "step": 21060,
      "train_runtime": 55213.9143,
      "train_tokens_per_second": 12156.876
    },
    {
      "epoch": 1.41832749798007,
      "grad_norm": 0.7737329220955402,
      "learning_rate": 1.9462733278766853e-06,
      "loss": 0.29945170879364014,
      "num_input_tokens_seen": 671387816,
      "step": 21065,
      "train_runtime": 55226.8907,
      "train_tokens_per_second": 12156.901
    },
    {
      "epoch": 1.4186641529760302,
      "grad_norm": 0.8138047644878008,
      "learning_rate": 1.9441800990846553e-06,
      "loss": 0.2919990062713623,
      "num_input_tokens_seen": 671546744,
      "step": 21070,
      "train_runtime": 55239.4876,
      "train_tokens_per_second": 12157.005
    },
    {
      "epoch": 1.4190008079719902,
      "grad_norm": 0.7519332793938329,
      "learning_rate": 1.942087724842915e-06,
      "loss": 0.3107193946838379,
      "num_input_tokens_seen": 671701408,
      "step": 21075,
      "train_runtime": 55252.6722,
      "train_tokens_per_second": 12156.904
    },
    {
      "epoch": 1.4193374629679505,
      "grad_norm": 0.8664060770979677,
      "learning_rate": 1.9399962057365944e-06,
      "loss": 0.2897336959838867,
      "num_input_tokens_seen": 671855264,
      "step": 21080,
      "train_runtime": 55266.1698,
      "train_tokens_per_second": 12156.718
    },
    {
      "epoch": 1.4196741179639105,
      "grad_norm": 0.8240649286629831,
      "learning_rate": 1.937905542350574e-06,
      "loss": 0.3101287603378296,
      "num_input_tokens_seen": 672016216,
      "step": 21085,
      "train_runtime": 55279.0619,
      "train_tokens_per_second": 12156.795
    },
    {
      "epoch": 1.4200107729598708,
      "grad_norm": 0.899627650566381,
      "learning_rate": 1.935815735269506e-06,
      "loss": 0.31273272037506106,
      "num_input_tokens_seen": 672173280,
      "step": 21090,
      "train_runtime": 55292.0327,
      "train_tokens_per_second": 12156.784
    },
    {
      "epoch": 1.4203474279558308,
      "grad_norm": 0.9294466370019883,
      "learning_rate": 1.933726785077797e-06,
      "loss": 0.3313465595245361,
      "num_input_tokens_seen": 672330608,
      "step": 21095,
      "train_runtime": 55304.4842,
      "train_tokens_per_second": 12156.891
    },
    {
      "epoch": 1.420684082951791,
      "grad_norm": 0.7881630642096581,
      "learning_rate": 1.9316386923596146e-06,
      "loss": 0.3223684310913086,
      "num_input_tokens_seen": 672487096,
      "step": 21100,
      "train_runtime": 55317.0599,
      "train_tokens_per_second": 12156.957
    },
    {
      "epoch": 1.421020737947751,
      "grad_norm": 0.8709967776505805,
      "learning_rate": 1.929551457698888e-06,
      "loss": 0.3101195335388184,
      "num_input_tokens_seen": 672645736,
      "step": 21105,
      "train_runtime": 55330.0097,
      "train_tokens_per_second": 12156.978
    },
    {
      "epoch": 1.4213573929437113,
      "grad_norm": 0.7868017586093501,
      "learning_rate": 1.9274650816793044e-06,
      "loss": 0.3017850160598755,
      "num_input_tokens_seen": 672807176,
      "step": 21110,
      "train_runtime": 55342.4293,
      "train_tokens_per_second": 12157.167
    },
    {
      "epoch": 1.4216940479396714,
      "grad_norm": 0.9329412150289632,
      "learning_rate": 1.9253795648843136e-06,
      "loss": 0.31035542488098145,
      "num_input_tokens_seen": 672965712,
      "step": 21115,
      "train_runtime": 55355.4492,
      "train_tokens_per_second": 12157.172
    },
    {
      "epoch": 1.4220307029356316,
      "grad_norm": 0.8001133470664693,
      "learning_rate": 1.9232949078971236e-06,
      "loss": 0.30742316246032714,
      "num_input_tokens_seen": 673125840,
      "step": 21120,
      "train_runtime": 55368.4085,
      "train_tokens_per_second": 12157.218
    },
    {
      "epoch": 1.4223673579315916,
      "grad_norm": 0.8044515612861367,
      "learning_rate": 1.9212111113006998e-06,
      "loss": 0.29122295379638674,
      "num_input_tokens_seen": 673289680,
      "step": 21125,
      "train_runtime": 55380.8284,
      "train_tokens_per_second": 12157.45
    },
    {
      "epoch": 1.4227040129275519,
      "grad_norm": 0.813711849466303,
      "learning_rate": 1.919128175677775e-06,
      "loss": 0.28965320587158205,
      "num_input_tokens_seen": 673447264,
      "step": 21130,
      "train_runtime": 55394.3969,
      "train_tokens_per_second": 12157.317
    },
    {
      "epoch": 1.423040667923512,
      "grad_norm": 0.8173294841163761,
      "learning_rate": 1.917046101610827e-06,
      "loss": 0.2809044122695923,
      "num_input_tokens_seen": 673607112,
      "step": 21135,
      "train_runtime": 55407.1666,
      "train_tokens_per_second": 12157.4
    },
    {
      "epoch": 1.4233773229194722,
      "grad_norm": 0.7629773189349726,
      "learning_rate": 1.9149648896821087e-06,
      "loss": 0.307040810585022,
      "num_input_tokens_seen": 673769024,
      "step": 21140,
      "train_runtime": 55420.5074,
      "train_tokens_per_second": 12157.395
    },
    {
      "epoch": 1.4237139779154322,
      "grad_norm": 0.8424367858269722,
      "learning_rate": 1.912884540473621e-06,
      "loss": 0.2925193071365356,
      "num_input_tokens_seen": 673922984,
      "step": 21145,
      "train_runtime": 55434.3897,
      "train_tokens_per_second": 12157.128
    },
    {
      "epoch": 1.4240506329113924,
      "grad_norm": 0.8802231400199286,
      "learning_rate": 1.9108050545671283e-06,
      "loss": 0.2741716384887695,
      "num_input_tokens_seen": 674079824,
      "step": 21150,
      "train_runtime": 55448.1724,
      "train_tokens_per_second": 12156.935
    },
    {
      "epoch": 1.4243872879073525,
      "grad_norm": 0.8540227414032426,
      "learning_rate": 1.9087264325441507e-06,
      "loss": 0.26904964447021484,
      "num_input_tokens_seen": 674240032,
      "step": 21155,
      "train_runtime": 55461.4638,
      "train_tokens_per_second": 12156.91
    },
    {
      "epoch": 1.4247239429033127,
      "grad_norm": 0.9163676425391353,
      "learning_rate": 1.9066486749859686e-06,
      "loss": 0.32110090255737306,
      "num_input_tokens_seen": 674396560,
      "step": 21160,
      "train_runtime": 55474.3185,
      "train_tokens_per_second": 12156.915
    },
    {
      "epoch": 1.4250605978992728,
      "grad_norm": 0.7273802265068893,
      "learning_rate": 1.9045717824736209e-06,
      "loss": 0.3160089015960693,
      "num_input_tokens_seen": 674555832,
      "step": 21165,
      "train_runtime": 55486.9148,
      "train_tokens_per_second": 12157.025
    },
    {
      "epoch": 1.425397252895233,
      "grad_norm": 0.9264693794919668,
      "learning_rate": 1.9024957555879037e-06,
      "loss": 0.3039489507675171,
      "num_input_tokens_seen": 674707776,
      "step": 21170,
      "train_runtime": 55499.9146,
      "train_tokens_per_second": 12156.916
    },
    {
      "epoch": 1.425733907891193,
      "grad_norm": 0.8099773802174989,
      "learning_rate": 1.9004205949093695e-06,
      "loss": 0.2939807176589966,
      "num_input_tokens_seen": 674865328,
      "step": 21175,
      "train_runtime": 55512.9709,
      "train_tokens_per_second": 12156.894
    },
    {
      "epoch": 1.4260705628871533,
      "grad_norm": 0.8075039027333336,
      "learning_rate": 1.8983463010183355e-06,
      "loss": 0.3183480978012085,
      "num_input_tokens_seen": 675022584,
      "step": 21180,
      "train_runtime": 55526.5637,
      "train_tokens_per_second": 12156.751
    },
    {
      "epoch": 1.4264072178831133,
      "grad_norm": 0.7630924665018629,
      "learning_rate": 1.8962728744948644e-06,
      "loss": 0.30946836471557615,
      "num_input_tokens_seen": 675186424,
      "step": 21185,
      "train_runtime": 55538.9902,
      "train_tokens_per_second": 12156.981
    },
    {
      "epoch": 1.4267438728790736,
      "grad_norm": 0.8392343554852615,
      "learning_rate": 1.894200315918789e-06,
      "loss": 0.30475497245788574,
      "num_input_tokens_seen": 675349392,
      "step": 21190,
      "train_runtime": 55551.7948,
      "train_tokens_per_second": 12157.112
    },
    {
      "epoch": 1.4270805278750336,
      "grad_norm": 0.8109563508046852,
      "learning_rate": 1.8921286258696913e-06,
      "loss": 0.2869288444519043,
      "num_input_tokens_seen": 675500192,
      "step": 21195,
      "train_runtime": 55564.6299,
      "train_tokens_per_second": 12157.018
    },
    {
      "epoch": 1.4274171828709938,
      "grad_norm": 0.8413780035102365,
      "learning_rate": 1.890057804926913e-06,
      "loss": 0.30377175807952883,
      "num_input_tokens_seen": 675661744,
      "step": 21200,
      "train_runtime": 55577.1366,
      "train_tokens_per_second": 12157.189
    },
    {
      "epoch": 1.4277538378669539,
      "grad_norm": 0.8009082947160341,
      "learning_rate": 1.887987853669553e-06,
      "loss": 0.2659959077835083,
      "num_input_tokens_seen": 675823128,
      "step": 21205,
      "train_runtime": 55590.8956,
      "train_tokens_per_second": 12157.083
    },
    {
      "epoch": 1.4280904928629141,
      "grad_norm": 0.9297945141354858,
      "learning_rate": 1.8859187726764654e-06,
      "loss": 0.30696640014648435,
      "num_input_tokens_seen": 675983296,
      "step": 21210,
      "train_runtime": 55603.5415,
      "train_tokens_per_second": 12157.199
    },
    {
      "epoch": 1.4284271478588741,
      "grad_norm": 0.8342897761211544,
      "learning_rate": 1.8838505625262638e-06,
      "loss": 0.29467411041259767,
      "num_input_tokens_seen": 676144424,
      "step": 21215,
      "train_runtime": 55616.0197,
      "train_tokens_per_second": 12157.368
    },
    {
      "epoch": 1.4287638028548344,
      "grad_norm": 0.802338836125089,
      "learning_rate": 1.8817832237973154e-06,
      "loss": 0.2830294132232666,
      "num_input_tokens_seen": 676303272,
      "step": 21220,
      "train_runtime": 55628.4322,
      "train_tokens_per_second": 12157.511
    },
    {
      "epoch": 1.4291004578507944,
      "grad_norm": 0.8111032479295406,
      "learning_rate": 1.8797167570677439e-06,
      "loss": 0.27536394596099856,
      "num_input_tokens_seen": 676462536,
      "step": 21225,
      "train_runtime": 55640.8332,
      "train_tokens_per_second": 12157.664
    },
    {
      "epoch": 1.4294371128467547,
      "grad_norm": 0.8013847844029551,
      "learning_rate": 1.8776511629154342e-06,
      "loss": 0.3162542819976807,
      "num_input_tokens_seen": 676625088,
      "step": 21230,
      "train_runtime": 55653.5027,
      "train_tokens_per_second": 12157.817
    },
    {
      "epoch": 1.4297737678427147,
      "grad_norm": 0.8018879789604034,
      "learning_rate": 1.8755864419180176e-06,
      "loss": 0.2986203670501709,
      "num_input_tokens_seen": 676786488,
      "step": 21235,
      "train_runtime": 55666.6263,
      "train_tokens_per_second": 12157.85
    },
    {
      "epoch": 1.430110422838675,
      "grad_norm": 0.7782613857007677,
      "learning_rate": 1.8735225946528906e-06,
      "loss": 0.30481600761413574,
      "num_input_tokens_seen": 676948128,
      "step": 21240,
      "train_runtime": 55679.1479,
      "train_tokens_per_second": 12158.019
    },
    {
      "epoch": 1.430447077834635,
      "grad_norm": 0.8938079243742394,
      "learning_rate": 1.8714596216972008e-06,
      "loss": 0.3058157444000244,
      "num_input_tokens_seen": 677107872,
      "step": 21245,
      "train_runtime": 55692.2215,
      "train_tokens_per_second": 12158.033
    },
    {
      "epoch": 1.4307837328305952,
      "grad_norm": 0.7858481517675684,
      "learning_rate": 1.8693975236278512e-06,
      "loss": 0.2930753231048584,
      "num_input_tokens_seen": 677267064,
      "step": 21250,
      "train_runtime": 55705.0235,
      "train_tokens_per_second": 12158.097
    },
    {
      "epoch": 1.4311203878265553,
      "grad_norm": 0.8704437172846795,
      "learning_rate": 1.8673363010215017e-06,
      "loss": 0.3337438106536865,
      "num_input_tokens_seen": 677426112,
      "step": 21255,
      "train_runtime": 55718.8506,
      "train_tokens_per_second": 12157.934
    },
    {
      "epoch": 1.4314570428225155,
      "grad_norm": 0.81490648902403,
      "learning_rate": 1.8652759544545661e-06,
      "loss": 0.3077549457550049,
      "num_input_tokens_seen": 677589952,
      "step": 21260,
      "train_runtime": 55731.2705,
      "train_tokens_per_second": 12158.164
    },
    {
      "epoch": 1.4317936978184758,
      "grad_norm": 0.7867439030425964,
      "learning_rate": 1.863216484503214e-06,
      "loss": 0.296589994430542,
      "num_input_tokens_seen": 677747656,
      "step": 21265,
      "train_runtime": 55744.4655,
      "train_tokens_per_second": 12158.116
    },
    {
      "epoch": 1.4321303528144358,
      "grad_norm": 0.8600250138820671,
      "learning_rate": 1.8611578917433698e-06,
      "loss": 0.32840142250061033,
      "num_input_tokens_seen": 677909472,
      "step": 21270,
      "train_runtime": 55757.5593,
      "train_tokens_per_second": 12158.163
    },
    {
      "epoch": 1.4324670078103958,
      "grad_norm": 0.8382862473213238,
      "learning_rate": 1.85910017675071e-06,
      "loss": 0.3198842525482178,
      "num_input_tokens_seen": 678066912,
      "step": 21275,
      "train_runtime": 55770.8971,
      "train_tokens_per_second": 12158.078
    },
    {
      "epoch": 1.432803662806356,
      "grad_norm": 0.7281648267814856,
      "learning_rate": 1.8570433401006737e-06,
      "loss": 0.26290369033813477,
      "num_input_tokens_seen": 678229528,
      "step": 21280,
      "train_runtime": 55783.7702,
      "train_tokens_per_second": 12158.187
    },
    {
      "epoch": 1.4331403178023163,
      "grad_norm": 1.136206886167445,
      "learning_rate": 1.8549873823684418e-06,
      "loss": 0.27656700611114504,
      "num_input_tokens_seen": 678389488,
      "step": 21285,
      "train_runtime": 55796.8752,
      "train_tokens_per_second": 12158.199
    },
    {
      "epoch": 1.4334769727982763,
      "grad_norm": 0.7797648999275555,
      "learning_rate": 1.8529323041289616e-06,
      "loss": 0.3268596649169922,
      "num_input_tokens_seen": 678550680,
      "step": 21290,
      "train_runtime": 55810.0123,
      "train_tokens_per_second": 12158.225
    },
    {
      "epoch": 1.4338136277942364,
      "grad_norm": 0.7639621711925553,
      "learning_rate": 1.850878105956927e-06,
      "loss": 0.2793318271636963,
      "num_input_tokens_seen": 678710392,
      "step": 21295,
      "train_runtime": 55823.0178,
      "train_tokens_per_second": 12158.253
    },
    {
      "epoch": 1.4341502827901966,
      "grad_norm": 0.7463992125697302,
      "learning_rate": 1.8488247884267885e-06,
      "loss": 0.28080949783325193,
      "num_input_tokens_seen": 678873704,
      "step": 21300,
      "train_runtime": 55835.9203,
      "train_tokens_per_second": 12158.369
    },
    {
      "epoch": 1.4344869377861569,
      "grad_norm": 0.8047760558116408,
      "learning_rate": 1.8467723521127495e-06,
      "loss": 0.2996971130371094,
      "num_input_tokens_seen": 679033472,
      "step": 21305,
      "train_runtime": 55849.1126,
      "train_tokens_per_second": 12158.357
    },
    {
      "epoch": 1.434823592782117,
      "grad_norm": 0.8640964976594291,
      "learning_rate": 1.8447207975887671e-06,
      "loss": 0.29690799713134763,
      "num_input_tokens_seen": 679192064,
      "step": 21310,
      "train_runtime": 55862.1539,
      "train_tokens_per_second": 12158.358
    },
    {
      "epoch": 1.435160247778077,
      "grad_norm": 0.8599610073097582,
      "learning_rate": 1.8426701254285522e-06,
      "loss": 0.30842161178588867,
      "num_input_tokens_seen": 679352472,
      "step": 21315,
      "train_runtime": 55874.5807,
      "train_tokens_per_second": 12158.525
    },
    {
      "epoch": 1.4354969027740372,
      "grad_norm": 0.8479316932676714,
      "learning_rate": 1.840620336205569e-06,
      "loss": 0.32105112075805664,
      "num_input_tokens_seen": 679512592,
      "step": 21320,
      "train_runtime": 55887.9789,
      "train_tokens_per_second": 12158.475
    },
    {
      "epoch": 1.4358335577699974,
      "grad_norm": 0.7671230182886822,
      "learning_rate": 1.8385714304930324e-06,
      "loss": 0.2665998458862305,
      "num_input_tokens_seen": 679672936,
      "step": 21325,
      "train_runtime": 55900.384,
      "train_tokens_per_second": 12158.645
    },
    {
      "epoch": 1.4361702127659575,
      "grad_norm": 0.7776175599887056,
      "learning_rate": 1.8365234088639178e-06,
      "loss": 0.2932872295379639,
      "num_input_tokens_seen": 679833552,
      "step": 21330,
      "train_runtime": 55913.7732,
      "train_tokens_per_second": 12158.606
    },
    {
      "epoch": 1.4365068677619175,
      "grad_norm": 0.8517853062803455,
      "learning_rate": 1.834476271890941e-06,
      "loss": 0.2763814926147461,
      "num_input_tokens_seen": 679987920,
      "step": 21335,
      "train_runtime": 55926.6054,
      "train_tokens_per_second": 12158.577
    },
    {
      "epoch": 1.4368435227578777,
      "grad_norm": 0.9210536635646197,
      "learning_rate": 1.8324300201465823e-06,
      "loss": 0.2705744743347168,
      "num_input_tokens_seen": 680148872,
      "step": 21340,
      "train_runtime": 55939.2363,
      "train_tokens_per_second": 12158.709
    },
    {
      "epoch": 1.437180177753838,
      "grad_norm": 0.8839284294447675,
      "learning_rate": 1.8303846542030674e-06,
      "loss": 0.3186767816543579,
      "num_input_tokens_seen": 680303360,
      "step": 21345,
      "train_runtime": 55951.9717,
      "train_tokens_per_second": 12158.702
    },
    {
      "epoch": 1.437516832749798,
      "grad_norm": 0.7068151177368024,
      "learning_rate": 1.828340174632377e-06,
      "loss": 0.2777002573013306,
      "num_input_tokens_seen": 680465664,
      "step": 21350,
      "train_runtime": 55965.9224,
      "train_tokens_per_second": 12158.571
    },
    {
      "epoch": 1.437853487745758,
      "grad_norm": 0.8534727297405322,
      "learning_rate": 1.8262965820062434e-06,
      "loss": 0.2907694339752197,
      "num_input_tokens_seen": 680624496,
      "step": 21355,
      "train_runtime": 55979.1814,
      "train_tokens_per_second": 12158.529
    },
    {
      "epoch": 1.4381901427417183,
      "grad_norm": 0.7535014122583716,
      "learning_rate": 1.8242538768961499e-06,
      "loss": 0.28483853340148924,
      "num_input_tokens_seen": 680781576,
      "step": 21360,
      "train_runtime": 55992.4754,
      "train_tokens_per_second": 12158.448
    },
    {
      "epoch": 1.4385267977376786,
      "grad_norm": 0.8690268193552331,
      "learning_rate": 1.8222120598733317e-06,
      "loss": 0.28143222332000734,
      "num_input_tokens_seen": 680945416,
      "step": 21365,
      "train_runtime": 56004.8926,
      "train_tokens_per_second": 12158.677
    },
    {
      "epoch": 1.4388634527336386,
      "grad_norm": 0.7737322165539616,
      "learning_rate": 1.820171131508781e-06,
      "loss": 0.32750644683837893,
      "num_input_tokens_seen": 681101704,
      "step": 21370,
      "train_runtime": 56018.4947,
      "train_tokens_per_second": 12158.515
    },
    {
      "epoch": 1.4392001077295986,
      "grad_norm": 0.8089917567865031,
      "learning_rate": 1.8181310923732304e-06,
      "loss": 0.26973321437835696,
      "num_input_tokens_seen": 681257256,
      "step": 21375,
      "train_runtime": 56031.1052,
      "train_tokens_per_second": 12158.555
    },
    {
      "epoch": 1.4395367627255589,
      "grad_norm": 0.7570295101572517,
      "learning_rate": 1.8160919430371764e-06,
      "loss": 0.31278605461120607,
      "num_input_tokens_seen": 681414712,
      "step": 21380,
      "train_runtime": 56044.3987,
      "train_tokens_per_second": 12158.48
    },
    {
      "epoch": 1.439873417721519,
      "grad_norm": 0.7913222820346503,
      "learning_rate": 1.814053684070855e-06,
      "loss": 0.28987956047058105,
      "num_input_tokens_seen": 681575456,
      "step": 21385,
      "train_runtime": 56057.2879,
      "train_tokens_per_second": 12158.552
    },
    {
      "epoch": 1.4402100727174791,
      "grad_norm": 0.7461149458225613,
      "learning_rate": 1.8120163160442633e-06,
      "loss": 0.2691861867904663,
      "num_input_tokens_seen": 681736600,
      "step": 21390,
      "train_runtime": 56069.7804,
      "train_tokens_per_second": 12158.717
    },
    {
      "epoch": 1.4405467277134392,
      "grad_norm": 0.8518112031645841,
      "learning_rate": 1.8099798395271428e-06,
      "loss": 0.3051251173019409,
      "num_input_tokens_seen": 681899024,
      "step": 21395,
      "train_runtime": 56082.4088,
      "train_tokens_per_second": 12158.875
    },
    {
      "epoch": 1.4408833827093994,
      "grad_norm": 0.9029857919230043,
      "learning_rate": 1.807944255088988e-06,
      "loss": 0.26226377487182617,
      "num_input_tokens_seen": 682057120,
      "step": 21400,
      "train_runtime": 56095.8838,
      "train_tokens_per_second": 12158.773
    },
    {
      "epoch": 1.4412200377053597,
      "grad_norm": 0.8823333700329987,
      "learning_rate": 1.8059095632990437e-06,
      "loss": 0.2917477607727051,
      "num_input_tokens_seen": 682218112,
      "step": 21405,
      "train_runtime": 56109.6401,
      "train_tokens_per_second": 12158.661
    },
    {
      "epoch": 1.4415566927013197,
      "grad_norm": 0.9166061747778074,
      "learning_rate": 1.8038757647263045e-06,
      "loss": 0.30876636505126953,
      "num_input_tokens_seen": 682376680,
      "step": 21410,
      "train_runtime": 56122.126,
      "train_tokens_per_second": 12158.782
    },
    {
      "epoch": 1.4418933476972797,
      "grad_norm": 0.7395607613262293,
      "learning_rate": 1.8018428599395143e-06,
      "loss": 0.2834480762481689,
      "num_input_tokens_seen": 682535760,
      "step": 21415,
      "train_runtime": 56134.5499,
      "train_tokens_per_second": 12158.925
    },
    {
      "epoch": 1.44223000269324,
      "grad_norm": 0.7507365515737661,
      "learning_rate": 1.7998108495071743e-06,
      "loss": 0.2868551254272461,
      "num_input_tokens_seen": 682698272,
      "step": 21420,
      "train_runtime": 56147.2099,
      "train_tokens_per_second": 12159.077
    },
    {
      "epoch": 1.4425666576892002,
      "grad_norm": 0.7195802445551491,
      "learning_rate": 1.797779733997521e-06,
      "loss": 0.29749279022216796,
      "num_input_tokens_seen": 682859408,
      "step": 21425,
      "train_runtime": 56160.7357,
      "train_tokens_per_second": 12159.018
    },
    {
      "epoch": 1.4429033126851603,
      "grad_norm": 0.8038981856926783,
      "learning_rate": 1.7957495139785585e-06,
      "loss": 0.31753060817718504,
      "num_input_tokens_seen": 683022312,
      "step": 21430,
      "train_runtime": 56174.2291,
      "train_tokens_per_second": 12158.998
    },
    {
      "epoch": 1.4432399676811203,
      "grad_norm": 0.8101308316696099,
      "learning_rate": 1.7937201900180228e-06,
      "loss": 0.2885474681854248,
      "num_input_tokens_seen": 683181360,
      "step": 21435,
      "train_runtime": 56188.4501,
      "train_tokens_per_second": 12158.751
    },
    {
      "epoch": 1.4435766226770805,
      "grad_norm": 0.8721945293025303,
      "learning_rate": 1.7916917626834134e-06,
      "loss": 0.31629080772399903,
      "num_input_tokens_seen": 683343096,
      "step": 21440,
      "train_runtime": 56200.8975,
      "train_tokens_per_second": 12158.936
    },
    {
      "epoch": 1.4439132776730408,
      "grad_norm": 0.804847599439285,
      "learning_rate": 1.7896642325419722e-06,
      "loss": 0.3054671764373779,
      "num_input_tokens_seen": 683503256,
      "step": 21445,
      "train_runtime": 56213.6228,
      "train_tokens_per_second": 12159.032
    },
    {
      "epoch": 1.4442499326690008,
      "grad_norm": 0.8571638144056907,
      "learning_rate": 1.7876376001606915e-06,
      "loss": 0.2939744472503662,
      "num_input_tokens_seen": 683658648,
      "step": 21450,
      "train_runtime": 56227.3792,
      "train_tokens_per_second": 12158.821
    },
    {
      "epoch": 1.4445865876649608,
      "grad_norm": 0.7304948314736766,
      "learning_rate": 1.7856118661063128e-06,
      "loss": 0.2670135021209717,
      "num_input_tokens_seen": 683822488,
      "step": 21455,
      "train_runtime": 56239.8054,
      "train_tokens_per_second": 12159.048
    },
    {
      "epoch": 1.444923242660921,
      "grad_norm": 0.7626352295598008,
      "learning_rate": 1.7835870309453251e-06,
      "loss": 0.2866842746734619,
      "num_input_tokens_seen": 683982128,
      "step": 21460,
      "train_runtime": 56253.5652,
      "train_tokens_per_second": 12158.912
    },
    {
      "epoch": 1.4452598976568813,
      "grad_norm": 0.8566538435293638,
      "learning_rate": 1.7815630952439667e-06,
      "loss": 0.29722516536712645,
      "num_input_tokens_seen": 684142992,
      "step": 21465,
      "train_runtime": 56267.2544,
      "train_tokens_per_second": 12158.812
    },
    {
      "epoch": 1.4455965526528414,
      "grad_norm": 0.7140566201619843,
      "learning_rate": 1.7795400595682289e-06,
      "loss": 0.3011247396469116,
      "num_input_tokens_seen": 684302056,
      "step": 21470,
      "train_runtime": 56280.7372,
      "train_tokens_per_second": 12158.726
    },
    {
      "epoch": 1.4459332076488014,
      "grad_norm": 0.8596531447643275,
      "learning_rate": 1.7775179244838408e-06,
      "loss": 0.32985286712646483,
      "num_input_tokens_seen": 684454880,
      "step": 21475,
      "train_runtime": 56293.7311,
      "train_tokens_per_second": 12158.634
    },
    {
      "epoch": 1.4462698626447617,
      "grad_norm": 0.7616688972474056,
      "learning_rate": 1.7754966905562925e-06,
      "loss": 0.27933244705200194,
      "num_input_tokens_seen": 684611888,
      "step": 21480,
      "train_runtime": 56306.9349,
      "train_tokens_per_second": 12158.571
    },
    {
      "epoch": 1.446606517640722,
      "grad_norm": 1.0618426951338584,
      "learning_rate": 1.7734763583508098e-06,
      "loss": 0.28763225078582766,
      "num_input_tokens_seen": 684772608,
      "step": 21485,
      "train_runtime": 56320.9338,
      "train_tokens_per_second": 12158.403
    },
    {
      "epoch": 1.446943172636682,
      "grad_norm": 0.8140189460786941,
      "learning_rate": 1.7714569284323757e-06,
      "loss": 0.2967977046966553,
      "num_input_tokens_seen": 684935168,
      "step": 21490,
      "train_runtime": 56333.3347,
      "train_tokens_per_second": 12158.612
    },
    {
      "epoch": 1.447279827632642,
      "grad_norm": 0.8430648848131194,
      "learning_rate": 1.7694384013657173e-06,
      "loss": 0.2972251892089844,
      "num_input_tokens_seen": 685093464,
      "step": 21495,
      "train_runtime": 56346.178,
      "train_tokens_per_second": 12158.65
    },
    {
      "epoch": 1.4476164826286022,
      "grad_norm": 0.796870231179487,
      "learning_rate": 1.767420777715308e-06,
      "loss": 0.31137757301330565,
      "num_input_tokens_seen": 685257304,
      "step": 21500,
      "train_runtime": 56358.5915,
      "train_tokens_per_second": 12158.879
    },
    {
      "epoch": 1.4479531376245625,
      "grad_norm": 0.8133643909610662,
      "learning_rate": 1.7654040580453702e-06,
      "loss": 0.28451905250549314,
      "num_input_tokens_seen": 685417648,
      "step": 21505,
      "train_runtime": 56371.3117,
      "train_tokens_per_second": 12158.98
    },
    {
      "epoch": 1.4482897926205225,
      "grad_norm": 0.7210214934455141,
      "learning_rate": 1.7633882429198734e-06,
      "loss": 0.3130315065383911,
      "num_input_tokens_seen": 685578560,
      "step": 21510,
      "train_runtime": 56383.7591,
      "train_tokens_per_second": 12159.15
    },
    {
      "epoch": 1.4486264476164825,
      "grad_norm": 0.7551439242442248,
      "learning_rate": 1.7613733329025318e-06,
      "loss": 0.29886622428894044,
      "num_input_tokens_seen": 685742400,
      "step": 21515,
      "train_runtime": 56396.182,
      "train_tokens_per_second": 12159.376
    },
    {
      "epoch": 1.4489631026124428,
      "grad_norm": 0.7434773451987416,
      "learning_rate": 1.759359328556814e-06,
      "loss": 0.31097979545593263,
      "num_input_tokens_seen": 685903624,
      "step": 21520,
      "train_runtime": 56408.6727,
      "train_tokens_per_second": 12159.542
    },
    {
      "epoch": 1.449299757608403,
      "grad_norm": 0.7970450861926961,
      "learning_rate": 1.7573462304459227e-06,
      "loss": 0.3127882957458496,
      "num_input_tokens_seen": 686063696,
      "step": 21525,
      "train_runtime": 56421.6866,
      "train_tokens_per_second": 12159.574
    },
    {
      "epoch": 1.449636412604363,
      "grad_norm": 0.8405521775564402,
      "learning_rate": 1.7553340391328216e-06,
      "loss": 0.30340938568115233,
      "num_input_tokens_seen": 686222928,
      "step": 21530,
      "train_runtime": 56434.7584,
      "train_tokens_per_second": 12159.579
    },
    {
      "epoch": 1.449973067600323,
      "grad_norm": 0.8318611910972775,
      "learning_rate": 1.7533227551802068e-06,
      "loss": 0.3056394100189209,
      "num_input_tokens_seen": 686384208,
      "step": 21535,
      "train_runtime": 56447.2105,
      "train_tokens_per_second": 12159.754
    },
    {
      "epoch": 1.4503097225962833,
      "grad_norm": 0.7891807978480752,
      "learning_rate": 1.7513123791505321e-06,
      "loss": 0.3165531873703003,
      "num_input_tokens_seen": 686542424,
      "step": 21540,
      "train_runtime": 56460.5763,
      "train_tokens_per_second": 12159.678
    },
    {
      "epoch": 1.4506463775922436,
      "grad_norm": 0.811067578094932,
      "learning_rate": 1.7493029116059913e-06,
      "loss": 0.3050800085067749,
      "num_input_tokens_seen": 686700600,
      "step": 21545,
      "train_runtime": 56473.8854,
      "train_tokens_per_second": 12159.613
    },
    {
      "epoch": 1.4509830325882036,
      "grad_norm": 0.7653939457423206,
      "learning_rate": 1.7472943531085252e-06,
      "loss": 0.30484685897827146,
      "num_input_tokens_seen": 686863184,
      "step": 21550,
      "train_runtime": 56486.5602,
      "train_tokens_per_second": 12159.763
    },
    {
      "epoch": 1.4513196875841636,
      "grad_norm": 0.8491299705942827,
      "learning_rate": 1.7452867042198212e-06,
      "loss": 0.28639864921569824,
      "num_input_tokens_seen": 687021208,
      "step": 21555,
      "train_runtime": 56499.1375,
      "train_tokens_per_second": 12159.853
    },
    {
      "epoch": 1.4516563425801239,
      "grad_norm": 0.7884690898621628,
      "learning_rate": 1.7432799655013122e-06,
      "loss": 0.28300116062164304,
      "num_input_tokens_seen": 687183360,
      "step": 21560,
      "train_runtime": 56512.3409,
      "train_tokens_per_second": 12159.881
    },
    {
      "epoch": 1.4519929975760841,
      "grad_norm": 0.7838449284348402,
      "learning_rate": 1.741274137514174e-06,
      "loss": 0.30590014457702636,
      "num_input_tokens_seen": 687342176,
      "step": 21565,
      "train_runtime": 56525.671,
      "train_tokens_per_second": 12159.823
    },
    {
      "epoch": 1.4523296525720442,
      "grad_norm": 0.8129609340304309,
      "learning_rate": 1.7392692208193351e-06,
      "loss": 0.28932440280914307,
      "num_input_tokens_seen": 687500120,
      "step": 21570,
      "train_runtime": 56538.6912,
      "train_tokens_per_second": 12159.817
    },
    {
      "epoch": 1.4526663075680042,
      "grad_norm": 0.7990204851467065,
      "learning_rate": 1.7372652159774578e-06,
      "loss": 0.26776914596557616,
      "num_input_tokens_seen": 687661640,
      "step": 21575,
      "train_runtime": 56551.1845,
      "train_tokens_per_second": 12159.987
    },
    {
      "epoch": 1.4530029625639644,
      "grad_norm": 0.8100846372904865,
      "learning_rate": 1.7352621235489625e-06,
      "loss": 0.28267197608947753,
      "num_input_tokens_seen": 687822464,
      "step": 21580,
      "train_runtime": 56564.7069,
      "train_tokens_per_second": 12159.923
    },
    {
      "epoch": 1.4533396175599247,
      "grad_norm": 0.9413468826323067,
      "learning_rate": 1.733259944094e-06,
      "loss": 0.3131802320480347,
      "num_input_tokens_seen": 687979464,
      "step": 21585,
      "train_runtime": 56577.4555,
      "train_tokens_per_second": 12159.958
    },
    {
      "epoch": 1.4536762725558847,
      "grad_norm": 0.7667098667048667,
      "learning_rate": 1.7312586781724789e-06,
      "loss": 0.29518172740936277,
      "num_input_tokens_seen": 688139224,
      "step": 21590,
      "train_runtime": 56590.6632,
      "train_tokens_per_second": 12159.943
    },
    {
      "epoch": 1.4540129275518447,
      "grad_norm": 0.8543632845666106,
      "learning_rate": 1.7292583263440454e-06,
      "loss": 0.31833181381225584,
      "num_input_tokens_seen": 688295520,
      "step": 21595,
      "train_runtime": 56603.443,
      "train_tokens_per_second": 12159.959
    },
    {
      "epoch": 1.454349582547805,
      "grad_norm": 0.826204547232496,
      "learning_rate": 1.7272588891680914e-06,
      "loss": 0.2934000015258789,
      "num_input_tokens_seen": 688453280,
      "step": 21600,
      "train_runtime": 56616.7332,
      "train_tokens_per_second": 12159.891
    },
    {
      "epoch": 1.4546862375437652,
      "grad_norm": 0.9121824989175548,
      "learning_rate": 1.7252603672037527e-06,
      "loss": 0.28450870513916016,
      "num_input_tokens_seen": 688609416,
      "step": 21605,
      "train_runtime": 56629.8267,
      "train_tokens_per_second": 12159.836
    },
    {
      "epoch": 1.4550228925397253,
      "grad_norm": 0.8181807965862449,
      "learning_rate": 1.72326276100991e-06,
      "loss": 0.2809373140335083,
      "num_input_tokens_seen": 688772344,
      "step": 21610,
      "train_runtime": 56642.6144,
      "train_tokens_per_second": 12159.967
    },
    {
      "epoch": 1.4553595475356853,
      "grad_norm": 0.8029632686030225,
      "learning_rate": 1.721266071145186e-06,
      "loss": 0.3233592987060547,
      "num_input_tokens_seen": 688932688,
      "step": 21615,
      "train_runtime": 56655.2153,
      "train_tokens_per_second": 12160.093
    },
    {
      "epoch": 1.4556962025316456,
      "grad_norm": 0.7452444064061056,
      "learning_rate": 1.7192702981679531e-06,
      "loss": 0.30849566459655764,
      "num_input_tokens_seen": 689094272,
      "step": 21620,
      "train_runtime": 56667.9552,
      "train_tokens_per_second": 12160.211
    },
    {
      "epoch": 1.4560328575276058,
      "grad_norm": 0.7858216629109098,
      "learning_rate": 1.717275442636317e-06,
      "loss": 0.2630155563354492,
      "num_input_tokens_seen": 689256704,
      "step": 21625,
      "train_runtime": 56680.526,
      "train_tokens_per_second": 12160.379
    },
    {
      "epoch": 1.4563695125235658,
      "grad_norm": 1.0564395769040469,
      "learning_rate": 1.7152815051081384e-06,
      "loss": 0.31965184211730957,
      "num_input_tokens_seen": 689415328,
      "step": 21630,
      "train_runtime": 56693.8752,
      "train_tokens_per_second": 12160.314
    },
    {
      "epoch": 1.4567061675195259,
      "grad_norm": 0.73330799555457,
      "learning_rate": 1.7132884861410093e-06,
      "loss": 0.2977971315383911,
      "num_input_tokens_seen": 689572824,
      "step": 21635,
      "train_runtime": 56706.9176,
      "train_tokens_per_second": 12160.295
    },
    {
      "epoch": 1.4570428225154861,
      "grad_norm": 0.8478087734585148,
      "learning_rate": 1.7112963862922766e-06,
      "loss": 0.2808065891265869,
      "num_input_tokens_seen": 689729696,
      "step": 21640,
      "train_runtime": 56719.6182,
      "train_tokens_per_second": 12160.337
    },
    {
      "epoch": 1.4573794775114464,
      "grad_norm": 0.7912794573815,
      "learning_rate": 1.7093052061190224e-06,
      "loss": 0.30621819496154784,
      "num_input_tokens_seen": 689893360,
      "step": 21645,
      "train_runtime": 56732.6714,
      "train_tokens_per_second": 12160.424
    },
    {
      "epoch": 1.4577161325074064,
      "grad_norm": 0.8362175201732533,
      "learning_rate": 1.7073149461780742e-06,
      "loss": 0.2925211191177368,
      "num_input_tokens_seen": 690055544,
      "step": 21650,
      "train_runtime": 56745.2835,
      "train_tokens_per_second": 12160.58
    },
    {
      "epoch": 1.4580527875033664,
      "grad_norm": 0.9076278128604351,
      "learning_rate": 1.705325607026001e-06,
      "loss": 0.304974627494812,
      "num_input_tokens_seen": 690210416,
      "step": 21655,
      "train_runtime": 56757.8022,
      "train_tokens_per_second": 12160.626
    },
    {
      "epoch": 1.4583894424993267,
      "grad_norm": 0.8754132700959792,
      "learning_rate": 1.7033371892191164e-06,
      "loss": 0.30251388549804686,
      "num_input_tokens_seen": 690370128,
      "step": 21660,
      "train_runtime": 56770.5412,
      "train_tokens_per_second": 12160.711
    },
    {
      "epoch": 1.458726097495287,
      "grad_norm": 0.8607860752993293,
      "learning_rate": 1.7013496933134726e-06,
      "loss": 0.305312967300415,
      "num_input_tokens_seen": 690526664,
      "step": 21665,
      "train_runtime": 56783.3667,
      "train_tokens_per_second": 12160.721
    },
    {
      "epoch": 1.459062752491247,
      "grad_norm": 0.7833076191979462,
      "learning_rate": 1.6993631198648724e-06,
      "loss": 0.33922324180603025,
      "num_input_tokens_seen": 690686888,
      "step": 21670,
      "train_runtime": 56797.0151,
      "train_tokens_per_second": 12160.619
    },
    {
      "epoch": 1.459399407487207,
      "grad_norm": 0.8876253004366031,
      "learning_rate": 1.697377469428847e-06,
      "loss": 0.2973735809326172,
      "num_input_tokens_seen": 690849520,
      "step": 21675,
      "train_runtime": 56810.5513,
      "train_tokens_per_second": 12160.585
    },
    {
      "epoch": 1.4597360624831672,
      "grad_norm": 0.8473232698736883,
      "learning_rate": 1.6953927425606842e-06,
      "loss": 0.32192392349243165,
      "num_input_tokens_seen": 691009656,
      "step": 21680,
      "train_runtime": 56823.808,
      "train_tokens_per_second": 12160.566
    },
    {
      "epoch": 1.4600727174791275,
      "grad_norm": 0.7003133936070418,
      "learning_rate": 1.6934089398154002e-06,
      "loss": 0.27196240425109863,
      "num_input_tokens_seen": 691168592,
      "step": 21685,
      "train_runtime": 56837.0784,
      "train_tokens_per_second": 12160.523
    },
    {
      "epoch": 1.4604093724750875,
      "grad_norm": 0.7047252653125704,
      "learning_rate": 1.6914260617477634e-06,
      "loss": 0.2757792234420776,
      "num_input_tokens_seen": 691327792,
      "step": 21690,
      "train_runtime": 56849.9583,
      "train_tokens_per_second": 12160.568
    },
    {
      "epoch": 1.4607460274710475,
      "grad_norm": 0.7045483722420454,
      "learning_rate": 1.6894441089122788e-06,
      "loss": 0.2627258777618408,
      "num_input_tokens_seen": 691488248,
      "step": 21695,
      "train_runtime": 56862.9279,
      "train_tokens_per_second": 12160.616
    },
    {
      "epoch": 1.4610826824670078,
      "grad_norm": 0.9390560886824095,
      "learning_rate": 1.6874630818631926e-06,
      "loss": 0.3093214511871338,
      "num_input_tokens_seen": 691649584,
      "step": 21700,
      "train_runtime": 56875.3652,
      "train_tokens_per_second": 12160.794
    },
    {
      "epoch": 1.461419337462968,
      "grad_norm": 0.7600123658609799,
      "learning_rate": 1.685482981154492e-06,
      "loss": 0.30641195774078367,
      "num_input_tokens_seen": 691810648,
      "step": 21705,
      "train_runtime": 56889.3407,
      "train_tokens_per_second": 12160.637
    },
    {
      "epoch": 1.461755992458928,
      "grad_norm": 0.7703978041383395,
      "learning_rate": 1.683503807339907e-06,
      "loss": 0.2964465141296387,
      "num_input_tokens_seen": 691971304,
      "step": 21710,
      "train_runtime": 56901.7958,
      "train_tokens_per_second": 12160.799
    },
    {
      "epoch": 1.462092647454888,
      "grad_norm": 0.8071186234033529,
      "learning_rate": 1.6815255609729047e-06,
      "loss": 0.3109596729278564,
      "num_input_tokens_seen": 692130800,
      "step": 21715,
      "train_runtime": 56914.454,
      "train_tokens_per_second": 12160.897
    },
    {
      "epoch": 1.4624293024508483,
      "grad_norm": 0.9074560559812094,
      "learning_rate": 1.6795482426067012e-06,
      "loss": 0.3002187967300415,
      "num_input_tokens_seen": 692284200,
      "step": 21720,
      "train_runtime": 56927.332,
      "train_tokens_per_second": 12160.84
    },
    {
      "epoch": 1.4627659574468086,
      "grad_norm": 0.9118115520023707,
      "learning_rate": 1.6775718527942403e-06,
      "loss": 0.2810234785079956,
      "num_input_tokens_seen": 692434768,
      "step": 21725,
      "train_runtime": 56941.6154,
      "train_tokens_per_second": 12160.434
    },
    {
      "epoch": 1.4631026124427686,
      "grad_norm": 1.338477873739346,
      "learning_rate": 1.675596392088219e-06,
      "loss": 0.2986829996109009,
      "num_input_tokens_seen": 692593432,
      "step": 21730,
      "train_runtime": 56954.1775,
      "train_tokens_per_second": 12160.538
    },
    {
      "epoch": 1.4634392674387287,
      "grad_norm": 0.7924862839798946,
      "learning_rate": 1.6736218610410633e-06,
      "loss": 0.2923915386199951,
      "num_input_tokens_seen": 692757272,
      "step": 21735,
      "train_runtime": 56966.6095,
      "train_tokens_per_second": 12160.76
    },
    {
      "epoch": 1.463775922434689,
      "grad_norm": 0.7805534920410327,
      "learning_rate": 1.6716482602049483e-06,
      "loss": 0.2890747547149658,
      "num_input_tokens_seen": 692913440,
      "step": 21740,
      "train_runtime": 56979.6811,
      "train_tokens_per_second": 12160.711
    },
    {
      "epoch": 1.4641125774306492,
      "grad_norm": 1.0264225372370543,
      "learning_rate": 1.669675590131784e-06,
      "loss": 0.3072577714920044,
      "num_input_tokens_seen": 693067128,
      "step": 21745,
      "train_runtime": 56993.1543,
      "train_tokens_per_second": 12160.533
    },
    {
      "epoch": 1.4644492324266092,
      "grad_norm": 0.8339566810095834,
      "learning_rate": 1.6677038513732214e-06,
      "loss": 0.3025707721710205,
      "num_input_tokens_seen": 693227848,
      "step": 21750,
      "train_runtime": 57006.3233,
      "train_tokens_per_second": 12160.543
    },
    {
      "epoch": 1.4647858874225694,
      "grad_norm": 0.8820807153206564,
      "learning_rate": 1.6657330444806508e-06,
      "loss": 0.3012739419937134,
      "num_input_tokens_seen": 693388168,
      "step": 21755,
      "train_runtime": 57020.5387,
      "train_tokens_per_second": 12160.323
    },
    {
      "epoch": 1.4651225424185295,
      "grad_norm": 0.8057176722233966,
      "learning_rate": 1.6637631700052021e-06,
      "loss": 0.29248919486999514,
      "num_input_tokens_seen": 693549880,
      "step": 21760,
      "train_runtime": 57033.6587,
      "train_tokens_per_second": 12160.361
    },
    {
      "epoch": 1.4654591974144897,
      "grad_norm": 0.788429603686,
      "learning_rate": 1.6617942284977428e-06,
      "loss": 0.27145137786865237,
      "num_input_tokens_seen": 693712880,
      "step": 21765,
      "train_runtime": 57046.4965,
      "train_tokens_per_second": 12160.482
    },
    {
      "epoch": 1.4657958524104497,
      "grad_norm": 0.8922539153296827,
      "learning_rate": 1.6598262205088862e-06,
      "loss": 0.3073990821838379,
      "num_input_tokens_seen": 693874712,
      "step": 21770,
      "train_runtime": 57059.7056,
      "train_tokens_per_second": 12160.503
    },
    {
      "epoch": 1.46613250740641,
      "grad_norm": 0.9074965378405806,
      "learning_rate": 1.6578591465889722e-06,
      "loss": 0.30682687759399413,
      "num_input_tokens_seen": 694036360,
      "step": 21775,
      "train_runtime": 57072.2493,
      "train_tokens_per_second": 12160.662
    },
    {
      "epoch": 1.46646916240237,
      "grad_norm": 0.802719267695498,
      "learning_rate": 1.6558930072880942e-06,
      "loss": 0.31342990398406984,
      "num_input_tokens_seen": 694197208,
      "step": 21780,
      "train_runtime": 57085.2144,
      "train_tokens_per_second": 12160.718
    },
    {
      "epoch": 1.4668058173983303,
      "grad_norm": 0.7944538935643626,
      "learning_rate": 1.6539278031560695e-06,
      "loss": 0.3204882383346558,
      "num_input_tokens_seen": 694354592,
      "step": 21785,
      "train_runtime": 57097.718,
      "train_tokens_per_second": 12160.812
    },
    {
      "epoch": 1.4671424723942903,
      "grad_norm": 0.946971906503643,
      "learning_rate": 1.6519635347424672e-06,
      "loss": 0.30272531509399414,
      "num_input_tokens_seen": 694509792,
      "step": 21790,
      "train_runtime": 57110.5879,
      "train_tokens_per_second": 12160.789
    },
    {
      "epoch": 1.4674791273902505,
      "grad_norm": 0.8256759423233608,
      "learning_rate": 1.650000202596586e-06,
      "loss": 0.3298978328704834,
      "num_input_tokens_seen": 694672520,
      "step": 21795,
      "train_runtime": 57123.0275,
      "train_tokens_per_second": 12160.989
    },
    {
      "epoch": 1.4678157823862106,
      "grad_norm": 0.9610444906539193,
      "learning_rate": 1.6480378072674659e-06,
      "loss": 0.304123067855835,
      "num_input_tokens_seen": 694834184,
      "step": 21800,
      "train_runtime": 57136.1132,
      "train_tokens_per_second": 12161.033
    },
    {
      "epoch": 1.4681524373821708,
      "grad_norm": 0.7554751854144708,
      "learning_rate": 1.646076349303884e-06,
      "loss": 0.287105131149292,
      "num_input_tokens_seen": 694995024,
      "step": 21805,
      "train_runtime": 57149.8773,
      "train_tokens_per_second": 12160.919
    },
    {
      "epoch": 1.4684890923781309,
      "grad_norm": 0.8640938887275199,
      "learning_rate": 1.6441158292543557e-06,
      "loss": 0.3035315990447998,
      "num_input_tokens_seen": 695155736,
      "step": 21810,
      "train_runtime": 57162.6254,
      "train_tokens_per_second": 12161.018
    },
    {
      "epoch": 1.468825747374091,
      "grad_norm": 0.7617390162940705,
      "learning_rate": 1.6421562476671339e-06,
      "loss": 0.33049280643463136,
      "num_input_tokens_seen": 695315528,
      "step": 21815,
      "train_runtime": 57175.1943,
      "train_tokens_per_second": 12161.14
    },
    {
      "epoch": 1.4691624023700511,
      "grad_norm": 0.7478745100343435,
      "learning_rate": 1.640197605090213e-06,
      "loss": 0.27458982467651366,
      "num_input_tokens_seen": 695476840,
      "step": 21820,
      "train_runtime": 57188.3996,
      "train_tokens_per_second": 12161.152
    },
    {
      "epoch": 1.4694990573660114,
      "grad_norm": 0.7683446394142356,
      "learning_rate": 1.6382399020713146e-06,
      "loss": 0.29114446640014646,
      "num_input_tokens_seen": 695629176,
      "step": 21825,
      "train_runtime": 57202.143,
      "train_tokens_per_second": 12160.894
    },
    {
      "epoch": 1.4698357123619714,
      "grad_norm": 0.8134183056000038,
      "learning_rate": 1.6362831391579115e-06,
      "loss": 0.27961149215698244,
      "num_input_tokens_seen": 695788352,
      "step": 21830,
      "train_runtime": 57214.8672,
      "train_tokens_per_second": 12160.971
    },
    {
      "epoch": 1.4701723673579317,
      "grad_norm": 0.8053622238046876,
      "learning_rate": 1.634327316897199e-06,
      "loss": 0.27769694328308103,
      "num_input_tokens_seen": 695947496,
      "step": 21835,
      "train_runtime": 57228.7127,
      "train_tokens_per_second": 12160.81
    },
    {
      "epoch": 1.4705090223538917,
      "grad_norm": 0.7177786556094375,
      "learning_rate": 1.6323724358361216e-06,
      "loss": 0.2857221603393555,
      "num_input_tokens_seen": 696103888,
      "step": 21840,
      "train_runtime": 57242.4539,
      "train_tokens_per_second": 12160.623
    },
    {
      "epoch": 1.470845677349852,
      "grad_norm": 0.8067103656560941,
      "learning_rate": 1.6304184965213537e-06,
      "loss": 0.28710260391235354,
      "num_input_tokens_seen": 696260920,
      "step": 21845,
      "train_runtime": 57255.2412,
      "train_tokens_per_second": 12160.649
    },
    {
      "epoch": 1.471182332345812,
      "grad_norm": 0.8138513564337244,
      "learning_rate": 1.6284654994993088e-06,
      "loss": 0.3102174043655396,
      "num_input_tokens_seen": 696424472,
      "step": 21850,
      "train_runtime": 57268.2816,
      "train_tokens_per_second": 12160.736
    },
    {
      "epoch": 1.4715189873417722,
      "grad_norm": 0.7643060201554904,
      "learning_rate": 1.6265134453161358e-06,
      "loss": 0.29467175006866453,
      "num_input_tokens_seen": 696583664,
      "step": 21855,
      "train_runtime": 57281.7107,
      "train_tokens_per_second": 12160.664
    },
    {
      "epoch": 1.4718556423377323,
      "grad_norm": 0.7472690398602665,
      "learning_rate": 1.6245623345177203e-06,
      "loss": 0.28908791542053225,
      "num_input_tokens_seen": 696742696,
      "step": 21860,
      "train_runtime": 57294.9559,
      "train_tokens_per_second": 12160.629
    },
    {
      "epoch": 1.4721922973336925,
      "grad_norm": 0.6787669045710994,
      "learning_rate": 1.622612167649683e-06,
      "loss": 0.28769631385803224,
      "num_input_tokens_seen": 696897816,
      "step": 21865,
      "train_runtime": 57307.7516,
      "train_tokens_per_second": 12160.62
    },
    {
      "epoch": 1.4725289523296525,
      "grad_norm": 0.721393453080723,
      "learning_rate": 1.6206629452573868e-06,
      "loss": 0.27122280597686765,
      "num_input_tokens_seen": 697059128,
      "step": 21870,
      "train_runtime": 57320.8263,
      "train_tokens_per_second": 12160.661
    },
    {
      "epoch": 1.4728656073256128,
      "grad_norm": 0.8095693695292882,
      "learning_rate": 1.618714667885919e-06,
      "loss": 0.28151531219482423,
      "num_input_tokens_seen": 697220856,
      "step": 21875,
      "train_runtime": 57334.1474,
      "train_tokens_per_second": 12160.656
    },
    {
      "epoch": 1.4732022623215728,
      "grad_norm": 0.8307888372213474,
      "learning_rate": 1.6167673360801151e-06,
      "loss": 0.28565587997436526,
      "num_input_tokens_seen": 697383408,
      "step": 21880,
      "train_runtime": 57347.4356,
      "train_tokens_per_second": 12160.673
    },
    {
      "epoch": 1.473538917317533,
      "grad_norm": 0.8557104639694532,
      "learning_rate": 1.6148209503845336e-06,
      "loss": 0.3425395965576172,
      "num_input_tokens_seen": 697543400,
      "step": 21885,
      "train_runtime": 57360.2743,
      "train_tokens_per_second": 12160.74
    },
    {
      "epoch": 1.473875572313493,
      "grad_norm": 0.8261641029621242,
      "learning_rate": 1.6128755113434808e-06,
      "loss": 0.29276294708251954,
      "num_input_tokens_seen": 697702640,
      "step": 21890,
      "train_runtime": 57373.3111,
      "train_tokens_per_second": 12160.753
    },
    {
      "epoch": 1.4742122273094533,
      "grad_norm": 0.7460022438197418,
      "learning_rate": 1.6109310195009908e-06,
      "loss": 0.29490392208099364,
      "num_input_tokens_seen": 697866184,
      "step": 21895,
      "train_runtime": 57385.744,
      "train_tokens_per_second": 12160.968
    },
    {
      "epoch": 1.4745488823054134,
      "grad_norm": 0.8588584589338877,
      "learning_rate": 1.6089874754008323e-06,
      "loss": 0.2767459869384766,
      "num_input_tokens_seen": 698022304,
      "step": 21900,
      "train_runtime": 57399.5422,
      "train_tokens_per_second": 12160.764
    },
    {
      "epoch": 1.4748855373013736,
      "grad_norm": 0.8104450979587057,
      "learning_rate": 1.6070448795865162e-06,
      "loss": 0.2919157028198242,
      "num_input_tokens_seen": 698182936,
      "step": 21905,
      "train_runtime": 57412.6653,
      "train_tokens_per_second": 12160.783
    },
    {
      "epoch": 1.4752221922973336,
      "grad_norm": 0.9028325624474686,
      "learning_rate": 1.605103232601277e-06,
      "loss": 0.29503540992736815,
      "num_input_tokens_seen": 698339728,
      "step": 21910,
      "train_runtime": 57425.5182,
      "train_tokens_per_second": 12160.791
    },
    {
      "epoch": 1.475558847293294,
      "grad_norm": 0.6623573356343531,
      "learning_rate": 1.603162534988096e-06,
      "loss": 0.25478668212890626,
      "num_input_tokens_seen": 698500000,
      "step": 21915,
      "train_runtime": 57438.1055,
      "train_tokens_per_second": 12160.916
    },
    {
      "epoch": 1.475895502289254,
      "grad_norm": 0.8076918869329043,
      "learning_rate": 1.6012227872896791e-06,
      "loss": 0.2746097564697266,
      "num_input_tokens_seen": 698657976,
      "step": 21920,
      "train_runtime": 57451.1495,
      "train_tokens_per_second": 12160.905
    },
    {
      "epoch": 1.4762321572852142,
      "grad_norm": 0.7659695292709491,
      "learning_rate": 1.599283990048472e-06,
      "loss": 0.2935265064239502,
      "num_input_tokens_seen": 698812176,
      "step": 21925,
      "train_runtime": 57464.5066,
      "train_tokens_per_second": 12160.762
    },
    {
      "epoch": 1.4765688122811742,
      "grad_norm": 0.8418017216775949,
      "learning_rate": 1.5973461438066533e-06,
      "loss": 0.29478678703308103,
      "num_input_tokens_seen": 698973104,
      "step": 21930,
      "train_runtime": 57477.5088,
      "train_tokens_per_second": 12160.811
    },
    {
      "epoch": 1.4769054672771345,
      "grad_norm": 0.7919703500226196,
      "learning_rate": 1.5954092491061351e-06,
      "loss": 0.29035129547119143,
      "num_input_tokens_seen": 699134968,
      "step": 21935,
      "train_runtime": 57490.6451,
      "train_tokens_per_second": 12160.847
    },
    {
      "epoch": 1.4772421222730945,
      "grad_norm": 0.6678113689151396,
      "learning_rate": 1.5934733064885637e-06,
      "loss": 0.2938823223114014,
      "num_input_tokens_seen": 699297352,
      "step": 21940,
      "train_runtime": 57503.2752,
      "train_tokens_per_second": 12161.0
    },
    {
      "epoch": 1.4775787772690547,
      "grad_norm": 0.8937976617522727,
      "learning_rate": 1.5915383164953203e-06,
      "loss": 0.2971139430999756,
      "num_input_tokens_seen": 699452064,
      "step": 21945,
      "train_runtime": 57516.166,
      "train_tokens_per_second": 12160.965
    },
    {
      "epoch": 1.4779154322650148,
      "grad_norm": 0.8957473532669773,
      "learning_rate": 1.5896042796675155e-06,
      "loss": 0.2998655319213867,
      "num_input_tokens_seen": 699607440,
      "step": 21950,
      "train_runtime": 57529.1895,
      "train_tokens_per_second": 12160.913
    },
    {
      "epoch": 1.478252087260975,
      "grad_norm": 0.780653909346158,
      "learning_rate": 1.5876711965460023e-06,
      "loss": 0.3008512020111084,
      "num_input_tokens_seen": 699769240,
      "step": 21955,
      "train_runtime": 57542.3704,
      "train_tokens_per_second": 12160.939
    },
    {
      "epoch": 1.478588742256935,
      "grad_norm": 0.7571595357338647,
      "learning_rate": 1.5857390676713546e-06,
      "loss": 0.2968494415283203,
      "num_input_tokens_seen": 699922848,
      "step": 21960,
      "train_runtime": 57556.0221,
      "train_tokens_per_second": 12160.723
    },
    {
      "epoch": 1.4789253972528953,
      "grad_norm": 0.7644449740339404,
      "learning_rate": 1.5838078935838913e-06,
      "loss": 0.2694641351699829,
      "num_input_tokens_seen": 700077872,
      "step": 21965,
      "train_runtime": 57568.8696,
      "train_tokens_per_second": 12160.702
    },
    {
      "epoch": 1.4792620522488553,
      "grad_norm": 0.8159690709468568,
      "learning_rate": 1.581877674823657e-06,
      "loss": 0.2921644926071167,
      "num_input_tokens_seen": 700232352,
      "step": 21970,
      "train_runtime": 57582.3183,
      "train_tokens_per_second": 12160.545
    },
    {
      "epoch": 1.4795987072448156,
      "grad_norm": 0.7132647017481912,
      "learning_rate": 1.5799484119304308e-06,
      "loss": 0.30087804794311523,
      "num_input_tokens_seen": 700390120,
      "step": 21975,
      "train_runtime": 57594.9229,
      "train_tokens_per_second": 12160.623
    },
    {
      "epoch": 1.4799353622407756,
      "grad_norm": 0.7854791737176585,
      "learning_rate": 1.578020105443726e-06,
      "loss": 0.2734895467758179,
      "num_input_tokens_seen": 700550664,
      "step": 21980,
      "train_runtime": 57607.8592,
      "train_tokens_per_second": 12160.679
    },
    {
      "epoch": 1.4802720172367358,
      "grad_norm": 0.9393453472689324,
      "learning_rate": 1.576092755902787e-06,
      "loss": 0.3157112836837769,
      "num_input_tokens_seen": 700710728,
      "step": 21985,
      "train_runtime": 57620.8357,
      "train_tokens_per_second": 12160.718
    },
    {
      "epoch": 1.4806086722326959,
      "grad_norm": 0.7854820115765618,
      "learning_rate": 1.5741663638465903e-06,
      "loss": 0.3182875871658325,
      "num_input_tokens_seen": 700863096,
      "step": 21990,
      "train_runtime": 57633.3745,
      "train_tokens_per_second": 12160.716
    },
    {
      "epoch": 1.4809453272286561,
      "grad_norm": 0.808013459384627,
      "learning_rate": 1.5722409298138463e-06,
      "loss": 0.2827549934387207,
      "num_input_tokens_seen": 701018928,
      "step": 21995,
      "train_runtime": 57646.7984,
      "train_tokens_per_second": 12160.587
    },
    {
      "epoch": 1.4812819822246162,
      "grad_norm": 1.0220902810696029,
      "learning_rate": 1.570316454342994e-06,
      "loss": 0.3213972568511963,
      "num_input_tokens_seen": 701178224,
      "step": 22000,
      "train_runtime": 57660.8298,
      "train_tokens_per_second": 12160.391
    },
    {
      "epoch": 1.4816186372205764,
      "grad_norm": 0.7058270491010112,
      "learning_rate": 1.568392937972212e-06,
      "loss": 0.29758486747741697,
      "num_input_tokens_seen": 701342064,
      "step": 22005,
      "train_runtime": 57673.2655,
      "train_tokens_per_second": 12160.61
    },
    {
      "epoch": 1.4819552922165364,
      "grad_norm": 0.8648784223813537,
      "learning_rate": 1.5664703812393988e-06,
      "loss": 0.2885399580001831,
      "num_input_tokens_seen": 701505904,
      "step": 22010,
      "train_runtime": 57685.6923,
      "train_tokens_per_second": 12160.83
    },
    {
      "epoch": 1.4822919472124967,
      "grad_norm": 0.7900199187374539,
      "learning_rate": 1.5645487846821967e-06,
      "loss": 0.2926321506500244,
      "num_input_tokens_seen": 701668400,
      "step": 22015,
      "train_runtime": 57699.0809,
      "train_tokens_per_second": 12160.825
    },
    {
      "epoch": 1.4826286022084567,
      "grad_norm": 0.7888468998175585,
      "learning_rate": 1.5626281488379724e-06,
      "loss": 0.3058467388153076,
      "num_input_tokens_seen": 701829584,
      "step": 22020,
      "train_runtime": 57712.3597,
      "train_tokens_per_second": 12160.819
    },
    {
      "epoch": 1.482965257204417,
      "grad_norm": 0.7881499624685758,
      "learning_rate": 1.560708474243825e-06,
      "loss": 0.27547659873962405,
      "num_input_tokens_seen": 701991776,
      "step": 22025,
      "train_runtime": 57726.3047,
      "train_tokens_per_second": 12160.691
    },
    {
      "epoch": 1.483301912200377,
      "grad_norm": 0.9682973109498847,
      "learning_rate": 1.5587897614365854e-06,
      "loss": 0.3041182518005371,
      "num_input_tokens_seen": 702154296,
      "step": 22030,
      "train_runtime": 57738.9642,
      "train_tokens_per_second": 12160.84
    },
    {
      "epoch": 1.4836385671963372,
      "grad_norm": 0.8078596499091372,
      "learning_rate": 1.5568720109528163e-06,
      "loss": 0.2979097366333008,
      "num_input_tokens_seen": 702316168,
      "step": 22035,
      "train_runtime": 57752.1685,
      "train_tokens_per_second": 12160.862
    },
    {
      "epoch": 1.4839752221922973,
      "grad_norm": 0.8198291866225513,
      "learning_rate": 1.5549552233288095e-06,
      "loss": 0.318677282333374,
      "num_input_tokens_seen": 702470552,
      "step": 22040,
      "train_runtime": 57764.7047,
      "train_tokens_per_second": 12160.896
    },
    {
      "epoch": 1.4843118771882575,
      "grad_norm": 0.7682584202868828,
      "learning_rate": 1.5530393991005893e-06,
      "loss": 0.2923455715179443,
      "num_input_tokens_seen": 702631880,
      "step": 22045,
      "train_runtime": 57777.67,
      "train_tokens_per_second": 12160.959
    },
    {
      "epoch": 1.4846485321842176,
      "grad_norm": 0.7964315205715882,
      "learning_rate": 1.551124538803908e-06,
      "loss": 0.29217450618743895,
      "num_input_tokens_seen": 702790928,
      "step": 22050,
      "train_runtime": 57790.2816,
      "train_tokens_per_second": 12161.057
    },
    {
      "epoch": 1.4849851871801778,
      "grad_norm": 0.8168480951740427,
      "learning_rate": 1.5492106429742547e-06,
      "loss": 0.30246269702911377,
      "num_input_tokens_seen": 702951992,
      "step": 22055,
      "train_runtime": 57803.4446,
      "train_tokens_per_second": 12161.074
    },
    {
      "epoch": 1.4853218421761378,
      "grad_norm": 0.8624772194104877,
      "learning_rate": 1.5472977121468386e-06,
      "loss": 0.30541276931762695,
      "num_input_tokens_seen": 703113344,
      "step": 22060,
      "train_runtime": 57816.3273,
      "train_tokens_per_second": 12161.155
    },
    {
      "epoch": 1.485658497172098,
      "grad_norm": 0.8686719580731861,
      "learning_rate": 1.545385746856609e-06,
      "loss": 0.2763216495513916,
      "num_input_tokens_seen": 703267888,
      "step": 22065,
      "train_runtime": 57829.0322,
      "train_tokens_per_second": 12161.156
    },
    {
      "epoch": 1.485995152168058,
      "grad_norm": 0.7644751620951357,
      "learning_rate": 1.5434747476382388e-06,
      "loss": 0.2551461696624756,
      "num_input_tokens_seen": 703431032,
      "step": 22070,
      "train_runtime": 57841.9206,
      "train_tokens_per_second": 12161.267
    },
    {
      "epoch": 1.4863318071640184,
      "grad_norm": 0.9413585639050013,
      "learning_rate": 1.5415647150261337e-06,
      "loss": 0.2813568115234375,
      "num_input_tokens_seen": 703585856,
      "step": 22075,
      "train_runtime": 57855.2174,
      "train_tokens_per_second": 12161.148
    },
    {
      "epoch": 1.4866684621599784,
      "grad_norm": 0.8302429649355397,
      "learning_rate": 1.539655649554428e-06,
      "loss": 0.3002461910247803,
      "num_input_tokens_seen": 703746360,
      "step": 22080,
      "train_runtime": 57868.6384,
      "train_tokens_per_second": 12161.101
    },
    {
      "epoch": 1.4870051171559386,
      "grad_norm": 0.8390172593398922,
      "learning_rate": 1.5377475517569856e-06,
      "loss": 0.28984642028808594,
      "num_input_tokens_seen": 703902616,
      "step": 22085,
      "train_runtime": 57881.656,
      "train_tokens_per_second": 12161.066
    },
    {
      "epoch": 1.4873417721518987,
      "grad_norm": 0.8830222774369477,
      "learning_rate": 1.5358404221674e-06,
      "loss": 0.30376505851745605,
      "num_input_tokens_seen": 704061704,
      "step": 22090,
      "train_runtime": 57895.2346,
      "train_tokens_per_second": 12160.961
    },
    {
      "epoch": 1.487678427147859,
      "grad_norm": 0.7664812878488345,
      "learning_rate": 1.533934261318994e-06,
      "loss": 0.2926789283752441,
      "num_input_tokens_seen": 704219440,
      "step": 22095,
      "train_runtime": 57908.0521,
      "train_tokens_per_second": 12160.993
    },
    {
      "epoch": 1.488015082143819,
      "grad_norm": 0.7282356372573082,
      "learning_rate": 1.5320290697448177e-06,
      "loss": 0.2853104114532471,
      "num_input_tokens_seen": 704378704,
      "step": 22100,
      "train_runtime": 57921.0548,
      "train_tokens_per_second": 12161.013
    },
    {
      "epoch": 1.4883517371397792,
      "grad_norm": 0.8995352281082737,
      "learning_rate": 1.530124847977657e-06,
      "loss": 0.3070134878158569,
      "num_input_tokens_seen": 704540040,
      "step": 22105,
      "train_runtime": 57934.9604,
      "train_tokens_per_second": 12160.879
    },
    {
      "epoch": 1.4886883921357392,
      "grad_norm": 0.8458146075863114,
      "learning_rate": 1.528221596550014e-06,
      "loss": 0.2793126106262207,
      "num_input_tokens_seen": 704702584,
      "step": 22110,
      "train_runtime": 57948.3121,
      "train_tokens_per_second": 12160.882
    },
    {
      "epoch": 1.4890250471316995,
      "grad_norm": 0.809014193419703,
      "learning_rate": 1.5263193159941331e-06,
      "loss": 0.29033966064453126,
      "num_input_tokens_seen": 704863752,
      "step": 22115,
      "train_runtime": 57960.7829,
      "train_tokens_per_second": 12161.046
    },
    {
      "epoch": 1.4893617021276595,
      "grad_norm": 0.7676854559074024,
      "learning_rate": 1.5244180068419783e-06,
      "loss": 0.27314608097076415,
      "num_input_tokens_seen": 705024744,
      "step": 22120,
      "train_runtime": 57973.884,
      "train_tokens_per_second": 12161.075
    },
    {
      "epoch": 1.4896983571236198,
      "grad_norm": 0.9208978890852293,
      "learning_rate": 1.5225176696252453e-06,
      "loss": 0.3191307544708252,
      "num_input_tokens_seen": 705182712,
      "step": 22125,
      "train_runtime": 57986.9962,
      "train_tokens_per_second": 12161.049
    },
    {
      "epoch": 1.4900350121195798,
      "grad_norm": 0.9266664913064606,
      "learning_rate": 1.5206183048753575e-06,
      "loss": 0.3077064514160156,
      "num_input_tokens_seen": 705341608,
      "step": 22130,
      "train_runtime": 57999.6095,
      "train_tokens_per_second": 12161.144
    },
    {
      "epoch": 1.49037166711554,
      "grad_norm": 1.0508839842852964,
      "learning_rate": 1.5187199131234653e-06,
      "loss": 0.3335201501846313,
      "num_input_tokens_seen": 705499544,
      "step": 22135,
      "train_runtime": 58012.6307,
      "train_tokens_per_second": 12161.137
    },
    {
      "epoch": 1.4907083221115,
      "grad_norm": 0.7955013280795981,
      "learning_rate": 1.5168224949004484e-06,
      "loss": 0.30562918186187743,
      "num_input_tokens_seen": 705655032,
      "step": 22140,
      "train_runtime": 58025.7035,
      "train_tokens_per_second": 12161.077
    },
    {
      "epoch": 1.4910449771074603,
      "grad_norm": 0.7839344645502176,
      "learning_rate": 1.514926050736914e-06,
      "loss": 0.28303353786468505,
      "num_input_tokens_seen": 705816760,
      "step": 22145,
      "train_runtime": 58038.1992,
      "train_tokens_per_second": 12161.245
    },
    {
      "epoch": 1.4913816321034203,
      "grad_norm": 0.7730378068372598,
      "learning_rate": 1.5130305811631945e-06,
      "loss": 0.30832529067993164,
      "num_input_tokens_seen": 705976032,
      "step": 22150,
      "train_runtime": 58050.6702,
      "train_tokens_per_second": 12161.376
    },
    {
      "epoch": 1.4917182870993806,
      "grad_norm": 0.840435743762005,
      "learning_rate": 1.5111360867093571e-06,
      "loss": 0.3058335542678833,
      "num_input_tokens_seen": 706139752,
      "step": 22155,
      "train_runtime": 58063.7795,
      "train_tokens_per_second": 12161.45
    },
    {
      "epoch": 1.4920549420953406,
      "grad_norm": 0.8598024051610549,
      "learning_rate": 1.5092425679051848e-06,
      "loss": 0.2843809127807617,
      "num_input_tokens_seen": 706298832,
      "step": 22160,
      "train_runtime": 58076.4458,
      "train_tokens_per_second": 12161.537
    },
    {
      "epoch": 1.4923915970913009,
      "grad_norm": 0.864458486016635,
      "learning_rate": 1.5073500252801988e-06,
      "loss": 0.31615653038024905,
      "num_input_tokens_seen": 706459136,
      "step": 22165,
      "train_runtime": 58089.0851,
      "train_tokens_per_second": 12161.65
    },
    {
      "epoch": 1.492728252087261,
      "grad_norm": 0.8189887657231626,
      "learning_rate": 1.5054584593636412e-06,
      "loss": 0.2970459222793579,
      "num_input_tokens_seen": 706621512,
      "step": 22170,
      "train_runtime": 58102.3664,
      "train_tokens_per_second": 12161.665
    },
    {
      "epoch": 1.4930649070832211,
      "grad_norm": 0.8723747129265235,
      "learning_rate": 1.5035678706844814e-06,
      "loss": 0.31584391593933103,
      "num_input_tokens_seen": 706782264,
      "step": 22175,
      "train_runtime": 58115.5711,
      "train_tokens_per_second": 12161.668
    },
    {
      "epoch": 1.4934015620791812,
      "grad_norm": 0.7412093337505787,
      "learning_rate": 1.501678259771418e-06,
      "loss": 0.3040097713470459,
      "num_input_tokens_seen": 706943200,
      "step": 22180,
      "train_runtime": 58128.4511,
      "train_tokens_per_second": 12161.742
    },
    {
      "epoch": 1.4937382170751414,
      "grad_norm": 0.8987352374170409,
      "learning_rate": 1.499789627152874e-06,
      "loss": 0.3002941131591797,
      "num_input_tokens_seen": 707103960,
      "step": 22185,
      "train_runtime": 58141.2641,
      "train_tokens_per_second": 12161.826
    },
    {
      "epoch": 1.4940748720711015,
      "grad_norm": 0.8072341876850407,
      "learning_rate": 1.497901973356999e-06,
      "loss": 0.26718697547912595,
      "num_input_tokens_seen": 707264368,
      "step": 22190,
      "train_runtime": 58154.7745,
      "train_tokens_per_second": 12161.759
    },
    {
      "epoch": 1.4944115270670617,
      "grad_norm": 0.7275727890913124,
      "learning_rate": 1.4960152989116694e-06,
      "loss": 0.29034998416900637,
      "num_input_tokens_seen": 707424280,
      "step": 22195,
      "train_runtime": 58167.2673,
      "train_tokens_per_second": 12161.896
    },
    {
      "epoch": 1.4947481820630217,
      "grad_norm": 0.8496728849197415,
      "learning_rate": 1.4941296043444869e-06,
      "loss": 0.3018314361572266,
      "num_input_tokens_seen": 707582296,
      "step": 22200,
      "train_runtime": 58180.5199,
      "train_tokens_per_second": 12161.842
    },
    {
      "epoch": 1.495084837058982,
      "grad_norm": 0.7056862259275045,
      "learning_rate": 1.4922448901827847e-06,
      "loss": 0.28294010162353517,
      "num_input_tokens_seen": 707743624,
      "step": 22205,
      "train_runtime": 58193.0068,
      "train_tokens_per_second": 12162.005
    },
    {
      "epoch": 1.495421492054942,
      "grad_norm": 0.8956731023139126,
      "learning_rate": 1.4903611569536103e-06,
      "loss": 0.28267407417297363,
      "num_input_tokens_seen": 707891424,
      "step": 22210,
      "train_runtime": 58205.7386,
      "train_tokens_per_second": 12161.884
    },
    {
      "epoch": 1.4957581470509023,
      "grad_norm": 0.7947114245491367,
      "learning_rate": 1.488478405183748e-06,
      "loss": 0.3011770486831665,
      "num_input_tokens_seen": 708042944,
      "step": 22215,
      "train_runtime": 58219.0039,
      "train_tokens_per_second": 12161.715
    },
    {
      "epoch": 1.4960948020468623,
      "grad_norm": 0.7976741320470214,
      "learning_rate": 1.4865966353997024e-06,
      "loss": 0.2968148231506348,
      "num_input_tokens_seen": 708205680,
      "step": 22220,
      "train_runtime": 58231.9928,
      "train_tokens_per_second": 12161.797
    },
    {
      "epoch": 1.4964314570428225,
      "grad_norm": 0.7219478079552263,
      "learning_rate": 1.4847158481277047e-06,
      "loss": 0.2789992094039917,
      "num_input_tokens_seen": 708365968,
      "step": 22225,
      "train_runtime": 58244.7177,
      "train_tokens_per_second": 12161.892
    },
    {
      "epoch": 1.4967681120387826,
      "grad_norm": 0.881694178453512,
      "learning_rate": 1.48283604389371e-06,
      "loss": 0.27448248863220215,
      "num_input_tokens_seen": 708521232,
      "step": 22230,
      "train_runtime": 58257.5856,
      "train_tokens_per_second": 12161.871
    },
    {
      "epoch": 1.4971047670347428,
      "grad_norm": 0.6995602961530029,
      "learning_rate": 1.480957223223401e-06,
      "loss": 0.27577791213989256,
      "num_input_tokens_seen": 708684016,
      "step": 22235,
      "train_runtime": 58271.037,
      "train_tokens_per_second": 12161.857
    },
    {
      "epoch": 1.497441422030703,
      "grad_norm": 0.8239582683652602,
      "learning_rate": 1.4790793866421826e-06,
      "loss": 0.33448789119720457,
      "num_input_tokens_seen": 708839272,
      "step": 22240,
      "train_runtime": 58283.7134,
      "train_tokens_per_second": 12161.876
    },
    {
      "epoch": 1.497778077026663,
      "grad_norm": 0.8914818108155549,
      "learning_rate": 1.4772025346751862e-06,
      "loss": 0.2982901334762573,
      "num_input_tokens_seen": 708995072,
      "step": 22245,
      "train_runtime": 58296.4886,
      "train_tokens_per_second": 12161.883
    },
    {
      "epoch": 1.4981147320226231,
      "grad_norm": 0.7451035236587146,
      "learning_rate": 1.475326667847266e-06,
      "loss": 0.3039821147918701,
      "num_input_tokens_seen": 709155888,
      "step": 22250,
      "train_runtime": 58309.921,
      "train_tokens_per_second": 12161.839
    },
    {
      "epoch": 1.4984513870185834,
      "grad_norm": 0.9273829571900518,
      "learning_rate": 1.4734517866830073e-06,
      "loss": 0.3145290374755859,
      "num_input_tokens_seen": 709318632,
      "step": 22255,
      "train_runtime": 58322.6633,
      "train_tokens_per_second": 12161.973
    },
    {
      "epoch": 1.4987880420145436,
      "grad_norm": 0.8181070989350961,
      "learning_rate": 1.4715778917067081e-06,
      "loss": 0.30013322830200195,
      "num_input_tokens_seen": 709478800,
      "step": 22260,
      "train_runtime": 58336.0589,
      "train_tokens_per_second": 12161.925
    },
    {
      "epoch": 1.4991246970105037,
      "grad_norm": 0.7488580426882897,
      "learning_rate": 1.4697049834424016e-06,
      "loss": 0.27686998844146726,
      "num_input_tokens_seen": 709635488,
      "step": 22265,
      "train_runtime": 58348.7333,
      "train_tokens_per_second": 12161.969
    },
    {
      "epoch": 1.4994613520064637,
      "grad_norm": 0.8647920774965072,
      "learning_rate": 1.4678330624138387e-06,
      "loss": 0.2847649097442627,
      "num_input_tokens_seen": 709790816,
      "step": 22270,
      "train_runtime": 58361.5296,
      "train_tokens_per_second": 12161.964
    },
    {
      "epoch": 1.499798007002424,
      "grad_norm": 0.7800853967615423,
      "learning_rate": 1.465962129144497e-06,
      "loss": 0.30553445816040037,
      "num_input_tokens_seen": 709948712,
      "step": 22275,
      "train_runtime": 58374.0188,
      "train_tokens_per_second": 12162.067
    },
    {
      "epoch": 1.5001346619983842,
      "grad_norm": 0.9943155961981985,
      "learning_rate": 1.4640921841575766e-06,
      "loss": 0.3349766254425049,
      "num_input_tokens_seen": 710112096,
      "step": 22280,
      "train_runtime": 58387.0046,
      "train_tokens_per_second": 12162.16
    },
    {
      "epoch": 1.5004713169943442,
      "grad_norm": 1.5199964097145087,
      "learning_rate": 1.4622232279760018e-06,
      "loss": 0.2890611171722412,
      "num_input_tokens_seen": 710269416,
      "step": 22285,
      "train_runtime": 58399.7002,
      "train_tokens_per_second": 12162.21
    },
    {
      "epoch": 1.5008079719903042,
      "grad_norm": 0.8097233404721841,
      "learning_rate": 1.4603552611224202e-06,
      "loss": 0.3098331689834595,
      "num_input_tokens_seen": 710431184,
      "step": 22290,
      "train_runtime": 58412.8103,
      "train_tokens_per_second": 12162.25
    },
    {
      "epoch": 1.5011446269862645,
      "grad_norm": 0.7809107753388665,
      "learning_rate": 1.4584882841192027e-06,
      "loss": 0.30844788551330565,
      "num_input_tokens_seen": 710591216,
      "step": 22295,
      "train_runtime": 58426.6274,
      "train_tokens_per_second": 12162.113
    },
    {
      "epoch": 1.5014812819822247,
      "grad_norm": 0.8348887701436766,
      "learning_rate": 1.456622297488442e-06,
      "loss": 0.2780912399291992,
      "num_input_tokens_seen": 710746720,
      "step": 22300,
      "train_runtime": 58439.6655,
      "train_tokens_per_second": 12162.06
    },
    {
      "epoch": 1.5018179369781848,
      "grad_norm": 0.8328637003948696,
      "learning_rate": 1.4547573017519595e-06,
      "loss": 0.28893346786499025,
      "num_input_tokens_seen": 710905312,
      "step": 22305,
      "train_runtime": 58453.3965,
      "train_tokens_per_second": 12161.916
    },
    {
      "epoch": 1.5021545919741448,
      "grad_norm": 0.8164233513020881,
      "learning_rate": 1.4528932974312897e-06,
      "loss": 0.2727975368499756,
      "num_input_tokens_seen": 711069112,
      "step": 22310,
      "train_runtime": 58466.4917,
      "train_tokens_per_second": 12161.994
    },
    {
      "epoch": 1.502491246970105,
      "grad_norm": 0.8912331195664639,
      "learning_rate": 1.4510302850477005e-06,
      "loss": 0.2958809852600098,
      "num_input_tokens_seen": 711229784,
      "step": 22315,
      "train_runtime": 58480.7305,
      "train_tokens_per_second": 12161.78
    },
    {
      "epoch": 1.5028279019660653,
      "grad_norm": 0.7210165990084275,
      "learning_rate": 1.4491682651221746e-06,
      "loss": 0.274484920501709,
      "num_input_tokens_seen": 711387352,
      "step": 22320,
      "train_runtime": 58493.4348,
      "train_tokens_per_second": 12161.832
    },
    {
      "epoch": 1.5031645569620253,
      "grad_norm": 0.8532580037263516,
      "learning_rate": 1.447307238175421e-06,
      "loss": 0.28702850341796876,
      "num_input_tokens_seen": 711542800,
      "step": 22325,
      "train_runtime": 58506.9314,
      "train_tokens_per_second": 12161.684
    },
    {
      "epoch": 1.5035012119579854,
      "grad_norm": 0.9047123518370407,
      "learning_rate": 1.4454472047278688e-06,
      "loss": 0.29506919384002683,
      "num_input_tokens_seen": 711699336,
      "step": 22330,
      "train_runtime": 58519.9605,
      "train_tokens_per_second": 12161.651
    },
    {
      "epoch": 1.5038378669539456,
      "grad_norm": 0.7637916550949423,
      "learning_rate": 1.443588165299672e-06,
      "loss": 0.2757214069366455,
      "num_input_tokens_seen": 711859600,
      "step": 22335,
      "train_runtime": 58532.917,
      "train_tokens_per_second": 12161.697
    },
    {
      "epoch": 1.5041745219499059,
      "grad_norm": 0.8230997225570305,
      "learning_rate": 1.4417301204107037e-06,
      "loss": 0.31171507835388185,
      "num_input_tokens_seen": 712019120,
      "step": 22340,
      "train_runtime": 58546.1254,
      "train_tokens_per_second": 12161.678
    },
    {
      "epoch": 1.504511176945866,
      "grad_norm": 0.8223000651184691,
      "learning_rate": 1.4398730705805607e-06,
      "loss": 0.28942370414733887,
      "num_input_tokens_seen": 712178696,
      "step": 22345,
      "train_runtime": 58558.871,
      "train_tokens_per_second": 12161.756
    },
    {
      "epoch": 1.504847831941826,
      "grad_norm": 0.7684032323103801,
      "learning_rate": 1.43801701632856e-06,
      "loss": 0.2952467918395996,
      "num_input_tokens_seen": 712340560,
      "step": 22350,
      "train_runtime": 58571.4254,
      "train_tokens_per_second": 12161.913
    },
    {
      "epoch": 1.5051844869377862,
      "grad_norm": 0.8834924161927463,
      "learning_rate": 1.436161958173745e-06,
      "loss": 0.3219234704971313,
      "num_input_tokens_seen": 712503024,
      "step": 22355,
      "train_runtime": 58584.0785,
      "train_tokens_per_second": 12162.059
    },
    {
      "epoch": 1.5055211419337464,
      "grad_norm": 0.7279584088167114,
      "learning_rate": 1.4343078966348718e-06,
      "loss": 0.2985443115234375,
      "num_input_tokens_seen": 712662360,
      "step": 22360,
      "train_runtime": 58597.8737,
      "train_tokens_per_second": 12161.915
    },
    {
      "epoch": 1.5058577969297064,
      "grad_norm": 0.7999296939031575,
      "learning_rate": 1.432454832230426e-06,
      "loss": 0.3008028507232666,
      "num_input_tokens_seen": 712825848,
      "step": 22365,
      "train_runtime": 58610.8782,
      "train_tokens_per_second": 12162.006
    },
    {
      "epoch": 1.5061944519256665,
      "grad_norm": 0.7636266842246873,
      "learning_rate": 1.4306027654786103e-06,
      "loss": 0.27956788539886473,
      "num_input_tokens_seen": 712987768,
      "step": 22370,
      "train_runtime": 58623.8344,
      "train_tokens_per_second": 12162.08
    },
    {
      "epoch": 1.5065311069216267,
      "grad_norm": 0.7680347423752818,
      "learning_rate": 1.4287516968973491e-06,
      "loss": 0.29725937843322753,
      "num_input_tokens_seen": 713144992,
      "step": 22375,
      "train_runtime": 58636.5014,
      "train_tokens_per_second": 12162.134
    },
    {
      "epoch": 1.506867761917587,
      "grad_norm": 0.8195603605176168,
      "learning_rate": 1.4269016270042884e-06,
      "loss": 0.31431739330291747,
      "num_input_tokens_seen": 713307672,
      "step": 22380,
      "train_runtime": 58649.9476,
      "train_tokens_per_second": 12162.12
    },
    {
      "epoch": 1.507204416913547,
      "grad_norm": 0.8767570869419723,
      "learning_rate": 1.425052556316794e-06,
      "loss": 0.28953397274017334,
      "num_input_tokens_seen": 713466992,
      "step": 22385,
      "train_runtime": 58663.1566,
      "train_tokens_per_second": 12162.097
    },
    {
      "epoch": 1.507541071909507,
      "grad_norm": 0.9455128425810174,
      "learning_rate": 1.4232044853519522e-06,
      "loss": 0.29953551292419434,
      "num_input_tokens_seen": 713620984,
      "step": 22390,
      "train_runtime": 58675.8147,
      "train_tokens_per_second": 12162.098
    },
    {
      "epoch": 1.5078777269054673,
      "grad_norm": 0.6681376304839848,
      "learning_rate": 1.4213574146265708e-06,
      "loss": 0.28127517700195315,
      "num_input_tokens_seen": 713774832,
      "step": 22395,
      "train_runtime": 58689.1101,
      "train_tokens_per_second": 12161.964
    },
    {
      "epoch": 1.5082143819014275,
      "grad_norm": 0.835589882630449,
      "learning_rate": 1.4195113446571752e-06,
      "loss": 0.3473880529403687,
      "num_input_tokens_seen": 713932464,
      "step": 22400,
      "train_runtime": 58702.5166,
      "train_tokens_per_second": 12161.872
    },
    {
      "epoch": 1.5085510368973876,
      "grad_norm": 0.7268541500822289,
      "learning_rate": 1.417666275960018e-06,
      "loss": 0.2736452579498291,
      "num_input_tokens_seen": 714095528,
      "step": 22405,
      "train_runtime": 58715.5705,
      "train_tokens_per_second": 12161.945
    },
    {
      "epoch": 1.5088876918933476,
      "grad_norm": 0.7385779066804373,
      "learning_rate": 1.4158222090510605e-06,
      "loss": 0.2688118934631348,
      "num_input_tokens_seen": 714256616,
      "step": 22410,
      "train_runtime": 58728.005,
      "train_tokens_per_second": 12162.113
    },
    {
      "epoch": 1.5092243468893078,
      "grad_norm": 0.7832178835559508,
      "learning_rate": 1.4139791444459945e-06,
      "loss": 0.2954734802246094,
      "num_input_tokens_seen": 714416744,
      "step": 22415,
      "train_runtime": 58741.4938,
      "train_tokens_per_second": 12162.046
    },
    {
      "epoch": 1.509561001885268,
      "grad_norm": 0.939593572603419,
      "learning_rate": 1.4121370826602254e-06,
      "loss": 0.28285238742828367,
      "num_input_tokens_seen": 714572608,
      "step": 22420,
      "train_runtime": 58754.9388,
      "train_tokens_per_second": 12161.916
    },
    {
      "epoch": 1.5098976568812281,
      "grad_norm": 0.8123666936599141,
      "learning_rate": 1.4102960242088803e-06,
      "loss": 0.3100184679031372,
      "num_input_tokens_seen": 714733568,
      "step": 22425,
      "train_runtime": 58767.3769,
      "train_tokens_per_second": 12162.08
    },
    {
      "epoch": 1.5102343118771882,
      "grad_norm": 0.8125567213041536,
      "learning_rate": 1.4084559696068057e-06,
      "loss": 0.28680465221405027,
      "num_input_tokens_seen": 714894728,
      "step": 22430,
      "train_runtime": 58780.2707,
      "train_tokens_per_second": 12162.154
    },
    {
      "epoch": 1.5105709668731484,
      "grad_norm": 0.7331931906633005,
      "learning_rate": 1.406616919368567e-06,
      "loss": 0.30208990573883054,
      "num_input_tokens_seen": 715057648,
      "step": 22435,
      "train_runtime": 58793.2715,
      "train_tokens_per_second": 12162.236
    },
    {
      "epoch": 1.5109076218691087,
      "grad_norm": 0.8375579124249747,
      "learning_rate": 1.4047788740084474e-06,
      "loss": 0.28663761615753175,
      "num_input_tokens_seen": 715218912,
      "step": 22440,
      "train_runtime": 58806.2826,
      "train_tokens_per_second": 12162.287
    },
    {
      "epoch": 1.5112442768650687,
      "grad_norm": 0.7816313221337211,
      "learning_rate": 1.402941834040452e-06,
      "loss": 0.28301610946655276,
      "num_input_tokens_seen": 715380048,
      "step": 22445,
      "train_runtime": 58818.7658,
      "train_tokens_per_second": 12162.446
    },
    {
      "epoch": 1.5115809318610287,
      "grad_norm": 0.7990188400347819,
      "learning_rate": 1.4011057999783006e-06,
      "loss": 0.28008761405944826,
      "num_input_tokens_seen": 715535896,
      "step": 22450,
      "train_runtime": 58832.0122,
      "train_tokens_per_second": 12162.356
    },
    {
      "epoch": 1.511917586856989,
      "grad_norm": 0.7879227757586994,
      "learning_rate": 1.3992707723354398e-06,
      "loss": 0.2888712167739868,
      "num_input_tokens_seen": 715696216,
      "step": 22455,
      "train_runtime": 58844.9241,
      "train_tokens_per_second": 12162.412
    },
    {
      "epoch": 1.5122542418529492,
      "grad_norm": 0.7550642249395687,
      "learning_rate": 1.397436751625022e-06,
      "loss": 0.2787339210510254,
      "num_input_tokens_seen": 715856760,
      "step": 22460,
      "train_runtime": 58857.7312,
      "train_tokens_per_second": 12162.493
    },
    {
      "epoch": 1.5125908968489092,
      "grad_norm": 0.8335895334032981,
      "learning_rate": 1.3956037383599308e-06,
      "loss": 0.28026251792907714,
      "num_input_tokens_seen": 716019240,
      "step": 22465,
      "train_runtime": 58870.3975,
      "train_tokens_per_second": 12162.636
    },
    {
      "epoch": 1.5129275518448693,
      "grad_norm": 0.7381731464961705,
      "learning_rate": 1.39377173305276e-06,
      "loss": 0.2853630542755127,
      "num_input_tokens_seen": 716173976,
      "step": 22470,
      "train_runtime": 58883.3115,
      "train_tokens_per_second": 12162.597
    },
    {
      "epoch": 1.5132642068408295,
      "grad_norm": 0.8496550289629963,
      "learning_rate": 1.3919407362158254e-06,
      "loss": 0.31069705486297605,
      "num_input_tokens_seen": 716337016,
      "step": 22475,
      "train_runtime": 58896.1653,
      "train_tokens_per_second": 12162.711
    },
    {
      "epoch": 1.5136008618367898,
      "grad_norm": 0.8331130134598611,
      "learning_rate": 1.3901107483611582e-06,
      "loss": 0.288267707824707,
      "num_input_tokens_seen": 716490936,
      "step": 22480,
      "train_runtime": 58909.4002,
      "train_tokens_per_second": 12162.591
    },
    {
      "epoch": 1.5139375168327498,
      "grad_norm": 0.7718796041004999,
      "learning_rate": 1.3882817700005091e-06,
      "loss": 0.28669161796569825,
      "num_input_tokens_seen": 716649696,
      "step": 22485,
      "train_runtime": 58921.9462,
      "train_tokens_per_second": 12162.696
    },
    {
      "epoch": 1.5142741718287098,
      "grad_norm": 0.8767547836145987,
      "learning_rate": 1.3864538016453466e-06,
      "loss": 0.3005767822265625,
      "num_input_tokens_seen": 716803584,
      "step": 22490,
      "train_runtime": 58935.6282,
      "train_tokens_per_second": 12162.483
    },
    {
      "epoch": 1.51461082682467,
      "grad_norm": 0.9088718099376718,
      "learning_rate": 1.384626843806855e-06,
      "loss": 0.30231337547302245,
      "num_input_tokens_seen": 716958496,
      "step": 22495,
      "train_runtime": 58949.1198,
      "train_tokens_per_second": 12162.327
    },
    {
      "epoch": 1.5149474818206303,
      "grad_norm": 0.7588852953234072,
      "learning_rate": 1.3828008969959367e-06,
      "loss": 0.30952341556549073,
      "num_input_tokens_seen": 717121360,
      "step": 22500,
      "train_runtime": 58961.8928,
      "train_tokens_per_second": 12162.455
    },
    {
      "epoch": 1.5152841368165904,
      "grad_norm": 0.842371168826284,
      "learning_rate": 1.3809759617232166e-06,
      "loss": 0.28448638916015623,
      "num_input_tokens_seen": 717280976,
      "step": 22505,
      "train_runtime": 58975.2562,
      "train_tokens_per_second": 12162.405
    },
    {
      "epoch": 1.5156207918125504,
      "grad_norm": 0.7704161273473085,
      "learning_rate": 1.3791520384990247e-06,
      "loss": 0.2891338586807251,
      "num_input_tokens_seen": 717440384,
      "step": 22510,
      "train_runtime": 58988.9198,
      "train_tokens_per_second": 12162.291
    },
    {
      "epoch": 1.5159574468085106,
      "grad_norm": 0.724030706598745,
      "learning_rate": 1.3773291278334216e-06,
      "loss": 0.28957176208496094,
      "num_input_tokens_seen": 717600128,
      "step": 22515,
      "train_runtime": 59001.6464,
      "train_tokens_per_second": 12162.375
    },
    {
      "epoch": 1.5162941018044709,
      "grad_norm": 0.7647257817007969,
      "learning_rate": 1.3755072302361754e-06,
      "loss": 0.3053462505340576,
      "num_input_tokens_seen": 717763968,
      "step": 22520,
      "train_runtime": 59014.0707,
      "train_tokens_per_second": 12162.59
    },
    {
      "epoch": 1.516630756800431,
      "grad_norm": 0.7898838442855936,
      "learning_rate": 1.3736863462167749e-06,
      "loss": 0.28431057929992676,
      "num_input_tokens_seen": 717919152,
      "step": 22525,
      "train_runtime": 59027.3597,
      "train_tokens_per_second": 12162.481
    },
    {
      "epoch": 1.516967411796391,
      "grad_norm": 0.8192890951291476,
      "learning_rate": 1.3718664762844242e-06,
      "loss": 0.2807717561721802,
      "num_input_tokens_seen": 718078520,
      "step": 22530,
      "train_runtime": 59040.5826,
      "train_tokens_per_second": 12162.457
    },
    {
      "epoch": 1.5173040667923512,
      "grad_norm": 0.8158887654473395,
      "learning_rate": 1.370047620948044e-06,
      "loss": 0.29694461822509766,
      "num_input_tokens_seen": 718240608,
      "step": 22535,
      "train_runtime": 59053.1726,
      "train_tokens_per_second": 12162.608
    },
    {
      "epoch": 1.5176407217883114,
      "grad_norm": 0.7934396063151912,
      "learning_rate": 1.368229780716272e-06,
      "loss": 0.3102105617523193,
      "num_input_tokens_seen": 718397904,
      "step": 22540,
      "train_runtime": 59066.2801,
      "train_tokens_per_second": 12162.572
    },
    {
      "epoch": 1.5179773767842715,
      "grad_norm": 0.910528152197629,
      "learning_rate": 1.3664129560974604e-06,
      "loss": 0.2921405553817749,
      "num_input_tokens_seen": 718554992,
      "step": 22545,
      "train_runtime": 59079.3707,
      "train_tokens_per_second": 12162.536
    },
    {
      "epoch": 1.5183140317802315,
      "grad_norm": 0.8860137257012383,
      "learning_rate": 1.3645971475996777e-06,
      "loss": 0.2833313226699829,
      "num_input_tokens_seen": 718710184,
      "step": 22550,
      "train_runtime": 59092.7031,
      "train_tokens_per_second": 12162.418
    },
    {
      "epoch": 1.5186506867761917,
      "grad_norm": 0.8524522876712192,
      "learning_rate": 1.3627823557307135e-06,
      "loss": 0.2982018947601318,
      "num_input_tokens_seen": 718873376,
      "step": 22555,
      "train_runtime": 59105.5968,
      "train_tokens_per_second": 12162.526
    },
    {
      "epoch": 1.518987341772152,
      "grad_norm": 0.688286493113632,
      "learning_rate": 1.3609685809980616e-06,
      "loss": 0.30071923732757566,
      "num_input_tokens_seen": 719036056,
      "step": 22560,
      "train_runtime": 59118.2963,
      "train_tokens_per_second": 12162.665
    },
    {
      "epoch": 1.519323996768112,
      "grad_norm": 0.7973915683926959,
      "learning_rate": 1.3591558239089443e-06,
      "loss": 0.28338260650634767,
      "num_input_tokens_seen": 719198320,
      "step": 22565,
      "train_runtime": 59131.5915,
      "train_tokens_per_second": 12162.675
    },
    {
      "epoch": 1.519660651764072,
      "grad_norm": 0.8626309267824943,
      "learning_rate": 1.3573440849702902e-06,
      "loss": 0.3180524826049805,
      "num_input_tokens_seen": 719362160,
      "step": 22570,
      "train_runtime": 59144.0336,
      "train_tokens_per_second": 12162.886
    },
    {
      "epoch": 1.5199973067600323,
      "grad_norm": 0.8471297572397387,
      "learning_rate": 1.3555333646887476e-06,
      "loss": 0.3034017086029053,
      "num_input_tokens_seen": 719521640,
      "step": 22575,
      "train_runtime": 59156.7944,
      "train_tokens_per_second": 12162.959
    },
    {
      "epoch": 1.5203339617559926,
      "grad_norm": 0.7633412615725172,
      "learning_rate": 1.3537236635706779e-06,
      "loss": 0.28931436538696287,
      "num_input_tokens_seen": 719683928,
      "step": 22580,
      "train_runtime": 59169.4205,
      "train_tokens_per_second": 12163.106
    },
    {
      "epoch": 1.5206706167519526,
      "grad_norm": 0.8109797326643368,
      "learning_rate": 1.351914982122159e-06,
      "loss": 0.2645892143249512,
      "num_input_tokens_seen": 719846232,
      "step": 22585,
      "train_runtime": 59181.9802,
      "train_tokens_per_second": 12163.267
    },
    {
      "epoch": 1.5210072717479126,
      "grad_norm": 0.811607137207679,
      "learning_rate": 1.3501073208489823e-06,
      "loss": 0.2924932956695557,
      "num_input_tokens_seen": 720004384,
      "step": 22590,
      "train_runtime": 59195.6953,
      "train_tokens_per_second": 12163.121
    },
    {
      "epoch": 1.5213439267438729,
      "grad_norm": 0.8175464829852251,
      "learning_rate": 1.3483006802566546e-06,
      "loss": 0.3018640995025635,
      "num_input_tokens_seen": 720164416,
      "step": 22595,
      "train_runtime": 59208.4573,
      "train_tokens_per_second": 12163.202
    },
    {
      "epoch": 1.5216805817398331,
      "grad_norm": 0.875511541718078,
      "learning_rate": 1.3464950608503957e-06,
      "loss": 0.3094246625900269,
      "num_input_tokens_seen": 720323688,
      "step": 22600,
      "train_runtime": 59221.902,
      "train_tokens_per_second": 12163.13
    },
    {
      "epoch": 1.5220172367357931,
      "grad_norm": 0.8602197918749629,
      "learning_rate": 1.344690463135146e-06,
      "loss": 0.30721702575683596,
      "num_input_tokens_seen": 720485216,
      "step": 22605,
      "train_runtime": 59235.6512,
      "train_tokens_per_second": 12163.034
    },
    {
      "epoch": 1.5223538917317532,
      "grad_norm": 0.8021495233525912,
      "learning_rate": 1.3428868876155493e-06,
      "loss": 0.3159857988357544,
      "num_input_tokens_seen": 720635112,
      "step": 22610,
      "train_runtime": 59248.4272,
      "train_tokens_per_second": 12162.941
    },
    {
      "epoch": 1.5226905467277134,
      "grad_norm": 0.6938520051260284,
      "learning_rate": 1.3410843347959745e-06,
      "loss": 0.2776130437850952,
      "num_input_tokens_seen": 720792152,
      "step": 22615,
      "train_runtime": 59261.2745,
      "train_tokens_per_second": 12162.954
    },
    {
      "epoch": 1.5230272017236737,
      "grad_norm": 0.7495250598648787,
      "learning_rate": 1.3392828051804979e-06,
      "loss": 0.3007868766784668,
      "num_input_tokens_seen": 720954712,
      "step": 22620,
      "train_runtime": 59273.8851,
      "train_tokens_per_second": 12163.109
    },
    {
      "epoch": 1.5233638567196337,
      "grad_norm": 0.735830102701975,
      "learning_rate": 1.3374822992729114e-06,
      "loss": 0.3064920902252197,
      "num_input_tokens_seen": 721117416,
      "step": 22625,
      "train_runtime": 59286.6069,
      "train_tokens_per_second": 12163.243
    },
    {
      "epoch": 1.5237005117155937,
      "grad_norm": 0.7985835634234847,
      "learning_rate": 1.3356828175767212e-06,
      "loss": 0.2888817548751831,
      "num_input_tokens_seen": 721275280,
      "step": 22630,
      "train_runtime": 59300.425,
      "train_tokens_per_second": 12163.071
    },
    {
      "epoch": 1.524037166711554,
      "grad_norm": 0.739255684162013,
      "learning_rate": 1.3338843605951462e-06,
      "loss": 0.2747783899307251,
      "num_input_tokens_seen": 721438280,
      "step": 22635,
      "train_runtime": 59313.931,
      "train_tokens_per_second": 12163.05
    },
    {
      "epoch": 1.5243738217075142,
      "grad_norm": 0.7797345871202387,
      "learning_rate": 1.33208692883112e-06,
      "loss": 0.2792922019958496,
      "num_input_tokens_seen": 721602120,
      "step": 22640,
      "train_runtime": 59326.3537,
      "train_tokens_per_second": 12163.264
    },
    {
      "epoch": 1.5247104767034743,
      "grad_norm": 0.6916888648656614,
      "learning_rate": 1.3302905227872876e-06,
      "loss": 0.2778587818145752,
      "num_input_tokens_seen": 721765688,
      "step": 22645,
      "train_runtime": 59339.3711,
      "train_tokens_per_second": 12163.353
    },
    {
      "epoch": 1.5250471316994343,
      "grad_norm": 0.7997440011215657,
      "learning_rate": 1.328495142966007e-06,
      "loss": 0.29504852294921874,
      "num_input_tokens_seen": 721923184,
      "step": 22650,
      "train_runtime": 59352.963,
      "train_tokens_per_second": 12163.221
    },
    {
      "epoch": 1.5253837866953945,
      "grad_norm": 0.768358857052409,
      "learning_rate": 1.3267007898693552e-06,
      "loss": 0.2853509187698364,
      "num_input_tokens_seen": 722087024,
      "step": 22655,
      "train_runtime": 59365.4077,
      "train_tokens_per_second": 12163.431
    },
    {
      "epoch": 1.5257204416913548,
      "grad_norm": 0.8019756286141467,
      "learning_rate": 1.324907463999111e-06,
      "loss": 0.30272064208984373,
      "num_input_tokens_seen": 722249704,
      "step": 22660,
      "train_runtime": 59378.111,
      "train_tokens_per_second": 12163.568
    },
    {
      "epoch": 1.5260570966873148,
      "grad_norm": 0.891640594431585,
      "learning_rate": 1.3231151658567776e-06,
      "loss": 0.30310521125793455,
      "num_input_tokens_seen": 722403048,
      "step": 22665,
      "train_runtime": 59391.1817,
      "train_tokens_per_second": 12163.473
    },
    {
      "epoch": 1.5263937516832748,
      "grad_norm": 0.8837917343262014,
      "learning_rate": 1.321323895943563e-06,
      "loss": 0.3033485174179077,
      "num_input_tokens_seen": 722564720,
      "step": 22670,
      "train_runtime": 59403.7109,
      "train_tokens_per_second": 12163.629
    },
    {
      "epoch": 1.526730406679235,
      "grad_norm": 0.8414900977242138,
      "learning_rate": 1.3195336547603905e-06,
      "loss": 0.3113853454589844,
      "num_input_tokens_seen": 722725168,
      "step": 22675,
      "train_runtime": 59416.5372,
      "train_tokens_per_second": 12163.704
    },
    {
      "epoch": 1.5270670616751953,
      "grad_norm": 0.773084013538252,
      "learning_rate": 1.3177444428078951e-06,
      "loss": 0.2729278802871704,
      "num_input_tokens_seen": 722880848,
      "step": 22680,
      "train_runtime": 59429.602,
      "train_tokens_per_second": 12163.649
    },
    {
      "epoch": 1.5274037166711554,
      "grad_norm": 0.8713948220580527,
      "learning_rate": 1.3159562605864245e-06,
      "loss": 0.2988485336303711,
      "num_input_tokens_seen": 723042696,
      "step": 22685,
      "train_runtime": 59442.6691,
      "train_tokens_per_second": 12163.698
    },
    {
      "epoch": 1.5277403716671154,
      "grad_norm": 0.9670116580608046,
      "learning_rate": 1.3141691085960385e-06,
      "loss": 0.29040334224700926,
      "num_input_tokens_seen": 723206536,
      "step": 22690,
      "train_runtime": 59455.1127,
      "train_tokens_per_second": 12163.908
    },
    {
      "epoch": 1.5280770266630757,
      "grad_norm": 0.7650667220008859,
      "learning_rate": 1.3123829873365073e-06,
      "loss": 0.28452630043029786,
      "num_input_tokens_seen": 723364832,
      "step": 22695,
      "train_runtime": 59467.8747,
      "train_tokens_per_second": 12163.96
    },
    {
      "epoch": 1.528413681659036,
      "grad_norm": 0.767114090260498,
      "learning_rate": 1.3105978973073136e-06,
      "loss": 0.2738353729248047,
      "num_input_tokens_seen": 723526200,
      "step": 22700,
      "train_runtime": 59481.4943,
      "train_tokens_per_second": 12163.887
    },
    {
      "epoch": 1.528750336654996,
      "grad_norm": 0.7631789164878374,
      "learning_rate": 1.3088138390076565e-06,
      "loss": 0.29928605556488036,
      "num_input_tokens_seen": 723683688,
      "step": 22705,
      "train_runtime": 59493.9517,
      "train_tokens_per_second": 12163.988
    },
    {
      "epoch": 1.529086991650956,
      "grad_norm": 0.7925612796007708,
      "learning_rate": 1.3070308129364357e-06,
      "loss": 0.27777714729309083,
      "num_input_tokens_seen": 723844872,
      "step": 22710,
      "train_runtime": 59507.6674,
      "train_tokens_per_second": 12163.893
    },
    {
      "epoch": 1.5294236466469162,
      "grad_norm": 0.7922180074886799,
      "learning_rate": 1.3052488195922736e-06,
      "loss": 0.28077311515808107,
      "num_input_tokens_seen": 723999992,
      "step": 22715,
      "train_runtime": 59520.6383,
      "train_tokens_per_second": 12163.848
    },
    {
      "epoch": 1.5297603016428765,
      "grad_norm": 0.8195123139703518,
      "learning_rate": 1.3034678594734972e-06,
      "loss": 0.27829957008361816,
      "num_input_tokens_seen": 724161512,
      "step": 22720,
      "train_runtime": 59533.7273,
      "train_tokens_per_second": 12163.887
    },
    {
      "epoch": 1.5300969566388365,
      "grad_norm": 0.8919690102270895,
      "learning_rate": 1.3016879330781468e-06,
      "loss": 0.30177719593048097,
      "num_input_tokens_seen": 724316960,
      "step": 22725,
      "train_runtime": 59546.8936,
      "train_tokens_per_second": 12163.808
    },
    {
      "epoch": 1.5304336116347965,
      "grad_norm": 0.8242368615781638,
      "learning_rate": 1.2999090409039728e-06,
      "loss": 0.31191267967224123,
      "num_input_tokens_seen": 724480792,
      "step": 22730,
      "train_runtime": 59559.9919,
      "train_tokens_per_second": 12163.883
    },
    {
      "epoch": 1.5307702666307568,
      "grad_norm": 0.8314019220172901,
      "learning_rate": 1.2981311834484367e-06,
      "loss": 0.30935657024383545,
      "num_input_tokens_seen": 724641440,
      "step": 22735,
      "train_runtime": 59572.7804,
      "train_tokens_per_second": 12163.969
    },
    {
      "epoch": 1.531106921626717,
      "grad_norm": 0.7787195498991406,
      "learning_rate": 1.296354361208711e-06,
      "loss": 0.2740952491760254,
      "num_input_tokens_seen": 724803904,
      "step": 22740,
      "train_runtime": 59585.7214,
      "train_tokens_per_second": 12164.054
    },
    {
      "epoch": 1.531443576622677,
      "grad_norm": 0.8618997682714752,
      "learning_rate": 1.294578574681678e-06,
      "loss": 0.31820383071899416,
      "num_input_tokens_seen": 724959232,
      "step": 22745,
      "train_runtime": 59598.9377,
      "train_tokens_per_second": 12163.962
    },
    {
      "epoch": 1.531780231618637,
      "grad_norm": 0.8591799251237343,
      "learning_rate": 1.292803824363929e-06,
      "loss": 0.2783145189285278,
      "num_input_tokens_seen": 725118048,
      "step": 22750,
      "train_runtime": 59611.4971,
      "train_tokens_per_second": 12164.064
    },
    {
      "epoch": 1.5321168866145973,
      "grad_norm": 0.86335692065507,
      "learning_rate": 1.2910301107517726e-06,
      "loss": 0.31418445110321047,
      "num_input_tokens_seen": 725274520,
      "step": 22755,
      "train_runtime": 59624.0029,
      "train_tokens_per_second": 12164.137
    },
    {
      "epoch": 1.5324535416105576,
      "grad_norm": 0.869848542243531,
      "learning_rate": 1.2892574343412156e-06,
      "loss": 0.3068138360977173,
      "num_input_tokens_seen": 725430432,
      "step": 22760,
      "train_runtime": 59636.9236,
      "train_tokens_per_second": 12164.116
    },
    {
      "epoch": 1.5327901966065176,
      "grad_norm": 0.8262261087203683,
      "learning_rate": 1.2874857956279858e-06,
      "loss": 0.2983036994934082,
      "num_input_tokens_seen": 725589384,
      "step": 22765,
      "train_runtime": 59649.807,
      "train_tokens_per_second": 12164.153
    },
    {
      "epoch": 1.5331268516024776,
      "grad_norm": 0.8332741417095026,
      "learning_rate": 1.2857151951075147e-06,
      "loss": 0.29120535850524903,
      "num_input_tokens_seen": 725751832,
      "step": 22770,
      "train_runtime": 59663.129,
      "train_tokens_per_second": 12164.16
    },
    {
      "epoch": 1.5334635065984379,
      "grad_norm": 0.7566676849727013,
      "learning_rate": 1.2839456332749456e-06,
      "loss": 0.28792848587036135,
      "num_input_tokens_seen": 725911616,
      "step": 22775,
      "train_runtime": 59676.7401,
      "train_tokens_per_second": 12164.063
    },
    {
      "epoch": 1.5338001615943981,
      "grad_norm": 0.8243242187685503,
      "learning_rate": 1.2821771106251308e-06,
      "loss": 0.2932237148284912,
      "num_input_tokens_seen": 726070456,
      "step": 22780,
      "train_runtime": 59690.1077,
      "train_tokens_per_second": 12164.0
    },
    {
      "epoch": 1.5341368165903582,
      "grad_norm": 0.7456731965831106,
      "learning_rate": 1.2804096276526312e-06,
      "loss": 0.270353889465332,
      "num_input_tokens_seen": 726224976,
      "step": 22785,
      "train_runtime": 59702.6365,
      "train_tokens_per_second": 12164.035
    },
    {
      "epoch": 1.5344734715863182,
      "grad_norm": 0.9139920045965834,
      "learning_rate": 1.2786431848517184e-06,
      "loss": 0.2909053087234497,
      "num_input_tokens_seen": 726376784,
      "step": 22790,
      "train_runtime": 59715.4212,
      "train_tokens_per_second": 12163.973
    },
    {
      "epoch": 1.5348101265822784,
      "grad_norm": 0.9023343652822307,
      "learning_rate": 1.2768777827163725e-06,
      "loss": 0.3015217065811157,
      "num_input_tokens_seen": 726537256,
      "step": 22795,
      "train_runtime": 59728.4525,
      "train_tokens_per_second": 12164.006
    },
    {
      "epoch": 1.5351467815782387,
      "grad_norm": 0.8546881227187612,
      "learning_rate": 1.2751134217402811e-06,
      "loss": 0.3047091722488403,
      "num_input_tokens_seen": 726692936,
      "step": 22800,
      "train_runtime": 59742.7543,
      "train_tokens_per_second": 12163.7
    },
    {
      "epoch": 1.5354834365741987,
      "grad_norm": 0.6727762442555562,
      "learning_rate": 1.2733501024168465e-06,
      "loss": 0.2649812698364258,
      "num_input_tokens_seen": 726850352,
      "step": 22805,
      "train_runtime": 59755.2291,
      "train_tokens_per_second": 12163.795
    },
    {
      "epoch": 1.5358200915701588,
      "grad_norm": 0.8998478069133383,
      "learning_rate": 1.2715878252391695e-06,
      "loss": 0.301397442817688,
      "num_input_tokens_seen": 727005944,
      "step": 22810,
      "train_runtime": 59767.9732,
      "train_tokens_per_second": 12163.805
    },
    {
      "epoch": 1.536156746566119,
      "grad_norm": 0.7692228679204597,
      "learning_rate": 1.269826590700069e-06,
      "loss": 0.29273266792297364,
      "num_input_tokens_seen": 727166912,
      "step": 22815,
      "train_runtime": 59781.03,
      "train_tokens_per_second": 12163.84
    },
    {
      "epoch": 1.5364934015620793,
      "grad_norm": 1.195924350138272,
      "learning_rate": 1.2680663992920684e-06,
      "loss": 0.2927964210510254,
      "num_input_tokens_seen": 727328280,
      "step": 22820,
      "train_runtime": 59793.9811,
      "train_tokens_per_second": 12163.905
    },
    {
      "epoch": 1.5368300565580393,
      "grad_norm": 0.8623250586002384,
      "learning_rate": 1.2663072515073988e-06,
      "loss": 0.302318811416626,
      "num_input_tokens_seen": 727484264,
      "step": 22825,
      "train_runtime": 59807.1352,
      "train_tokens_per_second": 12163.837
    },
    {
      "epoch": 1.5371667115539993,
      "grad_norm": 0.7803982946842419,
      "learning_rate": 1.2645491478380002e-06,
      "loss": 0.29267148971557616,
      "num_input_tokens_seen": 727641632,
      "step": 22830,
      "train_runtime": 59820.3704,
      "train_tokens_per_second": 12163.777
    },
    {
      "epoch": 1.5375033665499596,
      "grad_norm": 0.9091394963001928,
      "learning_rate": 1.2627920887755212e-06,
      "loss": 0.2972847938537598,
      "num_input_tokens_seen": 727797728,
      "step": 22835,
      "train_runtime": 59833.9065,
      "train_tokens_per_second": 12163.634
    },
    {
      "epoch": 1.5378400215459198,
      "grad_norm": 0.7623492031549824,
      "learning_rate": 1.261036074811317e-06,
      "loss": 0.31333317756652834,
      "num_input_tokens_seen": 727958440,
      "step": 22840,
      "train_runtime": 59847.3437,
      "train_tokens_per_second": 12163.588
    },
    {
      "epoch": 1.5381766765418798,
      "grad_norm": 0.8041356773921878,
      "learning_rate": 1.2592811064364524e-06,
      "loss": 0.2968874931335449,
      "num_input_tokens_seen": 728118112,
      "step": 22845,
      "train_runtime": 59859.7767,
      "train_tokens_per_second": 12163.729
    },
    {
      "epoch": 1.5385133315378399,
      "grad_norm": 0.8883060070663998,
      "learning_rate": 1.2575271841416963e-06,
      "loss": 0.3248203992843628,
      "num_input_tokens_seen": 728266168,
      "step": 22850,
      "train_runtime": 59873.4152,
      "train_tokens_per_second": 12163.431
    },
    {
      "epoch": 1.5388499865338001,
      "grad_norm": 0.8347727872277496,
      "learning_rate": 1.2557743084175327e-06,
      "loss": 0.27903060913085936,
      "num_input_tokens_seen": 728426208,
      "step": 22855,
      "train_runtime": 59887.0464,
      "train_tokens_per_second": 12163.335
    },
    {
      "epoch": 1.5391866415297604,
      "grad_norm": 0.8186428263176526,
      "learning_rate": 1.254022479754141e-06,
      "loss": 0.28824586868286134,
      "num_input_tokens_seen": 728586456,
      "step": 22860,
      "train_runtime": 59900.3982,
      "train_tokens_per_second": 12163.299
    },
    {
      "epoch": 1.5395232965257204,
      "grad_norm": 0.8318001867188793,
      "learning_rate": 1.252271698641419e-06,
      "loss": 0.292581844329834,
      "num_input_tokens_seen": 728745144,
      "step": 22865,
      "train_runtime": 59912.9646,
      "train_tokens_per_second": 12163.397
    },
    {
      "epoch": 1.5398599515216804,
      "grad_norm": 0.7494089898058677,
      "learning_rate": 1.2505219655689659e-06,
      "loss": 0.29260101318359377,
      "num_input_tokens_seen": 728907088,
      "step": 22870,
      "train_runtime": 59925.5571,
      "train_tokens_per_second": 12163.543
    },
    {
      "epoch": 1.5401966065176407,
      "grad_norm": 0.7680761672967052,
      "learning_rate": 1.2487732810260877e-06,
      "loss": 0.291391921043396,
      "num_input_tokens_seen": 729066992,
      "step": 22875,
      "train_runtime": 59939.2737,
      "train_tokens_per_second": 12163.427
    },
    {
      "epoch": 1.540533261513601,
      "grad_norm": 0.7518227523846464,
      "learning_rate": 1.2470256455018027e-06,
      "loss": 0.2672252655029297,
      "num_input_tokens_seen": 729229976,
      "step": 22880,
      "train_runtime": 59951.7134,
      "train_tokens_per_second": 12163.622
    },
    {
      "epoch": 1.540869916509561,
      "grad_norm": 0.8190723467315617,
      "learning_rate": 1.2452790594848257e-06,
      "loss": 0.26885218620300294,
      "num_input_tokens_seen": 729393016,
      "step": 22885,
      "train_runtime": 59964.5586,
      "train_tokens_per_second": 12163.735
    },
    {
      "epoch": 1.541206571505521,
      "grad_norm": 0.9195826711343672,
      "learning_rate": 1.2435335234635892e-06,
      "loss": 0.31464276313781736,
      "num_input_tokens_seen": 729549096,
      "step": 22890,
      "train_runtime": 59977.2032,
      "train_tokens_per_second": 12163.773
    },
    {
      "epoch": 1.5415432265014812,
      "grad_norm": 0.7666475754739664,
      "learning_rate": 1.2417890379262221e-06,
      "loss": 0.29359161853790283,
      "num_input_tokens_seen": 729711384,
      "step": 22895,
      "train_runtime": 59990.6577,
      "train_tokens_per_second": 12163.75
    },
    {
      "epoch": 1.5418798814974415,
      "grad_norm": 0.8284494431771039,
      "learning_rate": 1.2400456033605672e-06,
      "loss": 0.26127352714538576,
      "num_input_tokens_seen": 729870440,
      "step": 22900,
      "train_runtime": 60004.3696,
      "train_tokens_per_second": 12163.621
    },
    {
      "epoch": 1.5422165364934015,
      "grad_norm": 0.7649351133065323,
      "learning_rate": 1.2383032202541706e-06,
      "loss": 0.31118364334106446,
      "num_input_tokens_seen": 730026080,
      "step": 22905,
      "train_runtime": 60017.2825,
      "train_tokens_per_second": 12163.598
    },
    {
      "epoch": 1.5425531914893615,
      "grad_norm": 0.7822003674051046,
      "learning_rate": 1.2365618890942827e-06,
      "loss": 0.3109957933425903,
      "num_input_tokens_seen": 730183000,
      "step": 22910,
      "train_runtime": 60029.7251,
      "train_tokens_per_second": 12163.691
    },
    {
      "epoch": 1.5428898464853218,
      "grad_norm": 0.8923914012903582,
      "learning_rate": 1.2348216103678618e-06,
      "loss": 0.32297348976135254,
      "num_input_tokens_seen": 730341512,
      "step": 22915,
      "train_runtime": 60042.2927,
      "train_tokens_per_second": 12163.785
    },
    {
      "epoch": 1.543226501481282,
      "grad_norm": 0.8574399113989104,
      "learning_rate": 1.2330823845615713e-06,
      "loss": 0.2824406623840332,
      "num_input_tokens_seen": 730497024,
      "step": 22920,
      "train_runtime": 60056.5156,
      "train_tokens_per_second": 12163.493
    },
    {
      "epoch": 1.543563156477242,
      "grad_norm": 0.7975582089846489,
      "learning_rate": 1.2313442121617785e-06,
      "loss": 0.296402645111084,
      "num_input_tokens_seen": 730659000,
      "step": 22925,
      "train_runtime": 60069.1072,
      "train_tokens_per_second": 12163.64
    },
    {
      "epoch": 1.543899811473202,
      "grad_norm": 0.8426178343566919,
      "learning_rate": 1.229607093654563e-06,
      "loss": 0.2669986248016357,
      "num_input_tokens_seen": 730818272,
      "step": 22930,
      "train_runtime": 60082.0052,
      "train_tokens_per_second": 12163.68
    },
    {
      "epoch": 1.5442364664691623,
      "grad_norm": 0.7771660924237656,
      "learning_rate": 1.2278710295256974e-06,
      "loss": 0.30116374492645265,
      "num_input_tokens_seen": 730980776,
      "step": 22935,
      "train_runtime": 60094.697,
      "train_tokens_per_second": 12163.815
    },
    {
      "epoch": 1.5445731214651226,
      "grad_norm": 0.780681085063772,
      "learning_rate": 1.2261360202606725e-06,
      "loss": 0.27906479835510256,
      "num_input_tokens_seen": 731143056,
      "step": 22940,
      "train_runtime": 60107.3269,
      "train_tokens_per_second": 12163.959
    },
    {
      "epoch": 1.5449097764610826,
      "grad_norm": 0.7816108151536917,
      "learning_rate": 1.2244020663446731e-06,
      "loss": 0.2986234903335571,
      "num_input_tokens_seen": 731302424,
      "step": 22945,
      "train_runtime": 60120.92,
      "train_tokens_per_second": 12163.86
    },
    {
      "epoch": 1.5452464314570427,
      "grad_norm": 0.9253823645201016,
      "learning_rate": 1.222669168262598e-06,
      "loss": 0.29760174751281737,
      "num_input_tokens_seen": 731465312,
      "step": 22950,
      "train_runtime": 60135.0003,
      "train_tokens_per_second": 12163.72
    },
    {
      "epoch": 1.545583086453003,
      "grad_norm": 0.7108731563129567,
      "learning_rate": 1.2209373264990448e-06,
      "loss": 0.27967774868011475,
      "num_input_tokens_seen": 731619840,
      "step": 22955,
      "train_runtime": 60147.5105,
      "train_tokens_per_second": 12163.759
    },
    {
      "epoch": 1.5459197414489632,
      "grad_norm": 0.866179924726618,
      "learning_rate": 1.2192065415383176e-06,
      "loss": 0.29625301361083983,
      "num_input_tokens_seen": 731780232,
      "step": 22960,
      "train_runtime": 60161.0845,
      "train_tokens_per_second": 12163.681
    },
    {
      "epoch": 1.5462563964449232,
      "grad_norm": 0.7861738486181785,
      "learning_rate": 1.217476813864425e-06,
      "loss": 0.30833067893981936,
      "num_input_tokens_seen": 731942872,
      "step": 22965,
      "train_runtime": 60173.7894,
      "train_tokens_per_second": 12163.815
    },
    {
      "epoch": 1.5465930514408834,
      "grad_norm": 0.7681230832599563,
      "learning_rate": 1.2157481439610801e-06,
      "loss": 0.2545867204666138,
      "num_input_tokens_seen": 732101920,
      "step": 22970,
      "train_runtime": 60187.0759,
      "train_tokens_per_second": 12163.773
    },
    {
      "epoch": 1.5469297064368437,
      "grad_norm": 0.9161972868847953,
      "learning_rate": 1.2140205323116976e-06,
      "loss": 0.30569305419921877,
      "num_input_tokens_seen": 732253920,
      "step": 22975,
      "train_runtime": 60200.428,
      "train_tokens_per_second": 12163.6
    },
    {
      "epoch": 1.5472663614328037,
      "grad_norm": 0.7399102515239029,
      "learning_rate": 1.2122939793994043e-06,
      "loss": 0.2779374599456787,
      "num_input_tokens_seen": 732412808,
      "step": 22980,
      "train_runtime": 60213.7169,
      "train_tokens_per_second": 12163.554
    },
    {
      "epoch": 1.5476030164287637,
      "grad_norm": 0.7216817617474057,
      "learning_rate": 1.2105684857070182e-06,
      "loss": 0.27604074478149415,
      "num_input_tokens_seen": 732575168,
      "step": 22985,
      "train_runtime": 60227.0392,
      "train_tokens_per_second": 12163.559
    },
    {
      "epoch": 1.547939671424724,
      "grad_norm": 0.8228090879319583,
      "learning_rate": 1.2088440517170729e-06,
      "loss": 0.2893498420715332,
      "num_input_tokens_seen": 732735504,
      "step": 22990,
      "train_runtime": 60239.4829,
      "train_tokens_per_second": 12163.708
    },
    {
      "epoch": 1.5482763264206842,
      "grad_norm": 0.8390520563755358,
      "learning_rate": 1.2071206779117994e-06,
      "loss": 0.28883676528930663,
      "num_input_tokens_seen": 732895888,
      "step": 22995,
      "train_runtime": 60252.8071,
      "train_tokens_per_second": 12163.68
    },
    {
      "epoch": 1.5486129814166443,
      "grad_norm": 0.9265247946455487,
      "learning_rate": 1.2053983647731337e-06,
      "loss": 0.3107041597366333,
      "num_input_tokens_seen": 733057520,
      "step": 23000,
      "train_runtime": 60265.3774,
      "train_tokens_per_second": 12163.825
    },
    {
      "epoch": 1.5489496364126043,
      "grad_norm": 0.7464637701076479,
      "learning_rate": 1.2036771127827152e-06,
      "loss": 0.286514139175415,
      "num_input_tokens_seen": 733213032,
      "step": 23005,
      "train_runtime": 60279.2163,
      "train_tokens_per_second": 12163.613
    },
    {
      "epoch": 1.5492862914085646,
      "grad_norm": 0.8158854773008193,
      "learning_rate": 1.2019569224218869e-06,
      "loss": 0.32193081378936766,
      "num_input_tokens_seen": 733376872,
      "step": 23010,
      "train_runtime": 60291.67,
      "train_tokens_per_second": 12163.818
    },
    {
      "epoch": 1.5496229464045248,
      "grad_norm": 0.7151331333285766,
      "learning_rate": 1.2002377941716936e-06,
      "loss": 0.26620028018951414,
      "num_input_tokens_seen": 733533208,
      "step": 23015,
      "train_runtime": 60305.1135,
      "train_tokens_per_second": 12163.698
    },
    {
      "epoch": 1.5499596014004848,
      "grad_norm": 1.0550421241107877,
      "learning_rate": 1.1985197285128853e-06,
      "loss": 0.30360434055328367,
      "num_input_tokens_seen": 733691768,
      "step": 23020,
      "train_runtime": 60318.2124,
      "train_tokens_per_second": 12163.686
    },
    {
      "epoch": 1.5502962563964449,
      "grad_norm": 0.7890582939753223,
      "learning_rate": 1.1968027259259107e-06,
      "loss": 0.2946274757385254,
      "num_input_tokens_seen": 733848368,
      "step": 23025,
      "train_runtime": 60331.8642,
      "train_tokens_per_second": 12163.529
    },
    {
      "epoch": 1.5506329113924051,
      "grad_norm": 0.8214821102297931,
      "learning_rate": 1.1950867868909293e-06,
      "loss": 0.26646084785461427,
      "num_input_tokens_seen": 734005896,
      "step": 23030,
      "train_runtime": 60345.6953,
      "train_tokens_per_second": 12163.351
    },
    {
      "epoch": 1.5509695663883654,
      "grad_norm": 0.8902687085608858,
      "learning_rate": 1.1933719118877923e-06,
      "loss": 0.30422391891479494,
      "num_input_tokens_seen": 734169736,
      "step": 23035,
      "train_runtime": 60358.1404,
      "train_tokens_per_second": 12163.558
    },
    {
      "epoch": 1.5513062213843254,
      "grad_norm": 1.3049478762340514,
      "learning_rate": 1.1916581013960632e-06,
      "loss": 0.25280778408050536,
      "num_input_tokens_seen": 734333576,
      "step": 23040,
      "train_runtime": 60370.5949,
      "train_tokens_per_second": 12163.762
    },
    {
      "epoch": 1.5516428763802854,
      "grad_norm": 0.8673003581992507,
      "learning_rate": 1.189945355895002e-06,
      "loss": 0.3032198667526245,
      "num_input_tokens_seen": 734494392,
      "step": 23045,
      "train_runtime": 60384.1006,
      "train_tokens_per_second": 12163.705
    },
    {
      "epoch": 1.5519795313762457,
      "grad_norm": 0.8543350321352065,
      "learning_rate": 1.1882336758635727e-06,
      "loss": 0.2696817398071289,
      "num_input_tokens_seen": 734658232,
      "step": 23050,
      "train_runtime": 60397.1422,
      "train_tokens_per_second": 12163.791
    },
    {
      "epoch": 1.552316186372206,
      "grad_norm": 0.9633308137903916,
      "learning_rate": 1.1865230617804412e-06,
      "loss": 0.3258500099182129,
      "num_input_tokens_seen": 734817896,
      "step": 23055,
      "train_runtime": 60409.5751,
      "train_tokens_per_second": 12163.931
    },
    {
      "epoch": 1.552652841368166,
      "grad_norm": 0.8221039646728788,
      "learning_rate": 1.1848135141239753e-06,
      "loss": 0.2778597116470337,
      "num_input_tokens_seen": 734977336,
      "step": 23060,
      "train_runtime": 60423.1955,
      "train_tokens_per_second": 12163.828
    },
    {
      "epoch": 1.552989496364126,
      "grad_norm": 0.8328168017044412,
      "learning_rate": 1.1831050333722438e-06,
      "loss": 0.3028880596160889,
      "num_input_tokens_seen": 735132912,
      "step": 23065,
      "train_runtime": 60436.2328,
      "train_tokens_per_second": 12163.778
    },
    {
      "epoch": 1.5533261513600862,
      "grad_norm": 0.7513946366880544,
      "learning_rate": 1.1813976200030191e-06,
      "loss": 0.27589375972747804,
      "num_input_tokens_seen": 735294800,
      "step": 23070,
      "train_runtime": 60450.4025,
      "train_tokens_per_second": 12163.605
    },
    {
      "epoch": 1.5536628063560465,
      "grad_norm": 0.7434952413858029,
      "learning_rate": 1.1796912744937717e-06,
      "loss": 0.2990890026092529,
      "num_input_tokens_seen": 735456896,
      "step": 23075,
      "train_runtime": 60463.6552,
      "train_tokens_per_second": 12163.62
    },
    {
      "epoch": 1.5539994613520065,
      "grad_norm": 0.7934732318439129,
      "learning_rate": 1.1779859973216802e-06,
      "loss": 0.29690263271331785,
      "num_input_tokens_seen": 735615128,
      "step": 23080,
      "train_runtime": 60477.2113,
      "train_tokens_per_second": 12163.509
    },
    {
      "epoch": 1.5543361163479665,
      "grad_norm": 0.8971167542616845,
      "learning_rate": 1.1762817889636142e-06,
      "loss": 0.29803128242492677,
      "num_input_tokens_seen": 735769640,
      "step": 23085,
      "train_runtime": 60491.1857,
      "train_tokens_per_second": 12163.254
    },
    {
      "epoch": 1.5546727713439268,
      "grad_norm": 0.8379450786254626,
      "learning_rate": 1.1745786498961542e-06,
      "loss": 0.3054941654205322,
      "num_input_tokens_seen": 735933480,
      "step": 23090,
      "train_runtime": 60503.9252,
      "train_tokens_per_second": 12163.401
    },
    {
      "epoch": 1.555009426339887,
      "grad_norm": 0.7856850757944269,
      "learning_rate": 1.1728765805955766e-06,
      "loss": 0.2800941467285156,
      "num_input_tokens_seen": 736091104,
      "step": 23095,
      "train_runtime": 60517.7252,
      "train_tokens_per_second": 12163.232
    },
    {
      "epoch": 1.555346081335847,
      "grad_norm": 0.9265194089950859,
      "learning_rate": 1.1711755815378595e-06,
      "loss": 0.3234413623809814,
      "num_input_tokens_seen": 736248368,
      "step": 23100,
      "train_runtime": 60531.3478,
      "train_tokens_per_second": 12163.092
    },
    {
      "epoch": 1.555682736331807,
      "grad_norm": 0.7847198133517144,
      "learning_rate": 1.1694756531986818e-06,
      "loss": 0.2967512130737305,
      "num_input_tokens_seen": 736412208,
      "step": 23105,
      "train_runtime": 60543.7789,
      "train_tokens_per_second": 12163.301
    },
    {
      "epoch": 1.5560193913277673,
      "grad_norm": 0.7875121322793062,
      "learning_rate": 1.1677767960534232e-06,
      "loss": 0.3001002073287964,
      "num_input_tokens_seen": 736568456,
      "step": 23110,
      "train_runtime": 60556.6453,
      "train_tokens_per_second": 12163.297
    },
    {
      "epoch": 1.5563560463237276,
      "grad_norm": 0.7759507921470586,
      "learning_rate": 1.166079010577163e-06,
      "loss": 0.3113714694976807,
      "num_input_tokens_seen": 736728152,
      "step": 23115,
      "train_runtime": 60569.8701,
      "train_tokens_per_second": 12163.278
    },
    {
      "epoch": 1.5566927013196876,
      "grad_norm": 0.8735265527346818,
      "learning_rate": 1.1643822972446823e-06,
      "loss": 0.3008978605270386,
      "num_input_tokens_seen": 736888448,
      "step": 23120,
      "train_runtime": 60582.5995,
      "train_tokens_per_second": 12163.368
    },
    {
      "epoch": 1.5570293563156477,
      "grad_norm": 0.788389552214716,
      "learning_rate": 1.1626866565304594e-06,
      "loss": 0.294887113571167,
      "num_input_tokens_seen": 737051608,
      "step": 23125,
      "train_runtime": 60595.0391,
      "train_tokens_per_second": 12163.564
    },
    {
      "epoch": 1.557366011311608,
      "grad_norm": 0.7713080581521571,
      "learning_rate": 1.16099208890868e-06,
      "loss": 0.288470196723938,
      "num_input_tokens_seen": 737210136,
      "step": 23130,
      "train_runtime": 60608.1437,
      "train_tokens_per_second": 12163.549
    },
    {
      "epoch": 1.5577026663075682,
      "grad_norm": 2.148170817130903,
      "learning_rate": 1.1592985948532187e-06,
      "loss": 0.28723735809326173,
      "num_input_tokens_seen": 737367240,
      "step": 23135,
      "train_runtime": 60621.7966,
      "train_tokens_per_second": 12163.401
    },
    {
      "epoch": 1.5580393213035282,
      "grad_norm": 0.7937948471800973,
      "learning_rate": 1.1576061748376594e-06,
      "loss": 0.2882655620574951,
      "num_input_tokens_seen": 737523096,
      "step": 23140,
      "train_runtime": 60634.3628,
      "train_tokens_per_second": 12163.451
    },
    {
      "epoch": 1.5583759762994882,
      "grad_norm": 0.8047898549966961,
      "learning_rate": 1.1559148293352805e-06,
      "loss": 0.2820580959320068,
      "num_input_tokens_seen": 737682616,
      "step": 23145,
      "train_runtime": 60647.0877,
      "train_tokens_per_second": 12163.529
    },
    {
      "epoch": 1.5587126312954485,
      "grad_norm": 0.7668356841882308,
      "learning_rate": 1.154224558819062e-06,
      "loss": 0.2904337406158447,
      "num_input_tokens_seen": 737842976,
      "step": 23150,
      "train_runtime": 60659.9939,
      "train_tokens_per_second": 12163.585
    },
    {
      "epoch": 1.5590492862914087,
      "grad_norm": 0.811851104373591,
      "learning_rate": 1.1525353637616821e-06,
      "loss": 0.28730196952819825,
      "num_input_tokens_seen": 738001592,
      "step": 23155,
      "train_runtime": 60672.8691,
      "train_tokens_per_second": 12163.618
    },
    {
      "epoch": 1.5593859412873687,
      "grad_norm": 0.858925884628469,
      "learning_rate": 1.1508472446355196e-06,
      "loss": 0.29271821975708007,
      "num_input_tokens_seen": 738165432,
      "step": 23160,
      "train_runtime": 60685.2993,
      "train_tokens_per_second": 12163.826
    },
    {
      "epoch": 1.5597225962833288,
      "grad_norm": 0.8069614990464805,
      "learning_rate": 1.1491602019126501e-06,
      "loss": 0.3047060966491699,
      "num_input_tokens_seen": 738329128,
      "step": 23165,
      "train_runtime": 60698.3298,
      "train_tokens_per_second": 12163.912
    },
    {
      "epoch": 1.560059251279289,
      "grad_norm": 0.8502361412677573,
      "learning_rate": 1.1474742360648515e-06,
      "loss": 0.29802546501159666,
      "num_input_tokens_seen": 738489416,
      "step": 23170,
      "train_runtime": 60710.7583,
      "train_tokens_per_second": 12164.062
    },
    {
      "epoch": 1.5603959062752493,
      "grad_norm": 0.8611928572387201,
      "learning_rate": 1.1457893475635968e-06,
      "loss": 0.3120441913604736,
      "num_input_tokens_seen": 738645984,
      "step": 23175,
      "train_runtime": 60723.8576,
      "train_tokens_per_second": 12164.016
    },
    {
      "epoch": 1.5607325612712093,
      "grad_norm": 0.8554315750729933,
      "learning_rate": 1.144105536880063e-06,
      "loss": 0.2776580810546875,
      "num_input_tokens_seen": 738804832,
      "step": 23180,
      "train_runtime": 60737.5554,
      "train_tokens_per_second": 12163.888
    },
    {
      "epoch": 1.5610692162671693,
      "grad_norm": 0.8328715622881653,
      "learning_rate": 1.1424228044851176e-06,
      "loss": 0.2846327304840088,
      "num_input_tokens_seen": 738961256,
      "step": 23185,
      "train_runtime": 60751.8902,
      "train_tokens_per_second": 12163.593
    },
    {
      "epoch": 1.5614058712631296,
      "grad_norm": 0.7237227234164573,
      "learning_rate": 1.1407411508493355e-06,
      "loss": 0.28693265914916993,
      "num_input_tokens_seen": 739118856,
      "step": 23190,
      "train_runtime": 60764.9864,
      "train_tokens_per_second": 12163.565
    },
    {
      "epoch": 1.5617425262590898,
      "grad_norm": 0.7845524124349043,
      "learning_rate": 1.1390605764429846e-06,
      "loss": 0.2674920082092285,
      "num_input_tokens_seen": 739278488,
      "step": 23195,
      "train_runtime": 60777.8153,
      "train_tokens_per_second": 12163.624
    },
    {
      "epoch": 1.5620791812550499,
      "grad_norm": 0.8929289326485473,
      "learning_rate": 1.1373810817360314e-06,
      "loss": 0.2981806039810181,
      "num_input_tokens_seen": 739441800,
      "step": 23200,
      "train_runtime": 60790.7239,
      "train_tokens_per_second": 12163.727
    },
    {
      "epoch": 1.5624158362510099,
      "grad_norm": 0.9044612113654503,
      "learning_rate": 1.135702667198142e-06,
      "loss": 0.3152334690093994,
      "num_input_tokens_seen": 739603880,
      "step": 23205,
      "train_runtime": 60804.0461,
      "train_tokens_per_second": 12163.728
    },
    {
      "epoch": 1.5627524912469701,
      "grad_norm": 0.8378447701082983,
      "learning_rate": 1.134025333298679e-06,
      "loss": 0.26320223808288573,
      "num_input_tokens_seen": 739757192,
      "step": 23210,
      "train_runtime": 60816.8962,
      "train_tokens_per_second": 12163.679
    },
    {
      "epoch": 1.5630891462429304,
      "grad_norm": 0.8544615120702004,
      "learning_rate": 1.1323490805067045e-06,
      "loss": 0.2773989200592041,
      "num_input_tokens_seen": 739916776,
      "step": 23215,
      "train_runtime": 60830.2675,
      "train_tokens_per_second": 12163.629
    },
    {
      "epoch": 1.5634258012388904,
      "grad_norm": 0.8068933586041359,
      "learning_rate": 1.1306739092909757e-06,
      "loss": 0.2633871793746948,
      "num_input_tokens_seen": 740073848,
      "step": 23220,
      "train_runtime": 60843.3804,
      "train_tokens_per_second": 12163.589
    },
    {
      "epoch": 1.5637624562348504,
      "grad_norm": 0.7900715290821706,
      "learning_rate": 1.1289998201199493e-06,
      "loss": 0.27851572036743166,
      "num_input_tokens_seen": 740233400,
      "step": 23225,
      "train_runtime": 60856.757,
      "train_tokens_per_second": 12163.537
    },
    {
      "epoch": 1.5640991112308107,
      "grad_norm": 0.806297674238551,
      "learning_rate": 1.1273268134617816e-06,
      "loss": 0.3013458251953125,
      "num_input_tokens_seen": 740395416,
      "step": 23230,
      "train_runtime": 60869.3455,
      "train_tokens_per_second": 12163.683
    },
    {
      "epoch": 1.564435766226771,
      "grad_norm": 0.7196484840293191,
      "learning_rate": 1.1256548897843189e-06,
      "loss": 0.3004997491836548,
      "num_input_tokens_seen": 740558624,
      "step": 23235,
      "train_runtime": 60882.3874,
      "train_tokens_per_second": 12163.758
    },
    {
      "epoch": 1.564772421222731,
      "grad_norm": 0.7640318488229205,
      "learning_rate": 1.1239840495551136e-06,
      "loss": 0.2844345808029175,
      "num_input_tokens_seen": 740721184,
      "step": 23240,
      "train_runtime": 60894.9931,
      "train_tokens_per_second": 12163.909
    },
    {
      "epoch": 1.565109076218691,
      "grad_norm": 0.8648528222788187,
      "learning_rate": 1.1223142932414087e-06,
      "loss": 0.32747530937194824,
      "num_input_tokens_seen": 740885024,
      "step": 23245,
      "train_runtime": 60907.4387,
      "train_tokens_per_second": 12164.114
    },
    {
      "epoch": 1.5654457312146512,
      "grad_norm": 0.7210810976826967,
      "learning_rate": 1.120645621310147e-06,
      "loss": 0.29173855781555175,
      "num_input_tokens_seen": 741045944,
      "step": 23250,
      "train_runtime": 60919.9115,
      "train_tokens_per_second": 12164.265
    },
    {
      "epoch": 1.5657823862106115,
      "grad_norm": 0.9652057816898654,
      "learning_rate": 1.1189780342279666e-06,
      "loss": 0.2980844259262085,
      "num_input_tokens_seen": 741197304,
      "step": 23255,
      "train_runtime": 60932.5106,
      "train_tokens_per_second": 12164.234
    },
    {
      "epoch": 1.5661190412065715,
      "grad_norm": 0.7632355133382442,
      "learning_rate": 1.1173115324612033e-06,
      "loss": 0.3174387693405151,
      "num_input_tokens_seen": 741352024,
      "step": 23260,
      "train_runtime": 60945.6082,
      "train_tokens_per_second": 12164.158
    },
    {
      "epoch": 1.5664556962025316,
      "grad_norm": 0.8524277751319136,
      "learning_rate": 1.1156461164758891e-06,
      "loss": 0.3171722412109375,
      "num_input_tokens_seen": 741515864,
      "step": 23265,
      "train_runtime": 60958.0332,
      "train_tokens_per_second": 12164.367
    },
    {
      "epoch": 1.5667923511984918,
      "grad_norm": 0.7388589419155768,
      "learning_rate": 1.1139817867377522e-06,
      "loss": 0.24607739448547364,
      "num_input_tokens_seen": 741678272,
      "step": 23270,
      "train_runtime": 60970.6731,
      "train_tokens_per_second": 12164.509
    },
    {
      "epoch": 1.567129006194452,
      "grad_norm": 0.7999273112283958,
      "learning_rate": 1.1123185437122153e-06,
      "loss": 0.2833959341049194,
      "num_input_tokens_seen": 741842112,
      "step": 23275,
      "train_runtime": 60983.111,
      "train_tokens_per_second": 12164.714
    },
    {
      "epoch": 1.567465661190412,
      "grad_norm": 0.8532553354759879,
      "learning_rate": 1.1106563878644038e-06,
      "loss": 0.28699569702148436,
      "num_input_tokens_seen": 742002120,
      "step": 23280,
      "train_runtime": 60995.6004,
      "train_tokens_per_second": 12164.847
    },
    {
      "epoch": 1.5678023161863721,
      "grad_norm": 0.7991473194536528,
      "learning_rate": 1.1089953196591286e-06,
      "loss": 0.2935714483261108,
      "num_input_tokens_seen": 742162880,
      "step": 23285,
      "train_runtime": 61009.1363,
      "train_tokens_per_second": 12164.783
    },
    {
      "epoch": 1.5681389711823324,
      "grad_norm": 0.8974969088409925,
      "learning_rate": 1.107335339560906e-06,
      "loss": 0.30866389274597167,
      "num_input_tokens_seen": 742318360,
      "step": 23290,
      "train_runtime": 61022.8719,
      "train_tokens_per_second": 12164.592
    },
    {
      "epoch": 1.5684756261782926,
      "grad_norm": 0.8467021168415199,
      "learning_rate": 1.1056764480339427e-06,
      "loss": 0.294776177406311,
      "num_input_tokens_seen": 742482040,
      "step": 23295,
      "train_runtime": 61035.9754,
      "train_tokens_per_second": 12164.662
    },
    {
      "epoch": 1.5688122811742526,
      "grad_norm": 0.8499248608204794,
      "learning_rate": 1.1040186455421425e-06,
      "loss": 0.32110395431518557,
      "num_input_tokens_seen": 742644992,
      "step": 23300,
      "train_runtime": 61049.4657,
      "train_tokens_per_second": 12164.644
    },
    {
      "epoch": 1.5691489361702127,
      "grad_norm": 0.8372293846127322,
      "learning_rate": 1.1023619325491052e-06,
      "loss": 0.2966631889343262,
      "num_input_tokens_seen": 742804256,
      "step": 23305,
      "train_runtime": 61062.4203,
      "train_tokens_per_second": 12164.671
    },
    {
      "epoch": 1.569485591166173,
      "grad_norm": 0.685148142297708,
      "learning_rate": 1.1007063095181248e-06,
      "loss": 0.2775959730148315,
      "num_input_tokens_seen": 742963336,
      "step": 23310,
      "train_runtime": 61075.9682,
      "train_tokens_per_second": 12164.577
    },
    {
      "epoch": 1.5698222461621332,
      "grad_norm": 0.822782838837817,
      "learning_rate": 1.0990517769121905e-06,
      "loss": 0.3007147789001465,
      "num_input_tokens_seen": 743115912,
      "step": 23315,
      "train_runtime": 61089.1205,
      "train_tokens_per_second": 12164.456
    },
    {
      "epoch": 1.5701589011580932,
      "grad_norm": 0.9049653976399583,
      "learning_rate": 1.0973983351939876e-06,
      "loss": 0.30109505653381347,
      "num_input_tokens_seen": 743276432,
      "step": 23320,
      "train_runtime": 61102.483,
      "train_tokens_per_second": 12164.423
    },
    {
      "epoch": 1.5704955561540532,
      "grad_norm": 0.868828467049675,
      "learning_rate": 1.095745984825895e-06,
      "loss": 0.2956847667694092,
      "num_input_tokens_seen": 743435680,
      "step": 23325,
      "train_runtime": 61114.9034,
      "train_tokens_per_second": 12164.556
    },
    {
      "epoch": 1.5708322111500135,
      "grad_norm": 0.8686982955432532,
      "learning_rate": 1.0940947262699902e-06,
      "loss": 0.29397897720336913,
      "num_input_tokens_seen": 743591080,
      "step": 23330,
      "train_runtime": 61127.7418,
      "train_tokens_per_second": 12164.544
    },
    {
      "epoch": 1.5711688661459737,
      "grad_norm": 0.7251408874691271,
      "learning_rate": 1.092444559988038e-06,
      "loss": 0.28177633285522463,
      "num_input_tokens_seen": 743751736,
      "step": 23335,
      "train_runtime": 61140.9215,
      "train_tokens_per_second": 12164.549
    },
    {
      "epoch": 1.5715055211419338,
      "grad_norm": 0.7704535936860318,
      "learning_rate": 1.0907954864415048e-06,
      "loss": 0.293503999710083,
      "num_input_tokens_seen": 743913584,
      "step": 23340,
      "train_runtime": 61153.4451,
      "train_tokens_per_second": 12164.704
    },
    {
      "epoch": 1.5718421761378938,
      "grad_norm": 0.871551036206662,
      "learning_rate": 1.089147506091549e-06,
      "loss": 0.2725985527038574,
      "num_input_tokens_seen": 744072416,
      "step": 23345,
      "train_runtime": 61166.4124,
      "train_tokens_per_second": 12164.722
    },
    {
      "epoch": 1.572178831133854,
      "grad_norm": 0.9478749742803037,
      "learning_rate": 1.0875006193990217e-06,
      "loss": 0.3075448989868164,
      "num_input_tokens_seen": 744228888,
      "step": 23350,
      "train_runtime": 61179.8297,
      "train_tokens_per_second": 12164.612
    },
    {
      "epoch": 1.5725154861298143,
      "grad_norm": 0.8378236147152379,
      "learning_rate": 1.0858548268244706e-06,
      "loss": 0.28488545417785643,
      "num_input_tokens_seen": 744387760,
      "step": 23355,
      "train_runtime": 61192.4529,
      "train_tokens_per_second": 12164.699
    },
    {
      "epoch": 1.5728521411257743,
      "grad_norm": 0.8097247759842988,
      "learning_rate": 1.0842101288281359e-06,
      "loss": 0.288008975982666,
      "num_input_tokens_seen": 744548624,
      "step": 23360,
      "train_runtime": 61205.8871,
      "train_tokens_per_second": 12164.657
    },
    {
      "epoch": 1.5731887961217343,
      "grad_norm": 0.8129984259831883,
      "learning_rate": 1.082566525869952e-06,
      "loss": 0.2869732856750488,
      "num_input_tokens_seen": 744712352,
      "step": 23365,
      "train_runtime": 61218.9788,
      "train_tokens_per_second": 12164.73
    },
    {
      "epoch": 1.5735254511176946,
      "grad_norm": 0.8591360978323196,
      "learning_rate": 1.0809240184095476e-06,
      "loss": 0.31190996170043944,
      "num_input_tokens_seen": 744870784,
      "step": 23370,
      "train_runtime": 61231.8139,
      "train_tokens_per_second": 12164.768
    },
    {
      "epoch": 1.5738621061136548,
      "grad_norm": 0.7730497939452526,
      "learning_rate": 1.0792826069062429e-06,
      "loss": 0.27433960437774657,
      "num_input_tokens_seen": 745033336,
      "step": 23375,
      "train_runtime": 61244.7198,
      "train_tokens_per_second": 12164.858
    },
    {
      "epoch": 1.5741987611096149,
      "grad_norm": 0.7040122224160477,
      "learning_rate": 1.0776422918190576e-06,
      "loss": 0.30633959770202634,
      "num_input_tokens_seen": 745191712,
      "step": 23380,
      "train_runtime": 61257.2034,
      "train_tokens_per_second": 12164.965
    },
    {
      "epoch": 1.574535416105575,
      "grad_norm": 0.9112582956707095,
      "learning_rate": 1.0760030736066952e-06,
      "loss": 0.28393733501434326,
      "num_input_tokens_seen": 745346800,
      "step": 23385,
      "train_runtime": 61270.116,
      "train_tokens_per_second": 12164.932
    },
    {
      "epoch": 1.5748720711015352,
      "grad_norm": 0.7738243422361231,
      "learning_rate": 1.0743649527275619e-06,
      "loss": 0.28804926872253417,
      "num_input_tokens_seen": 745508296,
      "step": 23390,
      "train_runtime": 61283.2168,
      "train_tokens_per_second": 12164.967
    },
    {
      "epoch": 1.5752087260974954,
      "grad_norm": 0.7990054611874563,
      "learning_rate": 1.0727279296397514e-06,
      "loss": 0.29540624618530276,
      "num_input_tokens_seen": 745671592,
      "step": 23395,
      "train_runtime": 61296.2886,
      "train_tokens_per_second": 12165.037
    },
    {
      "epoch": 1.5755453810934554,
      "grad_norm": 0.9084433459683746,
      "learning_rate": 1.0710920048010526e-06,
      "loss": 0.31709871292114256,
      "num_input_tokens_seen": 745828784,
      "step": 23400,
      "train_runtime": 61308.728,
      "train_tokens_per_second": 12165.132
    },
    {
      "epoch": 1.5758820360894155,
      "grad_norm": 0.8420521370700118,
      "learning_rate": 1.0694571786689462e-06,
      "loss": 0.3073237657546997,
      "num_input_tokens_seen": 745992624,
      "step": 23405,
      "train_runtime": 61321.1557,
      "train_tokens_per_second": 12165.339
    },
    {
      "epoch": 1.5762186910853757,
      "grad_norm": 0.8925934933824081,
      "learning_rate": 1.067823451700606e-06,
      "loss": 0.28224446773529055,
      "num_input_tokens_seen": 746146096,
      "step": 23410,
      "train_runtime": 61333.6304,
      "train_tokens_per_second": 12165.367
    },
    {
      "epoch": 1.576555346081336,
      "grad_norm": 0.8935956607390586,
      "learning_rate": 1.066190824352899e-06,
      "loss": 0.2907134532928467,
      "num_input_tokens_seen": 746307544,
      "step": 23415,
      "train_runtime": 61346.7943,
      "train_tokens_per_second": 12165.388
    },
    {
      "epoch": 1.576892001077296,
      "grad_norm": 0.7808873564423955,
      "learning_rate": 1.0645592970823837e-06,
      "loss": 0.26576077938079834,
      "num_input_tokens_seen": 746463336,
      "step": 23420,
      "train_runtime": 61359.7407,
      "train_tokens_per_second": 12165.36
    },
    {
      "epoch": 1.577228656073256,
      "grad_norm": 0.771891208129401,
      "learning_rate": 1.0629288703453105e-06,
      "loss": 0.3038355350494385,
      "num_input_tokens_seen": 746625544,
      "step": 23425,
      "train_runtime": 61372.3443,
      "train_tokens_per_second": 12165.505
    },
    {
      "epoch": 1.5775653110692163,
      "grad_norm": 0.8560705484247997,
      "learning_rate": 1.061299544597627e-06,
      "loss": 0.29100608825683594,
      "num_input_tokens_seen": 746776008,
      "step": 23430,
      "train_runtime": 61385.1363,
      "train_tokens_per_second": 12165.421
    },
    {
      "epoch": 1.5779019660651765,
      "grad_norm": 0.7663357578575652,
      "learning_rate": 1.0596713202949637e-06,
      "loss": 0.297916579246521,
      "num_input_tokens_seen": 746938864,
      "step": 23435,
      "train_runtime": 61397.5995,
      "train_tokens_per_second": 12165.604
    },
    {
      "epoch": 1.5782386210611365,
      "grad_norm": 0.7797505610957637,
      "learning_rate": 1.0580441978926527e-06,
      "loss": 0.29828839302062987,
      "num_input_tokens_seen": 747098632,
      "step": 23440,
      "train_runtime": 61410.8318,
      "train_tokens_per_second": 12165.584
    },
    {
      "epoch": 1.5785752760570966,
      "grad_norm": 0.9435630949370322,
      "learning_rate": 1.056418177845711e-06,
      "loss": 0.2888603925704956,
      "num_input_tokens_seen": 747250464,
      "step": 23445,
      "train_runtime": 61423.716,
      "train_tokens_per_second": 12165.504
    },
    {
      "epoch": 1.5789119310530568,
      "grad_norm": 0.7393657273691062,
      "learning_rate": 1.0547932606088506e-06,
      "loss": 0.2955147743225098,
      "num_input_tokens_seen": 747409064,
      "step": 23450,
      "train_runtime": 61437.1285,
      "train_tokens_per_second": 12165.43
    },
    {
      "epoch": 1.579248586049017,
      "grad_norm": 0.8026188231971175,
      "learning_rate": 1.053169446636475e-06,
      "loss": 0.29681265354156494,
      "num_input_tokens_seen": 747571032,
      "step": 23455,
      "train_runtime": 61449.7081,
      "train_tokens_per_second": 12165.575
    },
    {
      "epoch": 1.579585241044977,
      "grad_norm": 0.9838278644291032,
      "learning_rate": 1.0515467363826782e-06,
      "loss": 0.299529504776001,
      "num_input_tokens_seen": 747731048,
      "step": 23460,
      "train_runtime": 61462.9109,
      "train_tokens_per_second": 12165.565
    },
    {
      "epoch": 1.5799218960409371,
      "grad_norm": 0.8347350106674356,
      "learning_rate": 1.0499251303012454e-06,
      "loss": 0.2828847646713257,
      "num_input_tokens_seen": 747893688,
      "step": 23465,
      "train_runtime": 61475.5948,
      "train_tokens_per_second": 12165.701
    },
    {
      "epoch": 1.5802585510368974,
      "grad_norm": 0.8744689452614091,
      "learning_rate": 1.0483046288456544e-06,
      "loss": 0.30875506401062014,
      "num_input_tokens_seen": 748046968,
      "step": 23470,
      "train_runtime": 61488.8129,
      "train_tokens_per_second": 12165.578
    },
    {
      "epoch": 1.5805952060328576,
      "grad_norm": 0.7978321641396979,
      "learning_rate": 1.0466852324690719e-06,
      "loss": 0.28389878273010255,
      "num_input_tokens_seen": 748205080,
      "step": 23475,
      "train_runtime": 61502.4089,
      "train_tokens_per_second": 12165.46
    },
    {
      "epoch": 1.5809318610288177,
      "grad_norm": 0.7982304213551433,
      "learning_rate": 1.0450669416243596e-06,
      "loss": 0.274608850479126,
      "num_input_tokens_seen": 748368520,
      "step": 23480,
      "train_runtime": 61516.0892,
      "train_tokens_per_second": 12165.411
    },
    {
      "epoch": 1.5812685160247777,
      "grad_norm": 0.7634355853805972,
      "learning_rate": 1.0434497567640639e-06,
      "loss": 0.29886023998260497,
      "num_input_tokens_seen": 748529888,
      "step": 23485,
      "train_runtime": 61529.1537,
      "train_tokens_per_second": 12165.451
    },
    {
      "epoch": 1.581605171020738,
      "grad_norm": 0.8666736886582709,
      "learning_rate": 1.0418336783404282e-06,
      "loss": 0.2986772537231445,
      "num_input_tokens_seen": 748687032,
      "step": 23490,
      "train_runtime": 61541.6751,
      "train_tokens_per_second": 12165.529
    },
    {
      "epoch": 1.5819418260166982,
      "grad_norm": 0.8037655437026634,
      "learning_rate": 1.0402187068053825e-06,
      "loss": 0.29055209159851075,
      "num_input_tokens_seen": 748844416,
      "step": 23495,
      "train_runtime": 61554.6716,
      "train_tokens_per_second": 12165.517
    },
    {
      "epoch": 1.5822784810126582,
      "grad_norm": 0.8342463481999536,
      "learning_rate": 1.0386048426105495e-06,
      "loss": 0.2872841119766235,
      "num_input_tokens_seen": 749002496,
      "step": 23500,
      "train_runtime": 61568.4343,
      "train_tokens_per_second": 12165.365
    },
    {
      "epoch": 1.5826151360086183,
      "grad_norm": 0.7872311353118867,
      "learning_rate": 1.0369920862072396e-06,
      "loss": 0.2815727710723877,
      "num_input_tokens_seen": 749156000,
      "step": 23505,
      "train_runtime": 61581.0605,
      "train_tokens_per_second": 12165.364
    },
    {
      "epoch": 1.5829517910045785,
      "grad_norm": 1.0684142289656602,
      "learning_rate": 1.0353804380464556e-06,
      "loss": 0.2715763092041016,
      "num_input_tokens_seen": 749317328,
      "step": 23510,
      "train_runtime": 61593.5583,
      "train_tokens_per_second": 12165.515
    },
    {
      "epoch": 1.5832884460005388,
      "grad_norm": 0.7640166452131792,
      "learning_rate": 1.0337698985788903e-06,
      "loss": 0.27533583641052245,
      "num_input_tokens_seen": 749480144,
      "step": 23515,
      "train_runtime": 61606.2338,
      "train_tokens_per_second": 12165.654
    },
    {
      "epoch": 1.5836251009964988,
      "grad_norm": 0.8178932490682378,
      "learning_rate": 1.0321604682549247e-06,
      "loss": 0.28578457832336424,
      "num_input_tokens_seen": 749643920,
      "step": 23520,
      "train_runtime": 61618.7191,
      "train_tokens_per_second": 12165.847
    },
    {
      "epoch": 1.5839617559924588,
      "grad_norm": 0.8784746701406516,
      "learning_rate": 1.0305521475246311e-06,
      "loss": 0.31900773048400877,
      "num_input_tokens_seen": 749797952,
      "step": 23525,
      "train_runtime": 61632.3772,
      "train_tokens_per_second": 12165.65
    },
    {
      "epoch": 1.584298410988419,
      "grad_norm": 0.8464043719963891,
      "learning_rate": 1.028944936837774e-06,
      "loss": 0.3088363170623779,
      "num_input_tokens_seen": 749952040,
      "step": 23530,
      "train_runtime": 61645.8559,
      "train_tokens_per_second": 12165.49
    },
    {
      "epoch": 1.5846350659843793,
      "grad_norm": 0.7911167392502553,
      "learning_rate": 1.0273388366438003e-06,
      "loss": 0.30842061042785646,
      "num_input_tokens_seen": 750114648,
      "step": 23535,
      "train_runtime": 61658.3251,
      "train_tokens_per_second": 12165.667
    },
    {
      "epoch": 1.5849717209803393,
      "grad_norm": 0.7578873092135829,
      "learning_rate": 1.0257338473918538e-06,
      "loss": 0.288736629486084,
      "num_input_tokens_seen": 750273424,
      "step": 23540,
      "train_runtime": 61670.7707,
      "train_tokens_per_second": 12165.786
    },
    {
      "epoch": 1.5853083759762994,
      "grad_norm": 0.7653009302192686,
      "learning_rate": 1.0241299695307644e-06,
      "loss": 0.2909852981567383,
      "num_input_tokens_seen": 750428872,
      "step": 23545,
      "train_runtime": 61683.9163,
      "train_tokens_per_second": 12165.714
    },
    {
      "epoch": 1.5856450309722596,
      "grad_norm": 0.7577211746878623,
      "learning_rate": 1.0225272035090506e-06,
      "loss": 0.28070902824401855,
      "num_input_tokens_seen": 750592712,
      "step": 23550,
      "train_runtime": 61696.3456,
      "train_tokens_per_second": 12165.918
    },
    {
      "epoch": 1.5859816859682199,
      "grad_norm": 0.8282012116532642,
      "learning_rate": 1.0209255497749209e-06,
      "loss": 0.30391921997070315,
      "num_input_tokens_seen": 750754904,
      "step": 23555,
      "train_runtime": 61708.9452,
      "train_tokens_per_second": 12166.063
    },
    {
      "epoch": 1.58631834096418,
      "grad_norm": 0.7903227302579233,
      "learning_rate": 1.0193250087762735e-06,
      "loss": 0.2886161804199219,
      "num_input_tokens_seen": 750914008,
      "step": 23560,
      "train_runtime": 61721.5417,
      "train_tokens_per_second": 12166.158
    },
    {
      "epoch": 1.58665499596014,
      "grad_norm": 0.6942488136841164,
      "learning_rate": 1.0177255809606934e-06,
      "loss": 0.29778280258178713,
      "num_input_tokens_seen": 751074368,
      "step": 23565,
      "train_runtime": 61735.106,
      "train_tokens_per_second": 12166.082
    },
    {
      "epoch": 1.5869916509561002,
      "grad_norm": 0.8980104470822609,
      "learning_rate": 1.0161272667754562e-06,
      "loss": 0.2906529903411865,
      "num_input_tokens_seen": 751237408,
      "step": 23570,
      "train_runtime": 61747.9652,
      "train_tokens_per_second": 12166.189
    },
    {
      "epoch": 1.5873283059520604,
      "grad_norm": 0.6902435429125515,
      "learning_rate": 1.014530066667524e-06,
      "loss": 0.2831205606460571,
      "num_input_tokens_seen": 751398784,
      "step": 23575,
      "train_runtime": 61760.4285,
      "train_tokens_per_second": 12166.347
    },
    {
      "epoch": 1.5876649609480205,
      "grad_norm": 0.7406565618833525,
      "learning_rate": 1.0129339810835526e-06,
      "loss": 0.2706297397613525,
      "num_input_tokens_seen": 751558640,
      "step": 23580,
      "train_runtime": 61773.6876,
      "train_tokens_per_second": 12166.323
    },
    {
      "epoch": 1.5880016159439805,
      "grad_norm": 0.8686686545311294,
      "learning_rate": 1.0113390104698767e-06,
      "loss": 0.27473959922790525,
      "num_input_tokens_seen": 751721840,
      "step": 23585,
      "train_runtime": 61786.5346,
      "train_tokens_per_second": 12166.435
    },
    {
      "epoch": 1.5883382709399407,
      "grad_norm": 0.7592500243821104,
      "learning_rate": 1.009745155272529e-06,
      "loss": 0.30735254287719727,
      "num_input_tokens_seen": 751884608,
      "step": 23590,
      "train_runtime": 61799.3825,
      "train_tokens_per_second": 12166.539
    },
    {
      "epoch": 1.588674925935901,
      "grad_norm": 0.7219638923153485,
      "learning_rate": 1.0081524159372246e-06,
      "loss": 0.29011518955230714,
      "num_input_tokens_seen": 752041320,
      "step": 23595,
      "train_runtime": 61813.5056,
      "train_tokens_per_second": 12166.295
    },
    {
      "epoch": 1.589011580931861,
      "grad_norm": 0.7952963497852942,
      "learning_rate": 1.0065607929093685e-06,
      "loss": 0.29025843143463137,
      "num_input_tokens_seen": 752200048,
      "step": 23600,
      "train_runtime": 61827.1031,
      "train_tokens_per_second": 12166.186
    },
    {
      "epoch": 1.589348235927821,
      "grad_norm": 0.8068418815456807,
      "learning_rate": 1.0049702866340521e-06,
      "loss": 0.2741154909133911,
      "num_input_tokens_seen": 752361456,
      "step": 23605,
      "train_runtime": 61840.2077,
      "train_tokens_per_second": 12166.218
    },
    {
      "epoch": 1.5896848909237813,
      "grad_norm": 0.7609310286232867,
      "learning_rate": 1.0033808975560556e-06,
      "loss": 0.26850142478942873,
      "num_input_tokens_seen": 752520640,
      "step": 23610,
      "train_runtime": 61853.8229,
      "train_tokens_per_second": 12166.114
    },
    {
      "epoch": 1.5900215459197415,
      "grad_norm": 0.8454710220906957,
      "learning_rate": 1.0017926261198473e-06,
      "loss": 0.29633140563964844,
      "num_input_tokens_seen": 752682600,
      "step": 23615,
      "train_runtime": 61866.977,
      "train_tokens_per_second": 12166.145
    },
    {
      "epoch": 1.5903582009157016,
      "grad_norm": 0.9184516817932522,
      "learning_rate": 1.0002054727695814e-06,
      "loss": 0.2838845491409302,
      "num_input_tokens_seen": 752839624,
      "step": 23620,
      "train_runtime": 61880.9459,
      "train_tokens_per_second": 12165.936
    },
    {
      "epoch": 1.5906948559116616,
      "grad_norm": 0.7817624173455998,
      "learning_rate": 9.986194379490993e-07,
      "loss": 0.297656774520874,
      "num_input_tokens_seen": 752996384,
      "step": 23625,
      "train_runtime": 61894.095,
      "train_tokens_per_second": 12165.884
    },
    {
      "epoch": 1.5910315109076218,
      "grad_norm": 0.8458505707728549,
      "learning_rate": 9.970345221019345e-07,
      "loss": 0.3040682315826416,
      "num_input_tokens_seen": 753151104,
      "step": 23630,
      "train_runtime": 61907.4641,
      "train_tokens_per_second": 12165.756
    },
    {
      "epoch": 1.591368165903582,
      "grad_norm": 0.9984066302606829,
      "learning_rate": 9.954507256712987e-07,
      "loss": 0.3052212715148926,
      "num_input_tokens_seen": 753309592,
      "step": 23635,
      "train_runtime": 61920.2593,
      "train_tokens_per_second": 12165.802
    },
    {
      "epoch": 1.5917048208995421,
      "grad_norm": 0.8079082440429876,
      "learning_rate": 9.938680491000991e-07,
      "loss": 0.2925656080245972,
      "num_input_tokens_seen": 753468032,
      "step": 23640,
      "train_runtime": 61933.1553,
      "train_tokens_per_second": 12165.827
    },
    {
      "epoch": 1.5920414758955022,
      "grad_norm": 0.7677942364658136,
      "learning_rate": 9.92286492830925e-07,
      "loss": 0.29314634799957273,
      "num_input_tokens_seen": 753630200,
      "step": 23645,
      "train_runtime": 61945.7809,
      "train_tokens_per_second": 12165.965
    },
    {
      "epoch": 1.5923781308914624,
      "grad_norm": 0.7501256049068767,
      "learning_rate": 9.907060573060535e-07,
      "loss": 0.30775933265686034,
      "num_input_tokens_seen": 753788936,
      "step": 23650,
      "train_runtime": 61958.3679,
      "train_tokens_per_second": 12166.055
    },
    {
      "epoch": 1.5927147858874227,
      "grad_norm": 0.7757216044188825,
      "learning_rate": 9.891267429674484e-07,
      "loss": 0.29603424072265627,
      "num_input_tokens_seen": 753947720,
      "step": 23655,
      "train_runtime": 61973.4905,
      "train_tokens_per_second": 12165.649
    },
    {
      "epoch": 1.5930514408833827,
      "grad_norm": 0.9124889965011846,
      "learning_rate": 9.8754855025676e-07,
      "loss": 0.2801192283630371,
      "num_input_tokens_seen": 754105888,
      "step": 23660,
      "train_runtime": 61986.6885,
      "train_tokens_per_second": 12165.61
    },
    {
      "epoch": 1.5933880958793427,
      "grad_norm": 0.8743539632201028,
      "learning_rate": 9.85971479615324e-07,
      "loss": 0.2901712656021118,
      "num_input_tokens_seen": 754267984,
      "step": 23665,
      "train_runtime": 61999.3659,
      "train_tokens_per_second": 12165.737
    },
    {
      "epoch": 1.593724750875303,
      "grad_norm": 0.7240708736595821,
      "learning_rate": 9.843955314841647e-07,
      "loss": 0.27735702991485595,
      "num_input_tokens_seen": 754427992,
      "step": 23670,
      "train_runtime": 62012.6497,
      "train_tokens_per_second": 12165.711
    },
    {
      "epoch": 1.5940614058712632,
      "grad_norm": 0.7956634194670543,
      "learning_rate": 9.82820706303989e-07,
      "loss": 0.28912575244903566,
      "num_input_tokens_seen": 754587704,
      "step": 23675,
      "train_runtime": 62025.7276,
      "train_tokens_per_second": 12165.721
    },
    {
      "epoch": 1.5943980608672232,
      "grad_norm": 0.7900671104903814,
      "learning_rate": 9.812470045151952e-07,
      "loss": 0.2891338586807251,
      "num_input_tokens_seen": 754745984,
      "step": 23680,
      "train_runtime": 62040.787,
      "train_tokens_per_second": 12165.319
    },
    {
      "epoch": 1.5947347158631833,
      "grad_norm": 0.8363698447560103,
      "learning_rate": 9.79674426557859e-07,
      "loss": 0.29004683494567873,
      "num_input_tokens_seen": 754901792,
      "step": 23685,
      "train_runtime": 62053.8159,
      "train_tokens_per_second": 12165.276
    },
    {
      "epoch": 1.5950713708591435,
      "grad_norm": 0.699101916970364,
      "learning_rate": 9.781029728717513e-07,
      "loss": 0.2821507453918457,
      "num_input_tokens_seen": 755061848,
      "step": 23690,
      "train_runtime": 62066.5316,
      "train_tokens_per_second": 12165.362
    },
    {
      "epoch": 1.5954080258551038,
      "grad_norm": 0.7641796399889572,
      "learning_rate": 9.765326438963218e-07,
      "loss": 0.30041255950927737,
      "num_input_tokens_seen": 755222400,
      "step": 23695,
      "train_runtime": 62079.7611,
      "train_tokens_per_second": 12165.356
    },
    {
      "epoch": 1.5957446808510638,
      "grad_norm": 0.7744527111588857,
      "learning_rate": 9.749634400707087e-07,
      "loss": 0.27595863342285154,
      "num_input_tokens_seen": 755380792,
      "step": 23700,
      "train_runtime": 62093.3695,
      "train_tokens_per_second": 12165.241
    },
    {
      "epoch": 1.5960813358470238,
      "grad_norm": 0.734421638172271,
      "learning_rate": 9.733953618337344e-07,
      "loss": 0.293001651763916,
      "num_input_tokens_seen": 755538408,
      "step": 23705,
      "train_runtime": 62106.6735,
      "train_tokens_per_second": 12165.173
    },
    {
      "epoch": 1.596417990842984,
      "grad_norm": 0.8363342511839025,
      "learning_rate": 9.718284096239077e-07,
      "loss": 0.32360756397247314,
      "num_input_tokens_seen": 755702248,
      "step": 23710,
      "train_runtime": 62119.1171,
      "train_tokens_per_second": 12165.373
    },
    {
      "epoch": 1.5967546458389443,
      "grad_norm": 0.8072185909567865,
      "learning_rate": 9.702625838794215e-07,
      "loss": 0.28470401763916015,
      "num_input_tokens_seen": 755859784,
      "step": 23715,
      "train_runtime": 62132.2815,
      "train_tokens_per_second": 12165.331
    },
    {
      "epoch": 1.5970913008349044,
      "grad_norm": 0.7808018944773587,
      "learning_rate": 9.686978850381535e-07,
      "loss": 0.284358286857605,
      "num_input_tokens_seen": 756022776,
      "step": 23720,
      "train_runtime": 62145.1256,
      "train_tokens_per_second": 12165.44
    },
    {
      "epoch": 1.5974279558308644,
      "grad_norm": 0.8129130908518589,
      "learning_rate": 9.67134313537666e-07,
      "loss": 0.2720520257949829,
      "num_input_tokens_seen": 756181168,
      "step": 23725,
      "train_runtime": 62158.7961,
      "train_tokens_per_second": 12165.312
    },
    {
      "epoch": 1.5977646108268246,
      "grad_norm": 2.105982538627707,
      "learning_rate": 9.6557186981521e-07,
      "loss": 0.305348801612854,
      "num_input_tokens_seen": 756341096,
      "step": 23730,
      "train_runtime": 62171.4801,
      "train_tokens_per_second": 12165.403
    },
    {
      "epoch": 1.5981012658227849,
      "grad_norm": 0.8120863702396408,
      "learning_rate": 9.640105543077133e-07,
      "loss": 0.29778907299041746,
      "num_input_tokens_seen": 756503600,
      "step": 23735,
      "train_runtime": 62184.5413,
      "train_tokens_per_second": 12165.461
    },
    {
      "epoch": 1.598437920818745,
      "grad_norm": 0.8340768863410528,
      "learning_rate": 9.624503674517972e-07,
      "loss": 0.2994217872619629,
      "num_input_tokens_seen": 756663888,
      "step": 23740,
      "train_runtime": 62196.9896,
      "train_tokens_per_second": 12165.603
    },
    {
      "epoch": 1.598774575814705,
      "grad_norm": 1.028529909831064,
      "learning_rate": 9.608913096837603e-07,
      "loss": 0.2685131311416626,
      "num_input_tokens_seen": 756827640,
      "step": 23745,
      "train_runtime": 62210.1021,
      "train_tokens_per_second": 12165.671
    },
    {
      "epoch": 1.5991112308106652,
      "grad_norm": 0.7757460874293973,
      "learning_rate": 9.593333814395889e-07,
      "loss": 0.3071715831756592,
      "num_input_tokens_seen": 756980088,
      "step": 23750,
      "train_runtime": 62222.9411,
      "train_tokens_per_second": 12165.611
    },
    {
      "epoch": 1.5994478858066254,
      "grad_norm": 0.759429298468551,
      "learning_rate": 9.577765831549529e-07,
      "loss": 0.2900444507598877,
      "num_input_tokens_seen": 757142560,
      "step": 23755,
      "train_runtime": 62236.2082,
      "train_tokens_per_second": 12165.628
    },
    {
      "epoch": 1.5997845408025855,
      "grad_norm": 0.7540992120671355,
      "learning_rate": 9.562209152652058e-07,
      "loss": 0.3045480251312256,
      "num_input_tokens_seen": 757299688,
      "step": 23760,
      "train_runtime": 62248.6255,
      "train_tokens_per_second": 12165.725
    },
    {
      "epoch": 1.6001211957985455,
      "grad_norm": 0.8511417040813506,
      "learning_rate": 9.54666378205385e-07,
      "loss": 0.31048357486724854,
      "num_input_tokens_seen": 757462920,
      "step": 23765,
      "train_runtime": 62261.5311,
      "train_tokens_per_second": 12165.825
    },
    {
      "epoch": 1.6004578507945058,
      "grad_norm": 0.7905626781475694,
      "learning_rate": 9.531129724102117e-07,
      "loss": 0.2925772428512573,
      "num_input_tokens_seen": 757623264,
      "step": 23770,
      "train_runtime": 62274.8925,
      "train_tokens_per_second": 12165.79
    },
    {
      "epoch": 1.600794505790466,
      "grad_norm": 0.7970022112572606,
      "learning_rate": 9.515606983140896e-07,
      "loss": 0.28814122676849363,
      "num_input_tokens_seen": 757783208,
      "step": 23775,
      "train_runtime": 62287.8552,
      "train_tokens_per_second": 12165.826
    },
    {
      "epoch": 1.601131160786426,
      "grad_norm": 0.8095590586517496,
      "learning_rate": 9.500095563511119e-07,
      "loss": 0.29329063892364504,
      "num_input_tokens_seen": 757944936,
      "step": 23780,
      "train_runtime": 62300.3432,
      "train_tokens_per_second": 12165.983
    },
    {
      "epoch": 1.601467815782386,
      "grad_norm": 0.7846797695107003,
      "learning_rate": 9.484595469550445e-07,
      "loss": 0.2858744621276855,
      "num_input_tokens_seen": 758106224,
      "step": 23785,
      "train_runtime": 62313.2989,
      "train_tokens_per_second": 12166.042
    },
    {
      "epoch": 1.6018044707783463,
      "grad_norm": 0.7561047913931127,
      "learning_rate": 9.469106705593462e-07,
      "loss": 0.2850042819976807,
      "num_input_tokens_seen": 758262888,
      "step": 23790,
      "train_runtime": 62326.8254,
      "train_tokens_per_second": 12165.915
    },
    {
      "epoch": 1.6021411257743066,
      "grad_norm": 0.8383870082780074,
      "learning_rate": 9.453629275971549e-07,
      "loss": 0.29554004669189454,
      "num_input_tokens_seen": 758421800,
      "step": 23795,
      "train_runtime": 62339.3403,
      "train_tokens_per_second": 12166.022
    },
    {
      "epoch": 1.6024777807702666,
      "grad_norm": 0.736498479841838,
      "learning_rate": 9.438163185012916e-07,
      "loss": 0.28136382102966306,
      "num_input_tokens_seen": 758575968,
      "step": 23800,
      "train_runtime": 62351.9331,
      "train_tokens_per_second": 12166.038
    },
    {
      "epoch": 1.6028144357662266,
      "grad_norm": 0.7641098663703605,
      "learning_rate": 9.422708437042604e-07,
      "loss": 0.26924445629119875,
      "num_input_tokens_seen": 758738456,
      "step": 23805,
      "train_runtime": 62364.375,
      "train_tokens_per_second": 12166.216
    },
    {
      "epoch": 1.6031510907621869,
      "grad_norm": 0.763982949181992,
      "learning_rate": 9.40726503638249e-07,
      "loss": 0.3191770076751709,
      "num_input_tokens_seen": 758896608,
      "step": 23810,
      "train_runtime": 62377.5477,
      "train_tokens_per_second": 12166.182
    },
    {
      "epoch": 1.6034877457581471,
      "grad_norm": 0.8202118561698065,
      "learning_rate": 9.391832987351268e-07,
      "loss": 0.29884905815124513,
      "num_input_tokens_seen": 759052480,
      "step": 23815,
      "train_runtime": 62390.2454,
      "train_tokens_per_second": 12166.204
    },
    {
      "epoch": 1.6038244007541071,
      "grad_norm": 0.8606186146560654,
      "learning_rate": 9.376412294264458e-07,
      "loss": 0.29393439292907714,
      "num_input_tokens_seen": 759211120,
      "step": 23820,
      "train_runtime": 62402.8138,
      "train_tokens_per_second": 12166.296
    },
    {
      "epoch": 1.6041610557500672,
      "grad_norm": 0.7100429244040334,
      "learning_rate": 9.361002961434401e-07,
      "loss": 0.2816779613494873,
      "num_input_tokens_seen": 759371640,
      "step": 23825,
      "train_runtime": 62416.0655,
      "train_tokens_per_second": 12166.285
    },
    {
      "epoch": 1.6044977107460274,
      "grad_norm": 0.862546006701609,
      "learning_rate": 9.345604993170299e-07,
      "loss": 0.3149986267089844,
      "num_input_tokens_seen": 759531776,
      "step": 23830,
      "train_runtime": 62428.9873,
      "train_tokens_per_second": 12166.332
    },
    {
      "epoch": 1.6048343657419877,
      "grad_norm": 0.8740335285982169,
      "learning_rate": 9.330218393778101e-07,
      "loss": 0.2918510437011719,
      "num_input_tokens_seen": 759686128,
      "step": 23835,
      "train_runtime": 62441.7941,
      "train_tokens_per_second": 12166.308
    },
    {
      "epoch": 1.6051710207379477,
      "grad_norm": 0.8463441127459306,
      "learning_rate": 9.314843167560656e-07,
      "loss": 0.28692121505737306,
      "num_input_tokens_seen": 759839216,
      "step": 23840,
      "train_runtime": 62454.3017,
      "train_tokens_per_second": 12166.323
    },
    {
      "epoch": 1.6055076757339077,
      "grad_norm": 0.7739159853638357,
      "learning_rate": 9.299479318817578e-07,
      "loss": 0.28337562084198,
      "num_input_tokens_seen": 759995656,
      "step": 23845,
      "train_runtime": 62467.1842,
      "train_tokens_per_second": 12166.318
    },
    {
      "epoch": 1.605844330729868,
      "grad_norm": 0.8445095272669748,
      "learning_rate": 9.284126851845316e-07,
      "loss": 0.2813390254974365,
      "num_input_tokens_seen": 760159496,
      "step": 23850,
      "train_runtime": 62479.616,
      "train_tokens_per_second": 12166.52
    },
    {
      "epoch": 1.6061809857258282,
      "grad_norm": 0.8061944828827718,
      "learning_rate": 9.268785770937172e-07,
      "loss": 0.29737081527709963,
      "num_input_tokens_seen": 760320520,
      "step": 23855,
      "train_runtime": 62492.7043,
      "train_tokens_per_second": 12166.549
    },
    {
      "epoch": 1.6065176407217883,
      "grad_norm": 0.9164530419514815,
      "learning_rate": 9.253456080383178e-07,
      "loss": 0.3048983573913574,
      "num_input_tokens_seen": 760481760,
      "step": 23860,
      "train_runtime": 62505.7277,
      "train_tokens_per_second": 12166.593
    },
    {
      "epoch": 1.6068542957177483,
      "grad_norm": 0.808491764444421,
      "learning_rate": 9.238137784470286e-07,
      "loss": 0.31508593559265136,
      "num_input_tokens_seen": 760634008,
      "step": 23865,
      "train_runtime": 62519.1102,
      "train_tokens_per_second": 12166.424
    },
    {
      "epoch": 1.6071909507137085,
      "grad_norm": 0.8793919761807834,
      "learning_rate": 9.222830887482153e-07,
      "loss": 0.27198152542114257,
      "num_input_tokens_seen": 760795600,
      "step": 23870,
      "train_runtime": 62532.7101,
      "train_tokens_per_second": 12166.362
    },
    {
      "epoch": 1.6075276057096688,
      "grad_norm": 0.9018421718930305,
      "learning_rate": 9.207535393699351e-07,
      "loss": 0.3121297597885132,
      "num_input_tokens_seen": 760949848,
      "step": 23875,
      "train_runtime": 62545.9747,
      "train_tokens_per_second": 12166.248
    },
    {
      "epoch": 1.6078642607056288,
      "grad_norm": 0.8258946859093949,
      "learning_rate": 9.192251307399197e-07,
      "loss": 0.30213170051574706,
      "num_input_tokens_seen": 761111240,
      "step": 23880,
      "train_runtime": 62559.5067,
      "train_tokens_per_second": 12166.196
    },
    {
      "epoch": 1.6082009157015889,
      "grad_norm": 0.8051883975711595,
      "learning_rate": 9.176978632855842e-07,
      "loss": 0.29917683601379397,
      "num_input_tokens_seen": 761269896,
      "step": 23885,
      "train_runtime": 62572.2454,
      "train_tokens_per_second": 12166.255
    },
    {
      "epoch": 1.608537570697549,
      "grad_norm": 0.8518695134779092,
      "learning_rate": 9.161717374340235e-07,
      "loss": 0.28396599292755126,
      "num_input_tokens_seen": 761432952,
      "step": 23890,
      "train_runtime": 62585.1253,
      "train_tokens_per_second": 12166.357
    },
    {
      "epoch": 1.6088742256935094,
      "grad_norm": 0.9069344781741441,
      "learning_rate": 9.14646753612014e-07,
      "loss": 0.2971592903137207,
      "num_input_tokens_seen": 761593472,
      "step": 23895,
      "train_runtime": 62598.9205,
      "train_tokens_per_second": 12166.24
    },
    {
      "epoch": 1.6092108806894694,
      "grad_norm": 0.7386370782188281,
      "learning_rate": 9.131229122460112e-07,
      "loss": 0.2967731475830078,
      "num_input_tokens_seen": 761757312,
      "step": 23900,
      "train_runtime": 62611.3584,
      "train_tokens_per_second": 12166.44
    },
    {
      "epoch": 1.6095475356854294,
      "grad_norm": 0.7768509211464094,
      "learning_rate": 9.116002137621566e-07,
      "loss": 0.3001314878463745,
      "num_input_tokens_seen": 761917216,
      "step": 23905,
      "train_runtime": 62624.2993,
      "train_tokens_per_second": 12166.479
    },
    {
      "epoch": 1.6098841906813897,
      "grad_norm": 0.8521260620788292,
      "learning_rate": 9.100786585862626e-07,
      "loss": 0.3259676218032837,
      "num_input_tokens_seen": 762073008,
      "step": 23910,
      "train_runtime": 62637.3959,
      "train_tokens_per_second": 12166.422
    },
    {
      "epoch": 1.61022084567735,
      "grad_norm": 0.8740086081050383,
      "learning_rate": 9.085582471438326e-07,
      "loss": 0.2971473217010498,
      "num_input_tokens_seen": 762233520,
      "step": 23915,
      "train_runtime": 62650.5497,
      "train_tokens_per_second": 12166.43
    },
    {
      "epoch": 1.61055750067331,
      "grad_norm": 0.8479815326973054,
      "learning_rate": 9.070389798600394e-07,
      "loss": 0.2925459146499634,
      "num_input_tokens_seen": 762388824,
      "step": 23920,
      "train_runtime": 62663.3493,
      "train_tokens_per_second": 12166.423
    },
    {
      "epoch": 1.61089415566927,
      "grad_norm": 0.8595847744997765,
      "learning_rate": 9.055208571597451e-07,
      "loss": 0.2985935926437378,
      "num_input_tokens_seen": 762548552,
      "step": 23925,
      "train_runtime": 62676.0562,
      "train_tokens_per_second": 12166.505
    },
    {
      "epoch": 1.6112308106652302,
      "grad_norm": 0.9508306098646893,
      "learning_rate": 9.040038794674872e-07,
      "loss": 0.3077725887298584,
      "num_input_tokens_seen": 762708328,
      "step": 23930,
      "train_runtime": 62689.199,
      "train_tokens_per_second": 12166.503
    },
    {
      "epoch": 1.6115674656611905,
      "grad_norm": 0.74859069790838,
      "learning_rate": 9.024880472074831e-07,
      "loss": 0.2688113212585449,
      "num_input_tokens_seen": 762867864,
      "step": 23935,
      "train_runtime": 62702.3344,
      "train_tokens_per_second": 12166.499
    },
    {
      "epoch": 1.6119041206571505,
      "grad_norm": 0.8101866821759104,
      "learning_rate": 9.009733608036308e-07,
      "loss": 0.30013370513916016,
      "num_input_tokens_seen": 763028280,
      "step": 23940,
      "train_runtime": 62714.7483,
      "train_tokens_per_second": 12166.648
    },
    {
      "epoch": 1.6122407756531107,
      "grad_norm": 0.9428810785294192,
      "learning_rate": 8.994598206795068e-07,
      "loss": 0.279677677154541,
      "num_input_tokens_seen": 763183752,
      "step": 23945,
      "train_runtime": 62727.7798,
      "train_tokens_per_second": 12166.599
    },
    {
      "epoch": 1.612577430649071,
      "grad_norm": 0.7337719320084028,
      "learning_rate": 8.979474272583672e-07,
      "loss": 0.2854472637176514,
      "num_input_tokens_seen": 763339344,
      "step": 23950,
      "train_runtime": 62740.8994,
      "train_tokens_per_second": 12166.535
    },
    {
      "epoch": 1.612914085645031,
      "grad_norm": 0.7969190508471568,
      "learning_rate": 8.964361809631517e-07,
      "loss": 0.29798264503479005,
      "num_input_tokens_seen": 763499776,
      "step": 23955,
      "train_runtime": 62753.3272,
      "train_tokens_per_second": 12166.682
    },
    {
      "epoch": 1.613250740640991,
      "grad_norm": 0.8482276013688518,
      "learning_rate": 8.949260822164707e-07,
      "loss": 0.29980831146240233,
      "num_input_tokens_seen": 763659024,
      "step": 23960,
      "train_runtime": 62767.3444,
      "train_tokens_per_second": 12166.502
    },
    {
      "epoch": 1.6135873956369513,
      "grad_norm": 0.7678386623510182,
      "learning_rate": 8.934171314406226e-07,
      "loss": 0.30230298042297366,
      "num_input_tokens_seen": 763821936,
      "step": 23965,
      "train_runtime": 62780.1289,
      "train_tokens_per_second": 12166.619
    },
    {
      "epoch": 1.6139240506329116,
      "grad_norm": 0.8064893242433274,
      "learning_rate": 8.919093290575765e-07,
      "loss": 0.2942954540252686,
      "num_input_tokens_seen": 763976568,
      "step": 23970,
      "train_runtime": 62793.1417,
      "train_tokens_per_second": 12166.561
    },
    {
      "epoch": 1.6142607056288716,
      "grad_norm": 0.8544901054127001,
      "learning_rate": 8.904026754889877e-07,
      "loss": 0.2950057744979858,
      "num_input_tokens_seen": 764136360,
      "step": 23975,
      "train_runtime": 62806.7229,
      "train_tokens_per_second": 12166.474
    },
    {
      "epoch": 1.6145973606248316,
      "grad_norm": 0.8924913419416338,
      "learning_rate": 8.888971711561867e-07,
      "loss": 0.31065855026245115,
      "num_input_tokens_seen": 764294456,
      "step": 23980,
      "train_runtime": 62819.4626,
      "train_tokens_per_second": 12166.523
    },
    {
      "epoch": 1.6149340156207919,
      "grad_norm": 0.9262072960785322,
      "learning_rate": 8.873928164801821e-07,
      "loss": 0.3403141975402832,
      "num_input_tokens_seen": 764452608,
      "step": 23985,
      "train_runtime": 62832.0475,
      "train_tokens_per_second": 12166.603
    },
    {
      "epoch": 1.6152706706167521,
      "grad_norm": 0.808070170410782,
      "learning_rate": 8.858896118816624e-07,
      "loss": 0.2961104869842529,
      "num_input_tokens_seen": 764600216,
      "step": 23990,
      "train_runtime": 62845.0414,
      "train_tokens_per_second": 12166.437
    },
    {
      "epoch": 1.6156073256127121,
      "grad_norm": 0.793795929528486,
      "learning_rate": 8.84387557780994e-07,
      "loss": 0.2917397737503052,
      "num_input_tokens_seen": 764755536,
      "step": 23995,
      "train_runtime": 62857.6182,
      "train_tokens_per_second": 12166.473
    },
    {
      "epoch": 1.6159439806086722,
      "grad_norm": 0.8207326650772828,
      "learning_rate": 8.828866545982196e-07,
      "loss": 0.27533702850341796,
      "num_input_tokens_seen": 764918488,
      "step": 24000,
      "train_runtime": 62870.4455,
      "train_tokens_per_second": 12166.583
    },
    {
      "epoch": 1.6162806356046324,
      "grad_norm": 0.807083008731918,
      "learning_rate": 8.813869027530664e-07,
      "loss": 0.29796686172485354,
      "num_input_tokens_seen": 765079184,
      "step": 24005,
      "train_runtime": 62882.8572,
      "train_tokens_per_second": 12166.737
    },
    {
      "epoch": 1.6166172906005927,
      "grad_norm": 0.8302061593282672,
      "learning_rate": 8.798883026649302e-07,
      "loss": 0.30478296279907224,
      "num_input_tokens_seen": 765240968,
      "step": 24010,
      "train_runtime": 62895.9421,
      "train_tokens_per_second": 12166.778
    },
    {
      "epoch": 1.6169539455965527,
      "grad_norm": 0.7336147187311574,
      "learning_rate": 8.783908547528947e-07,
      "loss": 0.2698078155517578,
      "num_input_tokens_seen": 765398312,
      "step": 24015,
      "train_runtime": 62908.9864,
      "train_tokens_per_second": 12166.756
    },
    {
      "epoch": 1.6172906005925127,
      "grad_norm": 0.8416134435279844,
      "learning_rate": 8.768945594357109e-07,
      "loss": 0.29384913444519045,
      "num_input_tokens_seen": 765555080,
      "step": 24020,
      "train_runtime": 62922.3438,
      "train_tokens_per_second": 12166.665
    },
    {
      "epoch": 1.617627255588473,
      "grad_norm": 0.8300741695905465,
      "learning_rate": 8.753994171318175e-07,
      "loss": 0.29502367973327637,
      "num_input_tokens_seen": 765709488,
      "step": 24025,
      "train_runtime": 62935.7849,
      "train_tokens_per_second": 12166.52
    },
    {
      "epoch": 1.6179639105844332,
      "grad_norm": 0.7823052689022453,
      "learning_rate": 8.739054282593245e-07,
      "loss": 0.3124000310897827,
      "num_input_tokens_seen": 765871520,
      "step": 24030,
      "train_runtime": 62948.5443,
      "train_tokens_per_second": 12166.628
    },
    {
      "epoch": 1.6183005655803933,
      "grad_norm": 0.8077522470188067,
      "learning_rate": 8.724125932360217e-07,
      "loss": 0.3132227182388306,
      "num_input_tokens_seen": 766030504,
      "step": 24035,
      "train_runtime": 62962.4612,
      "train_tokens_per_second": 12166.464
    },
    {
      "epoch": 1.6186372205763533,
      "grad_norm": 0.7610397747599223,
      "learning_rate": 8.70920912479375e-07,
      "loss": 0.30763969421386717,
      "num_input_tokens_seen": 766187000,
      "step": 24040,
      "train_runtime": 62975.347,
      "train_tokens_per_second": 12166.459
    },
    {
      "epoch": 1.6189738755723135,
      "grad_norm": 0.8297523286713625,
      "learning_rate": 8.694303864065285e-07,
      "loss": 0.30072593688964844,
      "num_input_tokens_seen": 766350144,
      "step": 24045,
      "train_runtime": 62988.2184,
      "train_tokens_per_second": 12166.563
    },
    {
      "epoch": 1.6193105305682738,
      "grad_norm": 0.7838228440717495,
      "learning_rate": 8.679410154343015e-07,
      "loss": 0.2808335781097412,
      "num_input_tokens_seen": 766509040,
      "step": 24050,
      "train_runtime": 63001.5735,
      "train_tokens_per_second": 12166.506
    },
    {
      "epoch": 1.6196471855642338,
      "grad_norm": 0.9291175282426894,
      "learning_rate": 8.664527999791961e-07,
      "loss": 0.3260240316390991,
      "num_input_tokens_seen": 766668720,
      "step": 24055,
      "train_runtime": 63014.9195,
      "train_tokens_per_second": 12166.464
    },
    {
      "epoch": 1.6199838405601938,
      "grad_norm": 0.8473704140035717,
      "learning_rate": 8.649657404573819e-07,
      "loss": 0.29311039447784426,
      "num_input_tokens_seen": 766832112,
      "step": 24060,
      "train_runtime": 63027.8574,
      "train_tokens_per_second": 12166.558
    },
    {
      "epoch": 1.620320495556154,
      "grad_norm": 0.9173578848697701,
      "learning_rate": 8.634798372847148e-07,
      "loss": 0.3103797912597656,
      "num_input_tokens_seen": 766991984,
      "step": 24065,
      "train_runtime": 63040.5792,
      "train_tokens_per_second": 12166.639
    },
    {
      "epoch": 1.6206571505521143,
      "grad_norm": 0.8451169695872937,
      "learning_rate": 8.619950908767189e-07,
      "loss": 0.26868252754211425,
      "num_input_tokens_seen": 767146712,
      "step": 24070,
      "train_runtime": 63053.468,
      "train_tokens_per_second": 12166.606
    },
    {
      "epoch": 1.6209938055480744,
      "grad_norm": 0.8244721094177448,
      "learning_rate": 8.605115016486016e-07,
      "loss": 0.2828901052474976,
      "num_input_tokens_seen": 767310552,
      "step": 24075,
      "train_runtime": 63065.9013,
      "train_tokens_per_second": 12166.805
    },
    {
      "epoch": 1.6213304605440344,
      "grad_norm": 0.9570790864965937,
      "learning_rate": 8.590290700152426e-07,
      "loss": 0.2815671920776367,
      "num_input_tokens_seen": 767471496,
      "step": 24080,
      "train_runtime": 63078.7731,
      "train_tokens_per_second": 12166.874
    },
    {
      "epoch": 1.6216671155399947,
      "grad_norm": 0.8345386623172896,
      "learning_rate": 8.575477963912005e-07,
      "loss": 0.2973333835601807,
      "num_input_tokens_seen": 767633560,
      "step": 24085,
      "train_runtime": 63091.884,
      "train_tokens_per_second": 12166.915
    },
    {
      "epoch": 1.622003770535955,
      "grad_norm": 0.8383253479804074,
      "learning_rate": 8.560676811907071e-07,
      "loss": 0.2976699352264404,
      "num_input_tokens_seen": 767795928,
      "step": 24090,
      "train_runtime": 63105.1958,
      "train_tokens_per_second": 12166.921
    },
    {
      "epoch": 1.622340425531915,
      "grad_norm": 0.8615703139623555,
      "learning_rate": 8.54588724827673e-07,
      "loss": 0.28347082138061525,
      "num_input_tokens_seen": 767948488,
      "step": 24095,
      "train_runtime": 63117.9038,
      "train_tokens_per_second": 12166.888
    },
    {
      "epoch": 1.622677080527875,
      "grad_norm": 0.849619425770004,
      "learning_rate": 8.531109277156818e-07,
      "loss": 0.29941763877868655,
      "num_input_tokens_seen": 768110632,
      "step": 24100,
      "train_runtime": 63130.8159,
      "train_tokens_per_second": 12166.968
    },
    {
      "epoch": 1.6230137355238352,
      "grad_norm": 0.6898343766474888,
      "learning_rate": 8.516342902679986e-07,
      "loss": 0.26726560592651366,
      "num_input_tokens_seen": 768272000,
      "step": 24105,
      "train_runtime": 63143.7565,
      "train_tokens_per_second": 12167.03
    },
    {
      "epoch": 1.6233503905197955,
      "grad_norm": 0.7843208623137609,
      "learning_rate": 8.501588128975557e-07,
      "loss": 0.28057262897491453,
      "num_input_tokens_seen": 768433744,
      "step": 24110,
      "train_runtime": 63157.1221,
      "train_tokens_per_second": 12167.016
    },
    {
      "epoch": 1.6236870455157555,
      "grad_norm": 0.751807847608498,
      "learning_rate": 8.486844960169704e-07,
      "loss": 0.2688769340515137,
      "num_input_tokens_seen": 768594160,
      "step": 24115,
      "train_runtime": 63170.5329,
      "train_tokens_per_second": 12166.973
    },
    {
      "epoch": 1.6240237005117155,
      "grad_norm": 0.8375231375872785,
      "learning_rate": 8.472113400385257e-07,
      "loss": 0.280105447769165,
      "num_input_tokens_seen": 768755832,
      "step": 24120,
      "train_runtime": 63183.592,
      "train_tokens_per_second": 12167.017
    },
    {
      "epoch": 1.6243603555076758,
      "grad_norm": 0.8365248711910799,
      "learning_rate": 8.457393453741886e-07,
      "loss": 0.2927290201187134,
      "num_input_tokens_seen": 768916224,
      "step": 24125,
      "train_runtime": 63196.9009,
      "train_tokens_per_second": 12166.993
    },
    {
      "epoch": 1.624697010503636,
      "grad_norm": 0.9052271176007426,
      "learning_rate": 8.442685124355965e-07,
      "loss": 0.2913320541381836,
      "num_input_tokens_seen": 769072768,
      "step": 24130,
      "train_runtime": 63209.9842,
      "train_tokens_per_second": 12166.951
    },
    {
      "epoch": 1.625033665499596,
      "grad_norm": 0.8045218226698183,
      "learning_rate": 8.427988416340633e-07,
      "loss": 0.3093292713165283,
      "num_input_tokens_seen": 769223880,
      "step": 24135,
      "train_runtime": 63223.8196,
      "train_tokens_per_second": 12166.678
    },
    {
      "epoch": 1.625370320495556,
      "grad_norm": 0.8876575969485171,
      "learning_rate": 8.413303333805777e-07,
      "loss": 0.29044246673583984,
      "num_input_tokens_seen": 769383144,
      "step": 24140,
      "train_runtime": 63237.6733,
      "train_tokens_per_second": 12166.531
    },
    {
      "epoch": 1.6257069754915163,
      "grad_norm": 0.9411217661060414,
      "learning_rate": 8.398629880858034e-07,
      "loss": 0.3149867057800293,
      "num_input_tokens_seen": 769545304,
      "step": 24145,
      "train_runtime": 63250.2657,
      "train_tokens_per_second": 12166.673
    },
    {
      "epoch": 1.6260436304874766,
      "grad_norm": 0.8002804179260032,
      "learning_rate": 8.383968061600778e-07,
      "loss": 0.3086592674255371,
      "num_input_tokens_seen": 769702512,
      "step": 24150,
      "train_runtime": 63263.9072,
      "train_tokens_per_second": 12166.535
    },
    {
      "epoch": 1.6263802854834366,
      "grad_norm": 0.7774754201494821,
      "learning_rate": 8.36931788013417e-07,
      "loss": 0.2835515022277832,
      "num_input_tokens_seen": 769863344,
      "step": 24155,
      "train_runtime": 63276.9583,
      "train_tokens_per_second": 12166.567
    },
    {
      "epoch": 1.6267169404793966,
      "grad_norm": 1.1161743661659889,
      "learning_rate": 8.354679340555044e-07,
      "loss": 0.3146426439285278,
      "num_input_tokens_seen": 770019560,
      "step": 24160,
      "train_runtime": 63289.9489,
      "train_tokens_per_second": 12166.538
    },
    {
      "epoch": 1.6270535954753569,
      "grad_norm": 0.7952408894900088,
      "learning_rate": 8.34005244695707e-07,
      "loss": 0.2780102252960205,
      "num_input_tokens_seen": 770176480,
      "step": 24165,
      "train_runtime": 63303.211,
      "train_tokens_per_second": 12166.468
    },
    {
      "epoch": 1.6273902504713171,
      "grad_norm": 0.7486200156296614,
      "learning_rate": 8.325437203430558e-07,
      "loss": 0.2834614753723145,
      "num_input_tokens_seen": 770339096,
      "step": 24170,
      "train_runtime": 63315.8924,
      "train_tokens_per_second": 12166.599
    },
    {
      "epoch": 1.6277269054672772,
      "grad_norm": 0.7737967800920933,
      "learning_rate": 8.310833614062652e-07,
      "loss": 0.2856858730316162,
      "num_input_tokens_seen": 770497872,
      "step": 24175,
      "train_runtime": 63328.5684,
      "train_tokens_per_second": 12166.671
    },
    {
      "epoch": 1.6280635604632372,
      "grad_norm": 0.9274625579728031,
      "learning_rate": 8.296241682937189e-07,
      "loss": 0.29600985050201417,
      "num_input_tokens_seen": 770661040,
      "step": 24180,
      "train_runtime": 63341.4437,
      "train_tokens_per_second": 12166.774
    },
    {
      "epoch": 1.6284002154591974,
      "grad_norm": 0.8079267319951388,
      "learning_rate": 8.281661414134761e-07,
      "loss": 0.2907759189605713,
      "num_input_tokens_seen": 770820912,
      "step": 24185,
      "train_runtime": 63354.5998,
      "train_tokens_per_second": 12166.771
    },
    {
      "epoch": 1.6287368704551577,
      "grad_norm": 0.845864441731933,
      "learning_rate": 8.267092811732686e-07,
      "loss": 0.3025081634521484,
      "num_input_tokens_seen": 770981824,
      "step": 24190,
      "train_runtime": 63367.0383,
      "train_tokens_per_second": 12166.922
    },
    {
      "epoch": 1.6290735254511177,
      "grad_norm": 0.7080033781103762,
      "learning_rate": 8.252535879805029e-07,
      "loss": 0.27605319023132324,
      "num_input_tokens_seen": 771145112,
      "step": 24195,
      "train_runtime": 63380.5946,
      "train_tokens_per_second": 12166.896
    },
    {
      "epoch": 1.6294101804470777,
      "grad_norm": 0.8033876998268532,
      "learning_rate": 8.237990622422576e-07,
      "loss": 0.27672591209411623,
      "num_input_tokens_seen": 771305248,
      "step": 24200,
      "train_runtime": 63393.4492,
      "train_tokens_per_second": 12166.955
    },
    {
      "epoch": 1.629746835443038,
      "grad_norm": 0.7655449967698873,
      "learning_rate": 8.223457043652905e-07,
      "loss": 0.30393733978271487,
      "num_input_tokens_seen": 771463680,
      "step": 24205,
      "train_runtime": 63406.3015,
      "train_tokens_per_second": 12166.988
    },
    {
      "epoch": 1.6300834904389983,
      "grad_norm": 0.8451137950840152,
      "learning_rate": 8.208935147560227e-07,
      "loss": 0.28078672885894773,
      "num_input_tokens_seen": 771613552,
      "step": 24210,
      "train_runtime": 63420.1784,
      "train_tokens_per_second": 12166.688
    },
    {
      "epoch": 1.6304201454349583,
      "grad_norm": 0.804429927757353,
      "learning_rate": 8.194424938205597e-07,
      "loss": 0.3120806455612183,
      "num_input_tokens_seen": 771777376,
      "step": 24215,
      "train_runtime": 63433.2778,
      "train_tokens_per_second": 12166.759
    },
    {
      "epoch": 1.6307568004309183,
      "grad_norm": 0.782631122733815,
      "learning_rate": 8.179926419646705e-07,
      "loss": 0.2855846881866455,
      "num_input_tokens_seen": 771938032,
      "step": 24220,
      "train_runtime": 63447.3338,
      "train_tokens_per_second": 12166.595
    },
    {
      "epoch": 1.6310934554268786,
      "grad_norm": 0.8600488533172096,
      "learning_rate": 8.165439595938047e-07,
      "loss": 0.3102189302444458,
      "num_input_tokens_seen": 772094568,
      "step": 24225,
      "train_runtime": 63460.8781,
      "train_tokens_per_second": 12166.465
    },
    {
      "epoch": 1.6314301104228388,
      "grad_norm": 0.806568039891321,
      "learning_rate": 8.150964471130801e-07,
      "loss": 0.27042691707611083,
      "num_input_tokens_seen": 772256024,
      "step": 24230,
      "train_runtime": 63473.8003,
      "train_tokens_per_second": 12166.532
    },
    {
      "epoch": 1.6317667654187988,
      "grad_norm": 0.8369493286590751,
      "learning_rate": 8.136501049272899e-07,
      "loss": 0.30075955390930176,
      "num_input_tokens_seen": 772415168,
      "step": 24235,
      "train_runtime": 63486.7184,
      "train_tokens_per_second": 12166.563
    },
    {
      "epoch": 1.6321034204147589,
      "grad_norm": 0.8156058326835,
      "learning_rate": 8.122049334408983e-07,
      "loss": 0.29486808776855467,
      "num_input_tokens_seen": 772571464,
      "step": 24240,
      "train_runtime": 63499.7164,
      "train_tokens_per_second": 12166.534
    },
    {
      "epoch": 1.6324400754107191,
      "grad_norm": 0.794288060930167,
      "learning_rate": 8.107609330580429e-07,
      "loss": 0.2836303234100342,
      "num_input_tokens_seen": 772730200,
      "step": 24245,
      "train_runtime": 63512.8394,
      "train_tokens_per_second": 12166.52
    },
    {
      "epoch": 1.6327767304066794,
      "grad_norm": 0.8035824180793449,
      "learning_rate": 8.093181041825332e-07,
      "loss": 0.3022428512573242,
      "num_input_tokens_seen": 772889904,
      "step": 24250,
      "train_runtime": 63525.4523,
      "train_tokens_per_second": 12166.618
    },
    {
      "epoch": 1.6331133854026394,
      "grad_norm": 0.7930371606299825,
      "learning_rate": 8.078764472178552e-07,
      "loss": 0.28709795475006106,
      "num_input_tokens_seen": 773050360,
      "step": 24255,
      "train_runtime": 63538.795,
      "train_tokens_per_second": 12166.588
    },
    {
      "epoch": 1.6334500403985994,
      "grad_norm": 0.7620219829466326,
      "learning_rate": 8.06435962567158e-07,
      "loss": 0.3325523376464844,
      "num_input_tokens_seen": 773200576,
      "step": 24260,
      "train_runtime": 63552.2196,
      "train_tokens_per_second": 12166.382
    },
    {
      "epoch": 1.6337866953945597,
      "grad_norm": 0.8221932630750418,
      "learning_rate": 8.049966506332746e-07,
      "loss": 0.26060056686401367,
      "num_input_tokens_seen": 773360104,
      "step": 24265,
      "train_runtime": 63565.2594,
      "train_tokens_per_second": 12166.396
    },
    {
      "epoch": 1.63412335039052,
      "grad_norm": 0.6856320664378047,
      "learning_rate": 8.035585118186984e-07,
      "loss": 0.2741750717163086,
      "num_input_tokens_seen": 773516544,
      "step": 24270,
      "train_runtime": 63578.3636,
      "train_tokens_per_second": 12166.349
    },
    {
      "epoch": 1.63446000538648,
      "grad_norm": 0.8581768641464982,
      "learning_rate": 8.021215465256038e-07,
      "loss": 0.2776967525482178,
      "num_input_tokens_seen": 773680384,
      "step": 24275,
      "train_runtime": 63590.7991,
      "train_tokens_per_second": 12166.546
    },
    {
      "epoch": 1.63479666038244,
      "grad_norm": 0.7821643611991164,
      "learning_rate": 8.006857551558328e-07,
      "loss": 0.29167101383209226,
      "num_input_tokens_seen": 773841632,
      "step": 24280,
      "train_runtime": 63603.2777,
      "train_tokens_per_second": 12166.694
    },
    {
      "epoch": 1.6351333153784002,
      "grad_norm": 0.8211554745480923,
      "learning_rate": 7.992511381108997e-07,
      "loss": 0.29289889335632324,
      "num_input_tokens_seen": 774005472,
      "step": 24285,
      "train_runtime": 63615.7292,
      "train_tokens_per_second": 12166.888
    },
    {
      "epoch": 1.6354699703743605,
      "grad_norm": 0.7986141595545517,
      "learning_rate": 7.978176957919909e-07,
      "loss": 0.29323616027832033,
      "num_input_tokens_seen": 774162896,
      "step": 24290,
      "train_runtime": 63628.9795,
      "train_tokens_per_second": 12166.829
    },
    {
      "epoch": 1.6358066253703205,
      "grad_norm": 0.7328296562966558,
      "learning_rate": 7.963854285999634e-07,
      "loss": 0.26825311183929446,
      "num_input_tokens_seen": 774323320,
      "step": 24295,
      "train_runtime": 63642.284,
      "train_tokens_per_second": 12166.806
    },
    {
      "epoch": 1.6361432803662805,
      "grad_norm": 0.8240197526386364,
      "learning_rate": 7.949543369353452e-07,
      "loss": 0.283182430267334,
      "num_input_tokens_seen": 774475992,
      "step": 24300,
      "train_runtime": 63654.9716,
      "train_tokens_per_second": 12166.779
    },
    {
      "epoch": 1.6364799353622408,
      "grad_norm": 0.7668292683811512,
      "learning_rate": 7.9352442119834e-07,
      "loss": 0.31056530475616456,
      "num_input_tokens_seen": 774633784,
      "step": 24305,
      "train_runtime": 63668.0438,
      "train_tokens_per_second": 12166.76
    },
    {
      "epoch": 1.636816590358201,
      "grad_norm": 0.8738225059299793,
      "learning_rate": 7.920956817888148e-07,
      "loss": 0.30580363273620603,
      "num_input_tokens_seen": 774792136,
      "step": 24310,
      "train_runtime": 63680.5945,
      "train_tokens_per_second": 12166.848
    },
    {
      "epoch": 1.637153245354161,
      "grad_norm": 0.8642533231809156,
      "learning_rate": 7.906681191063165e-07,
      "loss": 0.28445570468902587,
      "num_input_tokens_seen": 774953704,
      "step": 24315,
      "train_runtime": 63693.924,
      "train_tokens_per_second": 12166.839
    },
    {
      "epoch": 1.637489900350121,
      "grad_norm": 0.8458765554541205,
      "learning_rate": 7.892417335500541e-07,
      "loss": 0.29718751907348634,
      "num_input_tokens_seen": 775108032,
      "step": 24320,
      "train_runtime": 63707.1626,
      "train_tokens_per_second": 12166.733
    },
    {
      "epoch": 1.6378265553460813,
      "grad_norm": 0.8596582141905486,
      "learning_rate": 7.878165255189146e-07,
      "loss": 0.30609903335571287,
      "num_input_tokens_seen": 775263728,
      "step": 24325,
      "train_runtime": 63721.4602,
      "train_tokens_per_second": 12166.446
    },
    {
      "epoch": 1.6381632103420416,
      "grad_norm": 0.9126288868451309,
      "learning_rate": 7.863924954114522e-07,
      "loss": 0.2848400115966797,
      "num_input_tokens_seen": 775417728,
      "step": 24330,
      "train_runtime": 63734.9331,
      "train_tokens_per_second": 12166.291
    },
    {
      "epoch": 1.6384998653380016,
      "grad_norm": 0.738572347750081,
      "learning_rate": 7.849696436258919e-07,
      "loss": 0.2891238212585449,
      "num_input_tokens_seen": 775575368,
      "step": 24335,
      "train_runtime": 63747.9713,
      "train_tokens_per_second": 12166.275
    },
    {
      "epoch": 1.6388365203339617,
      "grad_norm": 0.824887729877554,
      "learning_rate": 7.835479705601307e-07,
      "loss": 0.30775496959686277,
      "num_input_tokens_seen": 775732568,
      "step": 24340,
      "train_runtime": 63761.221,
      "train_tokens_per_second": 12166.213
    },
    {
      "epoch": 1.639173175329922,
      "grad_norm": 0.702895576509229,
      "learning_rate": 7.821274766117337e-07,
      "loss": 0.2852254629135132,
      "num_input_tokens_seen": 775892552,
      "step": 24345,
      "train_runtime": 63774.0817,
      "train_tokens_per_second": 12166.268
    },
    {
      "epoch": 1.6395098303258822,
      "grad_norm": 0.8800750132058125,
      "learning_rate": 7.807081621779367e-07,
      "loss": 0.31917245388031007,
      "num_input_tokens_seen": 776052032,
      "step": 24350,
      "train_runtime": 63787.306,
      "train_tokens_per_second": 12166.246
    },
    {
      "epoch": 1.6398464853218422,
      "grad_norm": 0.8215854971835287,
      "learning_rate": 7.792900276556509e-07,
      "loss": 0.30937120914459226,
      "num_input_tokens_seen": 776215016,
      "step": 24355,
      "train_runtime": 63800.1336,
      "train_tokens_per_second": 12166.354
    },
    {
      "epoch": 1.6401831403178022,
      "grad_norm": 0.9790639792640997,
      "learning_rate": 7.778730734414469e-07,
      "loss": 0.29734656810760496,
      "num_input_tokens_seen": 776371536,
      "step": 24360,
      "train_runtime": 63813.3018,
      "train_tokens_per_second": 12166.296
    },
    {
      "epoch": 1.6405197953137625,
      "grad_norm": 0.8277688813243606,
      "learning_rate": 7.764572999315772e-07,
      "loss": 0.29772610664367677,
      "num_input_tokens_seen": 776533944,
      "step": 24365,
      "train_runtime": 63825.7413,
      "train_tokens_per_second": 12166.47
    },
    {
      "epoch": 1.6408564503097227,
      "grad_norm": 0.9025338683222252,
      "learning_rate": 7.750427075219536e-07,
      "loss": 0.30924651622772215,
      "num_input_tokens_seen": 776694664,
      "step": 24370,
      "train_runtime": 63839.2529,
      "train_tokens_per_second": 12166.412
    },
    {
      "epoch": 1.6411931053056827,
      "grad_norm": 0.8534588650036673,
      "learning_rate": 7.736292966081655e-07,
      "loss": 0.2794644355773926,
      "num_input_tokens_seen": 776848984,
      "step": 24375,
      "train_runtime": 63852.7888,
      "train_tokens_per_second": 12166.25
    },
    {
      "epoch": 1.6415297603016428,
      "grad_norm": 0.6731146122196049,
      "learning_rate": 7.722170675854673e-07,
      "loss": 0.2518440246582031,
      "num_input_tokens_seen": 777012168,
      "step": 24380,
      "train_runtime": 63865.6895,
      "train_tokens_per_second": 12166.347
    },
    {
      "epoch": 1.641866415297603,
      "grad_norm": 0.8778606499912358,
      "learning_rate": 7.70806020848785e-07,
      "loss": 0.28110690116882325,
      "num_input_tokens_seen": 777169536,
      "step": 24385,
      "train_runtime": 63878.5592,
      "train_tokens_per_second": 12166.36
    },
    {
      "epoch": 1.6422030702935633,
      "grad_norm": 0.7349206220743536,
      "learning_rate": 7.693961567927133e-07,
      "loss": 0.28679332733154295,
      "num_input_tokens_seen": 777328920,
      "step": 24390,
      "train_runtime": 63891.8442,
      "train_tokens_per_second": 12166.325
    },
    {
      "epoch": 1.6425397252895233,
      "grad_norm": 0.794626711975228,
      "learning_rate": 7.679874758115153e-07,
      "loss": 0.28293309211730955,
      "num_input_tokens_seen": 777489512,
      "step": 24395,
      "train_runtime": 63904.295,
      "train_tokens_per_second": 12166.467
    },
    {
      "epoch": 1.6428763802854833,
      "grad_norm": 0.7491062357049881,
      "learning_rate": 7.66579978299124e-07,
      "loss": 0.29605047702789306,
      "num_input_tokens_seen": 777651216,
      "step": 24400,
      "train_runtime": 63916.8433,
      "train_tokens_per_second": 12166.609
    },
    {
      "epoch": 1.6432130352814436,
      "grad_norm": 0.9334734303650634,
      "learning_rate": 7.651736646491447e-07,
      "loss": 0.31244916915893556,
      "num_input_tokens_seen": 777805072,
      "step": 24405,
      "train_runtime": 63929.6701,
      "train_tokens_per_second": 12166.574
    },
    {
      "epoch": 1.6435496902774038,
      "grad_norm": 0.7137461174549735,
      "learning_rate": 7.63768535254844e-07,
      "loss": 0.2654087543487549,
      "num_input_tokens_seen": 777961056,
      "step": 24410,
      "train_runtime": 63942.3061,
      "train_tokens_per_second": 12166.609
    },
    {
      "epoch": 1.6438863452733639,
      "grad_norm": 0.8075241409367051,
      "learning_rate": 7.623645905091664e-07,
      "loss": 0.3217033863067627,
      "num_input_tokens_seen": 778123672,
      "step": 24415,
      "train_runtime": 63956.2644,
      "train_tokens_per_second": 12166.497
    },
    {
      "epoch": 1.6442230002693239,
      "grad_norm": 0.8320110918344632,
      "learning_rate": 7.609618308047156e-07,
      "loss": 0.295696496963501,
      "num_input_tokens_seen": 778279768,
      "step": 24420,
      "train_runtime": 63969.0501,
      "train_tokens_per_second": 12166.505
    },
    {
      "epoch": 1.6445596552652841,
      "grad_norm": 0.818360548136631,
      "learning_rate": 7.595602565337729e-07,
      "loss": 0.2922187805175781,
      "num_input_tokens_seen": 778442360,
      "step": 24425,
      "train_runtime": 63981.759,
      "train_tokens_per_second": 12166.63
    },
    {
      "epoch": 1.6448963102612444,
      "grad_norm": 0.9051018105676516,
      "learning_rate": 7.581598680882829e-07,
      "loss": 0.32529101371765134,
      "num_input_tokens_seen": 778603096,
      "step": 24430,
      "train_runtime": 63994.6544,
      "train_tokens_per_second": 12166.69
    },
    {
      "epoch": 1.6452329652572044,
      "grad_norm": 0.8512258132001573,
      "learning_rate": 7.567606658598592e-07,
      "loss": 0.29698853492736815,
      "num_input_tokens_seen": 778764928,
      "step": 24435,
      "train_runtime": 64007.3131,
      "train_tokens_per_second": 12166.812
    },
    {
      "epoch": 1.6455696202531644,
      "grad_norm": 0.7737965517345735,
      "learning_rate": 7.553626502397848e-07,
      "loss": 0.3084287166595459,
      "num_input_tokens_seen": 778925096,
      "step": 24440,
      "train_runtime": 64020.1161,
      "train_tokens_per_second": 12166.88
    },
    {
      "epoch": 1.6459062752491247,
      "grad_norm": 0.8388135353213578,
      "learning_rate": 7.539658216190099e-07,
      "loss": 0.27610599994659424,
      "num_input_tokens_seen": 779087216,
      "step": 24445,
      "train_runtime": 64034.0301,
      "train_tokens_per_second": 12166.768
    },
    {
      "epoch": 1.646242930245085,
      "grad_norm": 0.8353793438820949,
      "learning_rate": 7.525701803881525e-07,
      "loss": 0.28893413543701174,
      "num_input_tokens_seen": 779244472,
      "step": 24450,
      "train_runtime": 64047.2685,
      "train_tokens_per_second": 12166.709
    },
    {
      "epoch": 1.646579585241045,
      "grad_norm": 0.8495842876277677,
      "learning_rate": 7.511757269375019e-07,
      "loss": 0.29696547985076904,
      "num_input_tokens_seen": 779401800,
      "step": 24455,
      "train_runtime": 64060.6533,
      "train_tokens_per_second": 12166.623
    },
    {
      "epoch": 1.646916240237005,
      "grad_norm": 0.7318834495535411,
      "learning_rate": 7.497824616570087e-07,
      "loss": 0.3057637929916382,
      "num_input_tokens_seen": 779557528,
      "step": 24460,
      "train_runtime": 64074.4667,
      "train_tokens_per_second": 12166.43
    },
    {
      "epoch": 1.6472528952329653,
      "grad_norm": 0.7519547033221803,
      "learning_rate": 7.483903849362995e-07,
      "loss": 0.30504086017608645,
      "num_input_tokens_seen": 779720736,
      "step": 24465,
      "train_runtime": 64087.3919,
      "train_tokens_per_second": 12166.523
    },
    {
      "epoch": 1.6475895502289255,
      "grad_norm": 0.8023276071812534,
      "learning_rate": 7.469994971646594e-07,
      "loss": 0.30136420726776125,
      "num_input_tokens_seen": 779884576,
      "step": 24470,
      "train_runtime": 64099.8358,
      "train_tokens_per_second": 12166.717
    },
    {
      "epoch": 1.6479262052248855,
      "grad_norm": 0.7999572249193063,
      "learning_rate": 7.456097987310495e-07,
      "loss": 0.3037379026412964,
      "num_input_tokens_seen": 780042880,
      "step": 24475,
      "train_runtime": 64112.497,
      "train_tokens_per_second": 12166.784
    },
    {
      "epoch": 1.6482628602208456,
      "grad_norm": 0.8891394877797107,
      "learning_rate": 7.442212900240931e-07,
      "loss": 0.3144464731216431,
      "num_input_tokens_seen": 780202224,
      "step": 24480,
      "train_runtime": 64126.0869,
      "train_tokens_per_second": 12166.69
    },
    {
      "epoch": 1.6485995152168058,
      "grad_norm": 0.8225060528988165,
      "learning_rate": 7.42833971432082e-07,
      "loss": 0.26826887130737304,
      "num_input_tokens_seen": 780357752,
      "step": 24485,
      "train_runtime": 64139.6515,
      "train_tokens_per_second": 12166.542
    },
    {
      "epoch": 1.648936170212766,
      "grad_norm": 0.852375756224124,
      "learning_rate": 7.414478433429756e-07,
      "loss": 0.3042499542236328,
      "num_input_tokens_seen": 780517424,
      "step": 24490,
      "train_runtime": 64152.9481,
      "train_tokens_per_second": 12166.509
    },
    {
      "epoch": 1.649272825208726,
      "grad_norm": 0.8297732205005874,
      "learning_rate": 7.400629061444003e-07,
      "loss": 0.28996710777282714,
      "num_input_tokens_seen": 780680168,
      "step": 24495,
      "train_runtime": 64165.6924,
      "train_tokens_per_second": 12166.629
    },
    {
      "epoch": 1.6496094802046861,
      "grad_norm": 0.7762853077289567,
      "learning_rate": 7.386791602236482e-07,
      "loss": 0.3154712200164795,
      "num_input_tokens_seen": 780842352,
      "step": 24500,
      "train_runtime": 64178.3105,
      "train_tokens_per_second": 12166.764
    },
    {
      "epoch": 1.6499461352006464,
      "grad_norm": 0.7358643541224151,
      "learning_rate": 7.372966059676834e-07,
      "loss": 0.34027299880981443,
      "num_input_tokens_seen": 781005312,
      "step": 24505,
      "train_runtime": 64190.7988,
      "train_tokens_per_second": 12166.936
    },
    {
      "epoch": 1.6502827901966066,
      "grad_norm": 0.7860742417587284,
      "learning_rate": 7.359152437631273e-07,
      "loss": 0.32011971473693845,
      "num_input_tokens_seen": 781165832,
      "step": 24510,
      "train_runtime": 64203.6236,
      "train_tokens_per_second": 12167.005
    },
    {
      "epoch": 1.6506194451925666,
      "grad_norm": 0.7915108321402856,
      "learning_rate": 7.345350739962781e-07,
      "loss": 0.2763681888580322,
      "num_input_tokens_seen": 781325080,
      "step": 24515,
      "train_runtime": 64217.0725,
      "train_tokens_per_second": 12166.937
    },
    {
      "epoch": 1.6509561001885267,
      "grad_norm": 0.7946283389120277,
      "learning_rate": 7.331560970530921e-07,
      "loss": 0.2875345230102539,
      "num_input_tokens_seen": 781485296,
      "step": 24520,
      "train_runtime": 64230.7141,
      "train_tokens_per_second": 12166.847
    },
    {
      "epoch": 1.651292755184487,
      "grad_norm": 0.7706121419902808,
      "learning_rate": 7.317783133191986e-07,
      "loss": 0.2798483371734619,
      "num_input_tokens_seen": 781645032,
      "step": 24525,
      "train_runtime": 64243.4483,
      "train_tokens_per_second": 12166.922
    },
    {
      "epoch": 1.6516294101804472,
      "grad_norm": 0.8180446553701978,
      "learning_rate": 7.304017231798893e-07,
      "loss": 0.2899933815002441,
      "num_input_tokens_seen": 781805960,
      "step": 24530,
      "train_runtime": 64256.6043,
      "train_tokens_per_second": 12166.935
    },
    {
      "epoch": 1.6519660651764072,
      "grad_norm": 0.8591443386862397,
      "learning_rate": 7.290263270201231e-07,
      "loss": 0.29896907806396483,
      "num_input_tokens_seen": 781967288,
      "step": 24535,
      "train_runtime": 64270.0333,
      "train_tokens_per_second": 12166.903
    },
    {
      "epoch": 1.6523027201723672,
      "grad_norm": 0.73914215931305,
      "learning_rate": 7.276521252245261e-07,
      "loss": 0.2778233528137207,
      "num_input_tokens_seen": 782124272,
      "step": 24540,
      "train_runtime": 64283.9905,
      "train_tokens_per_second": 12166.704
    },
    {
      "epoch": 1.6526393751683275,
      "grad_norm": 0.8563638866414667,
      "learning_rate": 7.262791181773882e-07,
      "loss": 0.2700481414794922,
      "num_input_tokens_seen": 782284712,
      "step": 24545,
      "train_runtime": 64296.5472,
      "train_tokens_per_second": 12166.823
    },
    {
      "epoch": 1.6529760301642877,
      "grad_norm": 0.8050080644355319,
      "learning_rate": 7.24907306262666e-07,
      "loss": 0.3003551483154297,
      "num_input_tokens_seen": 782445200,
      "step": 24550,
      "train_runtime": 64309.815,
      "train_tokens_per_second": 12166.808
    },
    {
      "epoch": 1.6533126851602478,
      "grad_norm": 0.8627966280927176,
      "learning_rate": 7.235366898639856e-07,
      "loss": 0.31953437328338624,
      "num_input_tokens_seen": 782605672,
      "step": 24555,
      "train_runtime": 64322.6497,
      "train_tokens_per_second": 12166.876
    },
    {
      "epoch": 1.6536493401562078,
      "grad_norm": 0.8239890846106095,
      "learning_rate": 7.221672693646309e-07,
      "loss": 0.3055120944976807,
      "num_input_tokens_seen": 782767248,
      "step": 24560,
      "train_runtime": 64335.724,
      "train_tokens_per_second": 12166.914
    },
    {
      "epoch": 1.653985995152168,
      "grad_norm": 0.8567067485281902,
      "learning_rate": 7.207990451475594e-07,
      "loss": 0.3114776611328125,
      "num_input_tokens_seen": 782926632,
      "step": 24565,
      "train_runtime": 64348.1903,
      "train_tokens_per_second": 12167.034
    },
    {
      "epoch": 1.6543226501481283,
      "grad_norm": 0.9293053788639143,
      "learning_rate": 7.194320175953901e-07,
      "loss": 0.2935426950454712,
      "num_input_tokens_seen": 783085000,
      "step": 24570,
      "train_runtime": 64362.2029,
      "train_tokens_per_second": 12166.846
    },
    {
      "epoch": 1.6546593051440883,
      "grad_norm": 0.7893763492333773,
      "learning_rate": 7.180661870904066e-07,
      "loss": 0.2772150278091431,
      "num_input_tokens_seen": 783246264,
      "step": 24575,
      "train_runtime": 64375.1269,
      "train_tokens_per_second": 12166.908
    },
    {
      "epoch": 1.6549959601400483,
      "grad_norm": 0.7859561604847938,
      "learning_rate": 7.167015540145599e-07,
      "loss": 0.2854712247848511,
      "num_input_tokens_seen": 783405080,
      "step": 24580,
      "train_runtime": 64387.6834,
      "train_tokens_per_second": 12167.002
    },
    {
      "epoch": 1.6553326151360086,
      "grad_norm": 1.140232536017856,
      "learning_rate": 7.153381187494657e-07,
      "loss": 0.314476466178894,
      "num_input_tokens_seen": 783564432,
      "step": 24585,
      "train_runtime": 64401.3051,
      "train_tokens_per_second": 12166.903
    },
    {
      "epoch": 1.6556692701319689,
      "grad_norm": 0.7641572550593998,
      "learning_rate": 7.139758816764036e-07,
      "loss": 0.27437562942504884,
      "num_input_tokens_seen": 783722520,
      "step": 24590,
      "train_runtime": 64415.5495,
      "train_tokens_per_second": 12166.667
    },
    {
      "epoch": 1.6560059251279289,
      "grad_norm": 0.7809144199716783,
      "learning_rate": 7.126148431763191e-07,
      "loss": 0.2809741735458374,
      "num_input_tokens_seen": 783886144,
      "step": 24595,
      "train_runtime": 64428.6143,
      "train_tokens_per_second": 12166.739
    },
    {
      "epoch": 1.656342580123889,
      "grad_norm": 0.7386572533465269,
      "learning_rate": 7.11255003629821e-07,
      "loss": 0.3045970439910889,
      "num_input_tokens_seen": 784048728,
      "step": 24600,
      "train_runtime": 64441.9942,
      "train_tokens_per_second": 12166.736
    },
    {
      "epoch": 1.6566792351198492,
      "grad_norm": 0.7579949196464796,
      "learning_rate": 7.098963634171874e-07,
      "loss": 0.28805837631225584,
      "num_input_tokens_seen": 784212568,
      "step": 24605,
      "train_runtime": 64454.4314,
      "train_tokens_per_second": 12166.93
    },
    {
      "epoch": 1.6570158901158094,
      "grad_norm": 0.9961820875125137,
      "learning_rate": 7.085389229183537e-07,
      "loss": 0.3036026954650879,
      "num_input_tokens_seen": 784366664,
      "step": 24610,
      "train_runtime": 64468.3023,
      "train_tokens_per_second": 12166.703
    },
    {
      "epoch": 1.6573525451117694,
      "grad_norm": 0.8250558416433791,
      "learning_rate": 7.071826825129264e-07,
      "loss": 0.2892672777175903,
      "num_input_tokens_seen": 784524104,
      "step": 24615,
      "train_runtime": 64481.0635,
      "train_tokens_per_second": 12166.736
    },
    {
      "epoch": 1.6576892001077295,
      "grad_norm": 0.8334326248391096,
      "learning_rate": 7.058276425801735e-07,
      "loss": 0.2980621576309204,
      "num_input_tokens_seen": 784682144,
      "step": 24620,
      "train_runtime": 64493.9979,
      "train_tokens_per_second": 12166.747
    },
    {
      "epoch": 1.6580258551036897,
      "grad_norm": 0.832889988012181,
      "learning_rate": 7.044738034990273e-07,
      "loss": 0.28281078338623045,
      "num_input_tokens_seen": 784836320,
      "step": 24625,
      "train_runtime": 64507.2966,
      "train_tokens_per_second": 12166.629
    },
    {
      "epoch": 1.65836251009965,
      "grad_norm": 0.8173642569194277,
      "learning_rate": 7.031211656480841e-07,
      "loss": 0.30190751552581785,
      "num_input_tokens_seen": 784995672,
      "step": 24630,
      "train_runtime": 64520.6157,
      "train_tokens_per_second": 12166.587
    },
    {
      "epoch": 1.65869916509561,
      "grad_norm": 0.8238521754438797,
      "learning_rate": 7.017697294056058e-07,
      "loss": 0.27278072834014894,
      "num_input_tokens_seen": 785151448,
      "step": 24635,
      "train_runtime": 64533.2624,
      "train_tokens_per_second": 12166.616
    },
    {
      "epoch": 1.65903582009157,
      "grad_norm": 0.8437614794475293,
      "learning_rate": 7.004194951495163e-07,
      "loss": 0.31250820159912107,
      "num_input_tokens_seen": 785312232,
      "step": 24640,
      "train_runtime": 64546.0994,
      "train_tokens_per_second": 12166.688
    },
    {
      "epoch": 1.6593724750875303,
      "grad_norm": 0.8414446713180203,
      "learning_rate": 6.990704632574046e-07,
      "loss": 0.2838765621185303,
      "num_input_tokens_seen": 785475392,
      "step": 24645,
      "train_runtime": 64558.9902,
      "train_tokens_per_second": 12166.786
    },
    {
      "epoch": 1.6597091300834905,
      "grad_norm": 0.8752982276305432,
      "learning_rate": 6.977226341065219e-07,
      "loss": 0.2985983371734619,
      "num_input_tokens_seen": 785635888,
      "step": 24650,
      "train_runtime": 64572.5933,
      "train_tokens_per_second": 12166.708
    },
    {
      "epoch": 1.6600457850794506,
      "grad_norm": 0.7170720497317911,
      "learning_rate": 6.96376008073788e-07,
      "loss": 0.3128868579864502,
      "num_input_tokens_seen": 785791488,
      "step": 24655,
      "train_runtime": 64585.6943,
      "train_tokens_per_second": 12166.649
    },
    {
      "epoch": 1.6603824400754106,
      "grad_norm": 0.8087022953927849,
      "learning_rate": 6.950305855357781e-07,
      "loss": 0.28070683479309083,
      "num_input_tokens_seen": 785951064,
      "step": 24660,
      "train_runtime": 64599.4302,
      "train_tokens_per_second": 12166.532
    },
    {
      "epoch": 1.6607190950713708,
      "grad_norm": 0.8065406231325483,
      "learning_rate": 6.936863668687383e-07,
      "loss": 0.2973364591598511,
      "num_input_tokens_seen": 786110616,
      "step": 24665,
      "train_runtime": 64612.7108,
      "train_tokens_per_second": 12166.501
    },
    {
      "epoch": 1.661055750067331,
      "grad_norm": 0.8741764427135748,
      "learning_rate": 6.923433524485745e-07,
      "loss": 0.3147925138473511,
      "num_input_tokens_seen": 786272752,
      "step": 24670,
      "train_runtime": 64625.2994,
      "train_tokens_per_second": 12166.64
    },
    {
      "epoch": 1.6613924050632911,
      "grad_norm": 0.7848960465835426,
      "learning_rate": 6.910015426508554e-07,
      "loss": 0.28102545738220214,
      "num_input_tokens_seen": 786432592,
      "step": 24675,
      "train_runtime": 64637.9967,
      "train_tokens_per_second": 12166.723
    },
    {
      "epoch": 1.6617290600592511,
      "grad_norm": 0.8781606751794255,
      "learning_rate": 6.896609378508152e-07,
      "loss": 0.28995914459228517,
      "num_input_tokens_seen": 786590112,
      "step": 24680,
      "train_runtime": 64651.2917,
      "train_tokens_per_second": 12166.657
    },
    {
      "epoch": 1.6620657150552114,
      "grad_norm": 0.7834359487259148,
      "learning_rate": 6.883215384233488e-07,
      "loss": 0.29813547134399415,
      "num_input_tokens_seen": 786752560,
      "step": 24685,
      "train_runtime": 64663.956,
      "train_tokens_per_second": 12166.787
    },
    {
      "epoch": 1.6624023700511716,
      "grad_norm": 0.7660412647829145,
      "learning_rate": 6.869833447430152e-07,
      "loss": 0.2689774990081787,
      "num_input_tokens_seen": 786908016,
      "step": 24690,
      "train_runtime": 64677.1263,
      "train_tokens_per_second": 12166.713
    },
    {
      "epoch": 1.6627390250471317,
      "grad_norm": 0.7915924640479196,
      "learning_rate": 6.856463571840361e-07,
      "loss": 0.29881303310394286,
      "num_input_tokens_seen": 787063768,
      "step": 24695,
      "train_runtime": 64691.2609,
      "train_tokens_per_second": 12166.462
    },
    {
      "epoch": 1.6630756800430917,
      "grad_norm": 0.8443564079002597,
      "learning_rate": 6.843105761202945e-07,
      "loss": 0.30872206687927245,
      "num_input_tokens_seen": 787225720,
      "step": 24700,
      "train_runtime": 64704.2796,
      "train_tokens_per_second": 12166.517
    },
    {
      "epoch": 1.663412335039052,
      "grad_norm": 0.8447607488022807,
      "learning_rate": 6.829760019253412e-07,
      "loss": 0.31563601493835447,
      "num_input_tokens_seen": 787386296,
      "step": 24705,
      "train_runtime": 64718.5162,
      "train_tokens_per_second": 12166.322
    },
    {
      "epoch": 1.6637489900350122,
      "grad_norm": 0.7691659766558674,
      "learning_rate": 6.816426349723804e-07,
      "loss": 0.2945215702056885,
      "num_input_tokens_seen": 787545984,
      "step": 24710,
      "train_runtime": 64731.8672,
      "train_tokens_per_second": 12166.279
    },
    {
      "epoch": 1.6640856450309722,
      "grad_norm": 0.7179484283114269,
      "learning_rate": 6.803104756342877e-07,
      "loss": 0.27899928092956544,
      "num_input_tokens_seen": 787709392,
      "step": 24715,
      "train_runtime": 64744.8911,
      "train_tokens_per_second": 12166.356
    },
    {
      "epoch": 1.6644223000269323,
      "grad_norm": 0.7709716945746821,
      "learning_rate": 6.789795242835962e-07,
      "loss": 0.2734822750091553,
      "num_input_tokens_seen": 787871016,
      "step": 24720,
      "train_runtime": 64757.4259,
      "train_tokens_per_second": 12166.497
    },
    {
      "epoch": 1.6647589550228925,
      "grad_norm": 0.8109986173128949,
      "learning_rate": 6.776497812925015e-07,
      "loss": 0.2983099460601807,
      "num_input_tokens_seen": 788026944,
      "step": 24725,
      "train_runtime": 64770.4586,
      "train_tokens_per_second": 12166.456
    },
    {
      "epoch": 1.6650956100188528,
      "grad_norm": 0.8183885397275192,
      "learning_rate": 6.763212470328628e-07,
      "loss": 0.3206261396408081,
      "num_input_tokens_seen": 788187024,
      "step": 24730,
      "train_runtime": 64784.3197,
      "train_tokens_per_second": 12166.324
    },
    {
      "epoch": 1.6654322650148128,
      "grad_norm": 0.8371204081076756,
      "learning_rate": 6.749939218762003e-07,
      "loss": 0.28403759002685547,
      "num_input_tokens_seen": 788346408,
      "step": 24735,
      "train_runtime": 64797.4674,
      "train_tokens_per_second": 12166.315
    },
    {
      "epoch": 1.6657689200107728,
      "grad_norm": 0.762067437460842,
      "learning_rate": 6.736678061936958e-07,
      "loss": 0.30754292011260986,
      "num_input_tokens_seen": 788510248,
      "step": 24740,
      "train_runtime": 64809.8992,
      "train_tokens_per_second": 12166.509
    },
    {
      "epoch": 1.666105575006733,
      "grad_norm": 0.887905166242567,
      "learning_rate": 6.723429003561937e-07,
      "loss": 0.3109133720397949,
      "num_input_tokens_seen": 788674064,
      "step": 24745,
      "train_runtime": 64823.0241,
      "train_tokens_per_second": 12166.573
    },
    {
      "epoch": 1.6664422300026933,
      "grad_norm": 0.818816776587006,
      "learning_rate": 6.710192047341984e-07,
      "loss": 0.2882524967193604,
      "num_input_tokens_seen": 788836864,
      "step": 24750,
      "train_runtime": 64836.4744,
      "train_tokens_per_second": 12166.56
    },
    {
      "epoch": 1.6667788849986533,
      "grad_norm": 0.846196048328415,
      "learning_rate": 6.69696719697881e-07,
      "loss": 0.2957979917526245,
      "num_input_tokens_seen": 788995008,
      "step": 24755,
      "train_runtime": 64849.3859,
      "train_tokens_per_second": 12166.576
    },
    {
      "epoch": 1.6671155399946134,
      "grad_norm": 0.8950515125724958,
      "learning_rate": 6.683754456170655e-07,
      "loss": 0.2931551218032837,
      "num_input_tokens_seen": 789150408,
      "step": 24760,
      "train_runtime": 64862.6985,
      "train_tokens_per_second": 12166.475
    },
    {
      "epoch": 1.6674521949905736,
      "grad_norm": 0.8790866423810335,
      "learning_rate": 6.670553828612448e-07,
      "loss": 0.27227044105529785,
      "num_input_tokens_seen": 789298096,
      "step": 24765,
      "train_runtime": 64875.4303,
      "train_tokens_per_second": 12166.364
    },
    {
      "epoch": 1.6677888499865339,
      "grad_norm": 0.7534159667693646,
      "learning_rate": 6.657365317995696e-07,
      "loss": 0.2873995304107666,
      "num_input_tokens_seen": 789456344,
      "step": 24770,
      "train_runtime": 64888.323,
      "train_tokens_per_second": 12166.385
    },
    {
      "epoch": 1.668125504982494,
      "grad_norm": 0.9008555406699663,
      "learning_rate": 6.644188928008521e-07,
      "loss": 0.3240548610687256,
      "num_input_tokens_seen": 789615872,
      "step": 24775,
      "train_runtime": 64901.5678,
      "train_tokens_per_second": 12166.361
    },
    {
      "epoch": 1.668462159978454,
      "grad_norm": 0.841304067912514,
      "learning_rate": 6.631024662335667e-07,
      "loss": 0.2955070972442627,
      "num_input_tokens_seen": 789777096,
      "step": 24780,
      "train_runtime": 64915.5285,
      "train_tokens_per_second": 12166.228
    },
    {
      "epoch": 1.6687988149744142,
      "grad_norm": 0.7886676019245754,
      "learning_rate": 6.617872524658469e-07,
      "loss": 0.29398746490478517,
      "num_input_tokens_seen": 789939712,
      "step": 24785,
      "train_runtime": 64928.3662,
      "train_tokens_per_second": 12166.327
    },
    {
      "epoch": 1.6691354699703744,
      "grad_norm": 0.911116559677418,
      "learning_rate": 6.604732518654882e-07,
      "loss": 0.32519259452819826,
      "num_input_tokens_seen": 790100112,
      "step": 24790,
      "train_runtime": 64941.6739,
      "train_tokens_per_second": 12166.303
    },
    {
      "epoch": 1.6694721249663345,
      "grad_norm": 0.9911429598771344,
      "learning_rate": 6.591604647999473e-07,
      "loss": 0.3286020755767822,
      "num_input_tokens_seen": 790261832,
      "step": 24795,
      "train_runtime": 64954.2364,
      "train_tokens_per_second": 12166.44
    },
    {
      "epoch": 1.6698087799622945,
      "grad_norm": 0.8196102888807713,
      "learning_rate": 6.57848891636339e-07,
      "loss": 0.2925854682922363,
      "num_input_tokens_seen": 790424512,
      "step": 24800,
      "train_runtime": 64967.4825,
      "train_tokens_per_second": 12166.464
    },
    {
      "epoch": 1.6701454349582547,
      "grad_norm": 0.8547180961356758,
      "learning_rate": 6.565385327414441e-07,
      "loss": 0.302751350402832,
      "num_input_tokens_seen": 790587536,
      "step": 24805,
      "train_runtime": 64979.9176,
      "train_tokens_per_second": 12166.644
    },
    {
      "epoch": 1.670482089954215,
      "grad_norm": 0.9427370040418245,
      "learning_rate": 6.552293884816968e-07,
      "loss": 0.28897724151611326,
      "num_input_tokens_seen": 790749664,
      "step": 24810,
      "train_runtime": 64992.5046,
      "train_tokens_per_second": 12166.782
    },
    {
      "epoch": 1.670818744950175,
      "grad_norm": 0.8119793074297755,
      "learning_rate": 6.539214592231968e-07,
      "loss": 0.30264132022857665,
      "num_input_tokens_seen": 790913160,
      "step": 24815,
      "train_runtime": 65005.5266,
      "train_tokens_per_second": 12166.86
    },
    {
      "epoch": 1.671155399946135,
      "grad_norm": 0.7406111247740471,
      "learning_rate": 6.526147453317028e-07,
      "loss": 0.2658879280090332,
      "num_input_tokens_seen": 791072448,
      "step": 24820,
      "train_runtime": 65018.6997,
      "train_tokens_per_second": 12166.845
    },
    {
      "epoch": 1.6714920549420953,
      "grad_norm": 0.7930303218007095,
      "learning_rate": 6.513092471726307e-07,
      "loss": 0.28693251609802245,
      "num_input_tokens_seen": 791236072,
      "step": 24825,
      "train_runtime": 65031.7537,
      "train_tokens_per_second": 12166.919
    },
    {
      "epoch": 1.6718287099380555,
      "grad_norm": 0.7575652909819368,
      "learning_rate": 6.500049651110635e-07,
      "loss": 0.27247180938720705,
      "num_input_tokens_seen": 791397040,
      "step": 24830,
      "train_runtime": 65044.2045,
      "train_tokens_per_second": 12167.065
    },
    {
      "epoch": 1.6721653649340156,
      "grad_norm": 0.7998019760301968,
      "learning_rate": 6.487018995117339e-07,
      "loss": 0.28848905563354493,
      "num_input_tokens_seen": 791560680,
      "step": 24835,
      "train_runtime": 65057.2534,
      "train_tokens_per_second": 12167.14
    },
    {
      "epoch": 1.6725020199299756,
      "grad_norm": 0.8296847140983191,
      "learning_rate": 6.474000507390455e-07,
      "loss": 0.30805058479309083,
      "num_input_tokens_seen": 791718304,
      "step": 24840,
      "train_runtime": 65069.9378,
      "train_tokens_per_second": 12167.19
    },
    {
      "epoch": 1.6728386749259359,
      "grad_norm": 0.7701757083170343,
      "learning_rate": 6.460994191570513e-07,
      "loss": 0.30128433704376223,
      "num_input_tokens_seen": 791876000,
      "step": 24845,
      "train_runtime": 65082.9422,
      "train_tokens_per_second": 12167.182
    },
    {
      "epoch": 1.673175329921896,
      "grad_norm": 0.7513119329211971,
      "learning_rate": 6.448000051294717e-07,
      "loss": 0.2735928535461426,
      "num_input_tokens_seen": 792034328,
      "step": 24850,
      "train_runtime": 65095.5789,
      "train_tokens_per_second": 12167.252
    },
    {
      "epoch": 1.6735119849178561,
      "grad_norm": 0.8225468389917898,
      "learning_rate": 6.435018090196832e-07,
      "loss": 0.308670711517334,
      "num_input_tokens_seen": 792194048,
      "step": 24855,
      "train_runtime": 65108.2782,
      "train_tokens_per_second": 12167.332
    },
    {
      "epoch": 1.6738486399138162,
      "grad_norm": 0.8609698368247336,
      "learning_rate": 6.42204831190722e-07,
      "loss": 0.28301177024841306,
      "num_input_tokens_seen": 792355728,
      "step": 24860,
      "train_runtime": 65120.8944,
      "train_tokens_per_second": 12167.458
    },
    {
      "epoch": 1.6741852949097764,
      "grad_norm": 0.7659254821924985,
      "learning_rate": 6.409090720052835e-07,
      "loss": 0.29714231491088866,
      "num_input_tokens_seen": 792515296,
      "step": 24865,
      "train_runtime": 65134.6343,
      "train_tokens_per_second": 12167.341
    },
    {
      "epoch": 1.6745219499057367,
      "grad_norm": 0.8604700645447717,
      "learning_rate": 6.396145318257229e-07,
      "loss": 0.27208235263824465,
      "num_input_tokens_seen": 792675656,
      "step": 24870,
      "train_runtime": 65147.0717,
      "train_tokens_per_second": 12167.479
    },
    {
      "epoch": 1.6748586049016967,
      "grad_norm": 0.8609426563584132,
      "learning_rate": 6.383212110140535e-07,
      "loss": 0.3103998899459839,
      "num_input_tokens_seen": 792837608,
      "step": 24875,
      "train_runtime": 65160.2245,
      "train_tokens_per_second": 12167.509
    },
    {
      "epoch": 1.6751952598976567,
      "grad_norm": 0.9586762874187554,
      "learning_rate": 6.370291099319509e-07,
      "loss": 0.31565978527069094,
      "num_input_tokens_seen": 792991128,
      "step": 24880,
      "train_runtime": 65173.4604,
      "train_tokens_per_second": 12167.393
    },
    {
      "epoch": 1.675531914893617,
      "grad_norm": 0.8551573410931695,
      "learning_rate": 6.357382289407433e-07,
      "loss": 0.2924635410308838,
      "num_input_tokens_seen": 793153024,
      "step": 24885,
      "train_runtime": 65185.9169,
      "train_tokens_per_second": 12167.552
    },
    {
      "epoch": 1.6758685698895772,
      "grad_norm": 0.8476076837495868,
      "learning_rate": 6.344485684014251e-07,
      "loss": 0.30182876586914065,
      "num_input_tokens_seen": 793313656,
      "step": 24890,
      "train_runtime": 65200.4934,
      "train_tokens_per_second": 12167.295
    },
    {
      "epoch": 1.6762052248855372,
      "grad_norm": 0.8405731887367258,
      "learning_rate": 6.33160128674642e-07,
      "loss": 0.3063340187072754,
      "num_input_tokens_seen": 793467792,
      "step": 24895,
      "train_runtime": 65214.331,
      "train_tokens_per_second": 12167.077
    },
    {
      "epoch": 1.6765418798814973,
      "grad_norm": 0.7686886313507418,
      "learning_rate": 6.318729101207055e-07,
      "loss": 0.2698678493499756,
      "num_input_tokens_seen": 793629256,
      "step": 24900,
      "train_runtime": 65226.8217,
      "train_tokens_per_second": 12167.223
    },
    {
      "epoch": 1.6768785348774575,
      "grad_norm": 0.7284589151207705,
      "learning_rate": 6.305869130995807e-07,
      "loss": 0.29496011734008787,
      "num_input_tokens_seen": 793787552,
      "step": 24905,
      "train_runtime": 65240.1136,
      "train_tokens_per_second": 12167.17
    },
    {
      "epoch": 1.6772151898734178,
      "grad_norm": 0.833853455823911,
      "learning_rate": 6.293021379708924e-07,
      "loss": 0.292670464515686,
      "num_input_tokens_seen": 793948200,
      "step": 24910,
      "train_runtime": 65253.2612,
      "train_tokens_per_second": 12167.18
    },
    {
      "epoch": 1.6775518448693778,
      "grad_norm": 0.9040130025802267,
      "learning_rate": 6.280185850939241e-07,
      "loss": 0.3286106586456299,
      "num_input_tokens_seen": 794106544,
      "step": 24915,
      "train_runtime": 65266.4184,
      "train_tokens_per_second": 12167.154
    },
    {
      "epoch": 1.677888499865338,
      "grad_norm": 0.9045331775157412,
      "learning_rate": 6.267362548276173e-07,
      "loss": 0.29216506481170657,
      "num_input_tokens_seen": 794260440,
      "step": 24920,
      "train_runtime": 65279.3056,
      "train_tokens_per_second": 12167.109
    },
    {
      "epoch": 1.6782251548612983,
      "grad_norm": 1.0288707002050386,
      "learning_rate": 6.254551475305703e-07,
      "loss": 0.2966667652130127,
      "num_input_tokens_seen": 794412768,
      "step": 24925,
      "train_runtime": 65292.4407,
      "train_tokens_per_second": 12166.995
    },
    {
      "epoch": 1.6785618098572583,
      "grad_norm": 0.8068678335477649,
      "learning_rate": 6.241752635610443e-07,
      "loss": 0.311280345916748,
      "num_input_tokens_seen": 794569672,
      "step": 24930,
      "train_runtime": 65305.232,
      "train_tokens_per_second": 12167.014
    },
    {
      "epoch": 1.6788984648532184,
      "grad_norm": 1.0258507048417855,
      "learning_rate": 6.228966032769496e-07,
      "loss": 0.2994679927825928,
      "num_input_tokens_seen": 794725976,
      "step": 24935,
      "train_runtime": 65317.6931,
      "train_tokens_per_second": 12167.086
    },
    {
      "epoch": 1.6792351198491786,
      "grad_norm": 0.8490317773296896,
      "learning_rate": 6.216191670358651e-07,
      "loss": 0.2810512065887451,
      "num_input_tokens_seen": 794885856,
      "step": 24940,
      "train_runtime": 65330.8312,
      "train_tokens_per_second": 12167.086
    },
    {
      "epoch": 1.6795717748451389,
      "grad_norm": 0.8489160808967282,
      "learning_rate": 6.203429551950163e-07,
      "loss": 0.3216299533843994,
      "num_input_tokens_seen": 795048208,
      "step": 24945,
      "train_runtime": 65344.1076,
      "train_tokens_per_second": 12167.099
    },
    {
      "epoch": 1.679908429841099,
      "grad_norm": 0.802119375622299,
      "learning_rate": 6.19067968111296e-07,
      "loss": 0.28259685039520266,
      "num_input_tokens_seen": 795202472,
      "step": 24950,
      "train_runtime": 65356.8625,
      "train_tokens_per_second": 12167.085
    },
    {
      "epoch": 1.680245084837059,
      "grad_norm": 0.7646118795940126,
      "learning_rate": 6.177942061412478e-07,
      "loss": 0.27660164833068845,
      "num_input_tokens_seen": 795359088,
      "step": 24955,
      "train_runtime": 65369.9021,
      "train_tokens_per_second": 12167.053
    },
    {
      "epoch": 1.6805817398330192,
      "grad_norm": 0.8584390639720236,
      "learning_rate": 6.16521669641077e-07,
      "loss": 0.29123899936676023,
      "num_input_tokens_seen": 795520752,
      "step": 24960,
      "train_runtime": 65383.0624,
      "train_tokens_per_second": 12167.077
    },
    {
      "epoch": 1.6809183948289794,
      "grad_norm": 0.8825573592954843,
      "learning_rate": 6.152503589666426e-07,
      "loss": 0.2991896629333496,
      "num_input_tokens_seen": 795683296,
      "step": 24965,
      "train_runtime": 65395.7305,
      "train_tokens_per_second": 12167.206
    },
    {
      "epoch": 1.6812550498249395,
      "grad_norm": 0.7423301422017123,
      "learning_rate": 6.139802744734636e-07,
      "loss": 0.27493629455566404,
      "num_input_tokens_seen": 795841856,
      "step": 24970,
      "train_runtime": 65409.2445,
      "train_tokens_per_second": 12167.116
    },
    {
      "epoch": 1.6815917048208995,
      "grad_norm": 0.8120249723915178,
      "learning_rate": 6.127114165167142e-07,
      "loss": 0.27087857723236086,
      "num_input_tokens_seen": 796000576,
      "step": 24975,
      "train_runtime": 65422.6031,
      "train_tokens_per_second": 12167.058
    },
    {
      "epoch": 1.6819283598168597,
      "grad_norm": 0.8694596864525649,
      "learning_rate": 6.11443785451229e-07,
      "loss": 0.29536535739898684,
      "num_input_tokens_seen": 796158048,
      "step": 24980,
      "train_runtime": 65435.4906,
      "train_tokens_per_second": 12167.068
    },
    {
      "epoch": 1.68226501481282,
      "grad_norm": 0.7954669130086321,
      "learning_rate": 6.101773816314926e-07,
      "loss": 0.28241686820983886,
      "num_input_tokens_seen": 796321600,
      "step": 24985,
      "train_runtime": 65448.4976,
      "train_tokens_per_second": 12167.149
    },
    {
      "epoch": 1.68260166980878,
      "grad_norm": 0.8158471587174305,
      "learning_rate": 6.089122054116558e-07,
      "loss": 0.2966434478759766,
      "num_input_tokens_seen": 796483744,
      "step": 24990,
      "train_runtime": 65461.3473,
      "train_tokens_per_second": 12167.237
    },
    {
      "epoch": 1.68293832480474,
      "grad_norm": 0.8239316003862425,
      "learning_rate": 6.076482571455161e-07,
      "loss": 0.29096825122833253,
      "num_input_tokens_seen": 796644136,
      "step": 24995,
      "train_runtime": 65473.7756,
      "train_tokens_per_second": 12167.377
    },
    {
      "epoch": 1.6832749798007003,
      "grad_norm": 0.8748118720362762,
      "learning_rate": 6.063855371865369e-07,
      "loss": 0.29922871589660643,
      "num_input_tokens_seen": 796795752,
      "step": 25000,
      "train_runtime": 65486.8272,
      "train_tokens_per_second": 12167.268
    },
    {
      "epoch": 1.6836116347966605,
      "grad_norm": 0.9029056520147132,
      "learning_rate": 6.051240458878316e-07,
      "loss": 0.3211724281311035,
      "num_input_tokens_seen": 796952408,
      "step": 25005,
      "train_runtime": 65499.6099,
      "train_tokens_per_second": 12167.285
    },
    {
      "epoch": 1.6839482897926206,
      "grad_norm": 0.8838281004255323,
      "learning_rate": 6.038637836021721e-07,
      "loss": 0.3115694046020508,
      "num_input_tokens_seen": 797112144,
      "step": 25010,
      "train_runtime": 65512.2602,
      "train_tokens_per_second": 12167.374
    },
    {
      "epoch": 1.6842849447885806,
      "grad_norm": 0.760497580542647,
      "learning_rate": 6.026047506819882e-07,
      "loss": 0.28313503265380857,
      "num_input_tokens_seen": 797268448,
      "step": 25015,
      "train_runtime": 65525.3849,
      "train_tokens_per_second": 12167.322
    },
    {
      "epoch": 1.6846215997845408,
      "grad_norm": 0.759568014344448,
      "learning_rate": 6.013469474793632e-07,
      "loss": 0.28777251243591306,
      "num_input_tokens_seen": 797429480,
      "step": 25020,
      "train_runtime": 65537.8376,
      "train_tokens_per_second": 12167.467
    },
    {
      "epoch": 1.684958254780501,
      "grad_norm": 0.8391435060120102,
      "learning_rate": 6.000903743460368e-07,
      "loss": 0.2923095226287842,
      "num_input_tokens_seen": 797591656,
      "step": 25025,
      "train_runtime": 65550.3769,
      "train_tokens_per_second": 12167.614
    },
    {
      "epoch": 1.6852949097764611,
      "grad_norm": 0.779009047921114,
      "learning_rate": 5.98835031633409e-07,
      "loss": 0.306101131439209,
      "num_input_tokens_seen": 797754744,
      "step": 25030,
      "train_runtime": 65563.5068,
      "train_tokens_per_second": 12167.664
    },
    {
      "epoch": 1.6856315647724212,
      "grad_norm": 0.8654898552594379,
      "learning_rate": 5.97580919692528e-07,
      "loss": 0.3088191032409668,
      "num_input_tokens_seen": 797917040,
      "step": 25035,
      "train_runtime": 65576.7921,
      "train_tokens_per_second": 12167.674
    },
    {
      "epoch": 1.6859682197683814,
      "grad_norm": 0.8153521505961947,
      "learning_rate": 5.963280388741072e-07,
      "loss": 0.2744007587432861,
      "num_input_tokens_seen": 798076840,
      "step": 25040,
      "train_runtime": 65590.5025,
      "train_tokens_per_second": 12167.567
    },
    {
      "epoch": 1.6863048747643417,
      "grad_norm": 0.9418888309743303,
      "learning_rate": 5.950763895285055e-07,
      "loss": 0.27949841022491456,
      "num_input_tokens_seen": 798229648,
      "step": 25045,
      "train_runtime": 65603.2589,
      "train_tokens_per_second": 12167.53
    },
    {
      "epoch": 1.6866415297603017,
      "grad_norm": 0.7379573626459642,
      "learning_rate": 5.938259720057465e-07,
      "loss": 0.28723607063293455,
      "num_input_tokens_seen": 798388304,
      "step": 25050,
      "train_runtime": 65616.2777,
      "train_tokens_per_second": 12167.534
    },
    {
      "epoch": 1.6869781847562617,
      "grad_norm": 0.8030639037252914,
      "learning_rate": 5.925767866555043e-07,
      "loss": 0.2873709201812744,
      "num_input_tokens_seen": 798544608,
      "step": 25055,
      "train_runtime": 65629.6022,
      "train_tokens_per_second": 12167.445
    },
    {
      "epoch": 1.687314839752222,
      "grad_norm": 0.9214551354387678,
      "learning_rate": 5.913288338271095e-07,
      "loss": 0.3072784423828125,
      "num_input_tokens_seen": 798707512,
      "step": 25060,
      "train_runtime": 65642.5952,
      "train_tokens_per_second": 12167.519
    },
    {
      "epoch": 1.6876514947481822,
      "grad_norm": 0.7634145241254645,
      "learning_rate": 5.900821138695478e-07,
      "loss": 0.28399407863616943,
      "num_input_tokens_seen": 798869000,
      "step": 25065,
      "train_runtime": 65655.3825,
      "train_tokens_per_second": 12167.609
    },
    {
      "epoch": 1.6879881497441422,
      "grad_norm": 0.79298226071443,
      "learning_rate": 5.888366271314605e-07,
      "loss": 0.28875155448913575,
      "num_input_tokens_seen": 799031968,
      "step": 25070,
      "train_runtime": 65668.1901,
      "train_tokens_per_second": 12167.717
    },
    {
      "epoch": 1.6883248047401023,
      "grad_norm": 0.8122251724242145,
      "learning_rate": 5.875923739611439e-07,
      "loss": 0.3123459339141846,
      "num_input_tokens_seen": 799190928,
      "step": 25075,
      "train_runtime": 65680.9091,
      "train_tokens_per_second": 12167.781
    },
    {
      "epoch": 1.6886614597360625,
      "grad_norm": 0.7205528364107072,
      "learning_rate": 5.863493547065513e-07,
      "loss": 0.28240170478820803,
      "num_input_tokens_seen": 799349656,
      "step": 25080,
      "train_runtime": 65693.7333,
      "train_tokens_per_second": 12167.822
    },
    {
      "epoch": 1.6889981147320228,
      "grad_norm": 0.7380578058180574,
      "learning_rate": 5.851075697152858e-07,
      "loss": 0.2742695093154907,
      "num_input_tokens_seen": 799512880,
      "step": 25085,
      "train_runtime": 65706.6368,
      "train_tokens_per_second": 12167.917
    },
    {
      "epoch": 1.6893347697279828,
      "grad_norm": 0.8490158962466555,
      "learning_rate": 5.838670193346118e-07,
      "loss": 0.30379667282104494,
      "num_input_tokens_seen": 799669112,
      "step": 25090,
      "train_runtime": 65719.2175,
      "train_tokens_per_second": 12167.965
    },
    {
      "epoch": 1.6896714247239428,
      "grad_norm": 0.8154353187052651,
      "learning_rate": 5.826277039114426e-07,
      "loss": 0.27211599349975585,
      "num_input_tokens_seen": 799831752,
      "step": 25095,
      "train_runtime": 65731.9464,
      "train_tokens_per_second": 12168.083
    },
    {
      "epoch": 1.690008079719903,
      "grad_norm": 0.8059264446728833,
      "learning_rate": 5.813896237923505e-07,
      "loss": 0.3278991222381592,
      "num_input_tokens_seen": 799993864,
      "step": 25100,
      "train_runtime": 65744.4972,
      "train_tokens_per_second": 12168.225
    },
    {
      "epoch": 1.6903447347158633,
      "grad_norm": 0.800496170335914,
      "learning_rate": 5.801527793235606e-07,
      "loss": 0.30397605895996094,
      "num_input_tokens_seen": 800156864,
      "step": 25105,
      "train_runtime": 65757.3347,
      "train_tokens_per_second": 12168.329
    },
    {
      "epoch": 1.6906813897118234,
      "grad_norm": 0.7501265440937339,
      "learning_rate": 5.789171708509516e-07,
      "loss": 0.285453724861145,
      "num_input_tokens_seen": 800314512,
      "step": 25110,
      "train_runtime": 65771.4995,
      "train_tokens_per_second": 12168.105
    },
    {
      "epoch": 1.6910180447077834,
      "grad_norm": 0.7684838344201916,
      "learning_rate": 5.776827987200584e-07,
      "loss": 0.307835054397583,
      "num_input_tokens_seen": 800473488,
      "step": 25115,
      "train_runtime": 65784.5323,
      "train_tokens_per_second": 12168.111
    },
    {
      "epoch": 1.6913546997037436,
      "grad_norm": 0.8267263091646561,
      "learning_rate": 5.764496632760691e-07,
      "loss": 0.274769926071167,
      "num_input_tokens_seen": 800627016,
      "step": 25120,
      "train_runtime": 65797.2195,
      "train_tokens_per_second": 12168.098
    },
    {
      "epoch": 1.6916913546997039,
      "grad_norm": 0.7602366286621726,
      "learning_rate": 5.752177648638241e-07,
      "loss": 0.2738208770751953,
      "num_input_tokens_seen": 800785592,
      "step": 25125,
      "train_runtime": 65809.8771,
      "train_tokens_per_second": 12168.167
    },
    {
      "epoch": 1.692028009695664,
      "grad_norm": 0.8855191529361517,
      "learning_rate": 5.739871038278239e-07,
      "loss": 0.3088421106338501,
      "num_input_tokens_seen": 800942568,
      "step": 25130,
      "train_runtime": 65822.8847,
      "train_tokens_per_second": 12168.147
    },
    {
      "epoch": 1.692364664691624,
      "grad_norm": 0.7763642930103014,
      "learning_rate": 5.727576805122142e-07,
      "loss": 0.2768913507461548,
      "num_input_tokens_seen": 801101544,
      "step": 25135,
      "train_runtime": 65836.1083,
      "train_tokens_per_second": 12168.118
    },
    {
      "epoch": 1.6927013196875842,
      "grad_norm": 0.7696662641219881,
      "learning_rate": 5.715294952608041e-07,
      "loss": 0.2835252285003662,
      "num_input_tokens_seen": 801264232,
      "step": 25140,
      "train_runtime": 65848.7502,
      "train_tokens_per_second": 12168.253
    },
    {
      "epoch": 1.6930379746835444,
      "grad_norm": 0.9588288462807922,
      "learning_rate": 5.703025484170471e-07,
      "loss": 0.3243107795715332,
      "num_input_tokens_seen": 801425240,
      "step": 25145,
      "train_runtime": 65861.2258,
      "train_tokens_per_second": 12168.392
    },
    {
      "epoch": 1.6933746296795045,
      "grad_norm": 0.8809431854390011,
      "learning_rate": 5.690768403240587e-07,
      "loss": 0.3107900857925415,
      "num_input_tokens_seen": 801587016,
      "step": 25150,
      "train_runtime": 65875.0499,
      "train_tokens_per_second": 12168.295
    },
    {
      "epoch": 1.6937112846754645,
      "grad_norm": 0.7938221178191538,
      "learning_rate": 5.67852371324602e-07,
      "loss": 0.30399086475372317,
      "num_input_tokens_seen": 801750680,
      "step": 25155,
      "train_runtime": 65888.1034,
      "train_tokens_per_second": 12168.368
    },
    {
      "epoch": 1.6940479396714248,
      "grad_norm": 0.8069563375262169,
      "learning_rate": 5.66629141761097e-07,
      "loss": 0.2899326324462891,
      "num_input_tokens_seen": 801911128,
      "step": 25160,
      "train_runtime": 65900.5449,
      "train_tokens_per_second": 12168.505
    },
    {
      "epoch": 1.694384594667385,
      "grad_norm": 0.7997407532466363,
      "learning_rate": 5.654071519756154e-07,
      "loss": 0.29592466354370117,
      "num_input_tokens_seen": 802074056,
      "step": 25165,
      "train_runtime": 65913.4002,
      "train_tokens_per_second": 12168.604
    },
    {
      "epoch": 1.694721249663345,
      "grad_norm": 0.8217694503633814,
      "learning_rate": 5.641864023098831e-07,
      "loss": 0.3291255712509155,
      "num_input_tokens_seen": 802235128,
      "step": 25170,
      "train_runtime": 65926.9011,
      "train_tokens_per_second": 12168.555
    },
    {
      "epoch": 1.695057904659305,
      "grad_norm": 0.792292176665687,
      "learning_rate": 5.629668931052773e-07,
      "loss": 0.3054348468780518,
      "num_input_tokens_seen": 802398624,
      "step": 25175,
      "train_runtime": 65939.9123,
      "train_tokens_per_second": 12168.633
    },
    {
      "epoch": 1.6953945596552653,
      "grad_norm": 0.8185731335853207,
      "learning_rate": 5.61748624702833e-07,
      "loss": 0.3245995044708252,
      "num_input_tokens_seen": 802558256,
      "step": 25180,
      "train_runtime": 65952.4274,
      "train_tokens_per_second": 12168.745
    },
    {
      "epoch": 1.6957312146512256,
      "grad_norm": 0.8777389401390597,
      "learning_rate": 5.605315974432313e-07,
      "loss": 0.3108147144317627,
      "num_input_tokens_seen": 802718096,
      "step": 25185,
      "train_runtime": 65965.1012,
      "train_tokens_per_second": 12168.83
    },
    {
      "epoch": 1.6960678696471856,
      "grad_norm": 0.756552154977697,
      "learning_rate": 5.593158116668146e-07,
      "loss": 0.2891570568084717,
      "num_input_tokens_seen": 802881936,
      "step": 25190,
      "train_runtime": 65977.544,
      "train_tokens_per_second": 12169.018
    },
    {
      "epoch": 1.6964045246431456,
      "grad_norm": 0.8816252492366896,
      "learning_rate": 5.581012677135683e-07,
      "loss": 0.29001004695892335,
      "num_input_tokens_seen": 803032584,
      "step": 25195,
      "train_runtime": 65991.0029,
      "train_tokens_per_second": 12168.819
    },
    {
      "epoch": 1.6967411796391059,
      "grad_norm": 0.8250004805569026,
      "learning_rate": 5.56887965923139e-07,
      "loss": 0.3027050971984863,
      "num_input_tokens_seen": 803193512,
      "step": 25200,
      "train_runtime": 66003.4904,
      "train_tokens_per_second": 12168.955
    },
    {
      "epoch": 1.6970778346350661,
      "grad_norm": 0.8384060628920201,
      "learning_rate": 5.556759066348227e-07,
      "loss": 0.29056997299194337,
      "num_input_tokens_seen": 803353376,
      "step": 25205,
      "train_runtime": 66015.9885,
      "train_tokens_per_second": 12169.073
    },
    {
      "epoch": 1.6974144896310261,
      "grad_norm": 0.7681827073711663,
      "learning_rate": 5.54465090187567e-07,
      "loss": 0.3030840396881104,
      "num_input_tokens_seen": 803512024,
      "step": 25210,
      "train_runtime": 66029.304,
      "train_tokens_per_second": 12169.022
    },
    {
      "epoch": 1.6977511446269862,
      "grad_norm": 0.795024711685822,
      "learning_rate": 5.532555169199727e-07,
      "loss": 0.28143887519836425,
      "num_input_tokens_seen": 803667336,
      "step": 25215,
      "train_runtime": 66042.6213,
      "train_tokens_per_second": 12168.919
    },
    {
      "epoch": 1.6980877996229464,
      "grad_norm": 0.7589717236028772,
      "learning_rate": 5.520471871702937e-07,
      "loss": 0.2832162618637085,
      "num_input_tokens_seen": 803827896,
      "step": 25220,
      "train_runtime": 66055.7154,
      "train_tokens_per_second": 12168.938
    },
    {
      "epoch": 1.6984244546189067,
      "grad_norm": 0.8627059760443402,
      "learning_rate": 5.508401012764341e-07,
      "loss": 0.2953542709350586,
      "num_input_tokens_seen": 803985080,
      "step": 25225,
      "train_runtime": 66069.0204,
      "train_tokens_per_second": 12168.866
    },
    {
      "epoch": 1.6987611096148667,
      "grad_norm": 0.75552966329654,
      "learning_rate": 5.496342595759552e-07,
      "loss": 0.268917179107666,
      "num_input_tokens_seen": 804148920,
      "step": 25230,
      "train_runtime": 66081.446,
      "train_tokens_per_second": 12169.058
    },
    {
      "epoch": 1.6990977646108267,
      "grad_norm": 0.7520443819579034,
      "learning_rate": 5.484296624060626e-07,
      "loss": 0.2827796220779419,
      "num_input_tokens_seen": 804311176,
      "step": 25235,
      "train_runtime": 66094.0401,
      "train_tokens_per_second": 12169.194
    },
    {
      "epoch": 1.699434419606787,
      "grad_norm": 0.793656659551414,
      "learning_rate": 5.472263101036212e-07,
      "loss": 0.29239912033081056,
      "num_input_tokens_seen": 804472584,
      "step": 25240,
      "train_runtime": 66107.3031,
      "train_tokens_per_second": 12169.194
    },
    {
      "epoch": 1.6997710746027472,
      "grad_norm": 0.9828065726238588,
      "learning_rate": 5.460242030051421e-07,
      "loss": 0.3105603218078613,
      "num_input_tokens_seen": 804631344,
      "step": 25245,
      "train_runtime": 66120.7819,
      "train_tokens_per_second": 12169.114
    },
    {
      "epoch": 1.7001077295987073,
      "grad_norm": 0.7515100376319965,
      "learning_rate": 5.448233414467924e-07,
      "loss": 0.30887694358825685,
      "num_input_tokens_seen": 804794872,
      "step": 25250,
      "train_runtime": 66133.8209,
      "train_tokens_per_second": 12169.188
    },
    {
      "epoch": 1.7004443845946673,
      "grad_norm": 0.7532891696818705,
      "learning_rate": 5.43623725764389e-07,
      "loss": 0.31564314365386964,
      "num_input_tokens_seen": 804956656,
      "step": 25255,
      "train_runtime": 66146.8369,
      "train_tokens_per_second": 12169.239
    },
    {
      "epoch": 1.7007810395906275,
      "grad_norm": 0.7971741527408678,
      "learning_rate": 5.424253562934001e-07,
      "loss": 0.28030054569244384,
      "num_input_tokens_seen": 805119344,
      "step": 25260,
      "train_runtime": 66159.4874,
      "train_tokens_per_second": 12169.371
    },
    {
      "epoch": 1.7011176945865878,
      "grad_norm": 0.79111863716471,
      "learning_rate": 5.412282333689461e-07,
      "loss": 0.30404462814331057,
      "num_input_tokens_seen": 805280840,
      "step": 25265,
      "train_runtime": 66173.1564,
      "train_tokens_per_second": 12169.298
    },
    {
      "epoch": 1.7014543495825478,
      "grad_norm": 0.8126911369748564,
      "learning_rate": 5.400323573257987e-07,
      "loss": 0.29868664741516116,
      "num_input_tokens_seen": 805442984,
      "step": 25270,
      "train_runtime": 66186.3881,
      "train_tokens_per_second": 12169.315
    },
    {
      "epoch": 1.7017910045785078,
      "grad_norm": 0.7531648950013857,
      "learning_rate": 5.388377284983803e-07,
      "loss": 0.2803764581680298,
      "num_input_tokens_seen": 805602968,
      "step": 25275,
      "train_runtime": 66199.1101,
      "train_tokens_per_second": 12169.393
    },
    {
      "epoch": 1.702127659574468,
      "grad_norm": 0.8686858212633283,
      "learning_rate": 5.37644347220767e-07,
      "loss": 0.3020791053771973,
      "num_input_tokens_seen": 805760112,
      "step": 25280,
      "train_runtime": 66211.7765,
      "train_tokens_per_second": 12169.438
    },
    {
      "epoch": 1.7024643145704283,
      "grad_norm": 0.8803022160278495,
      "learning_rate": 5.364522138266809e-07,
      "loss": 0.296372127532959,
      "num_input_tokens_seen": 805922272,
      "step": 25285,
      "train_runtime": 66224.3722,
      "train_tokens_per_second": 12169.572
    },
    {
      "epoch": 1.7028009695663884,
      "grad_norm": 0.8155451927873711,
      "learning_rate": 5.352613286495018e-07,
      "loss": 0.29341325759887693,
      "num_input_tokens_seen": 806077168,
      "step": 25290,
      "train_runtime": 66237.3282,
      "train_tokens_per_second": 12169.53
    },
    {
      "epoch": 1.7031376245623484,
      "grad_norm": 0.7520090991767232,
      "learning_rate": 5.340716920222527e-07,
      "loss": 0.2737943172454834,
      "num_input_tokens_seen": 806240136,
      "step": 25295,
      "train_runtime": 66250.3744,
      "train_tokens_per_second": 12169.594
    },
    {
      "epoch": 1.7034742795583087,
      "grad_norm": 0.962050645372596,
      "learning_rate": 5.32883304277616e-07,
      "loss": 0.3186910629272461,
      "num_input_tokens_seen": 806397272,
      "step": 25300,
      "train_runtime": 66263.47,
      "train_tokens_per_second": 12169.56
    },
    {
      "epoch": 1.703810934554269,
      "grad_norm": 0.8852503516110072,
      "learning_rate": 5.31696165747918e-07,
      "loss": 0.2684332370758057,
      "num_input_tokens_seen": 806557824,
      "step": 25305,
      "train_runtime": 66276.883,
      "train_tokens_per_second": 12169.52
    },
    {
      "epoch": 1.704147589550229,
      "grad_norm": 0.8404307815649437,
      "learning_rate": 5.305102767651393e-07,
      "loss": 0.29191365242004397,
      "num_input_tokens_seen": 806721296,
      "step": 25310,
      "train_runtime": 66289.3267,
      "train_tokens_per_second": 12169.701
    },
    {
      "epoch": 1.704484244546189,
      "grad_norm": 0.846110243738861,
      "learning_rate": 5.293256376609091e-07,
      "loss": 0.30517499446868895,
      "num_input_tokens_seen": 806881824,
      "step": 25315,
      "train_runtime": 66302.0578,
      "train_tokens_per_second": 12169.786
    },
    {
      "epoch": 1.7048208995421492,
      "grad_norm": 0.7807440591766549,
      "learning_rate": 5.281422487665089e-07,
      "loss": 0.28216400146484377,
      "num_input_tokens_seen": 807045664,
      "step": 25320,
      "train_runtime": 66314.498,
      "train_tokens_per_second": 12169.973
    },
    {
      "epoch": 1.7051575545381095,
      "grad_norm": 0.8635260531680488,
      "learning_rate": 5.269601104128674e-07,
      "loss": 0.2909599542617798,
      "num_input_tokens_seen": 807209504,
      "step": 25325,
      "train_runtime": 66326.9527,
      "train_tokens_per_second": 12170.158
    },
    {
      "epoch": 1.7054942095340695,
      "grad_norm": 0.7950375112966951,
      "learning_rate": 5.257792229305702e-07,
      "loss": 0.31408185958862306,
      "num_input_tokens_seen": 807366936,
      "step": 25330,
      "train_runtime": 66339.6443,
      "train_tokens_per_second": 12170.203
    },
    {
      "epoch": 1.7058308645300295,
      "grad_norm": 0.8419617185275647,
      "learning_rate": 5.245995866498438e-07,
      "loss": 0.2757680416107178,
      "num_input_tokens_seen": 807525288,
      "step": 25335,
      "train_runtime": 66352.277,
      "train_tokens_per_second": 12170.272
    },
    {
      "epoch": 1.7061675195259898,
      "grad_norm": 0.8276594986765249,
      "learning_rate": 5.234212019005741e-07,
      "loss": 0.3080322265625,
      "num_input_tokens_seen": 807682488,
      "step": 25340,
      "train_runtime": 66365.6176,
      "train_tokens_per_second": 12170.195
    },
    {
      "epoch": 1.70650417452195,
      "grad_norm": 0.8101383319675212,
      "learning_rate": 5.222440690122882e-07,
      "loss": 0.2870373010635376,
      "num_input_tokens_seen": 807841080,
      "step": 25345,
      "train_runtime": 66378.7291,
      "train_tokens_per_second": 12170.18
    },
    {
      "epoch": 1.70684082951791,
      "grad_norm": 0.7850745560927895,
      "learning_rate": 5.210681883141716e-07,
      "loss": 0.2790995121002197,
      "num_input_tokens_seen": 808002272,
      "step": 25350,
      "train_runtime": 66391.5753,
      "train_tokens_per_second": 12170.253
    },
    {
      "epoch": 1.70717748451387,
      "grad_norm": 0.8957888916427711,
      "learning_rate": 5.19893560135053e-07,
      "loss": 0.290134072303772,
      "num_input_tokens_seen": 808163320,
      "step": 25355,
      "train_runtime": 66404.0398,
      "train_tokens_per_second": 12170.394
    },
    {
      "epoch": 1.7075141395098303,
      "grad_norm": 0.7858696037693033,
      "learning_rate": 5.187201848034146e-07,
      "loss": 0.28968205451965334,
      "num_input_tokens_seen": 808322728,
      "step": 25360,
      "train_runtime": 66417.5816,
      "train_tokens_per_second": 12170.313
    },
    {
      "epoch": 1.7078507945057906,
      "grad_norm": 0.8550267368462614,
      "learning_rate": 5.175480626473873e-07,
      "loss": 0.28513302803039553,
      "num_input_tokens_seen": 808480216,
      "step": 25365,
      "train_runtime": 66430.0339,
      "train_tokens_per_second": 12170.402
    },
    {
      "epoch": 1.7081874495017506,
      "grad_norm": 0.8828328480467336,
      "learning_rate": 5.163771939947504e-07,
      "loss": 0.3155116081237793,
      "num_input_tokens_seen": 808639720,
      "step": 25370,
      "train_runtime": 66442.7997,
      "train_tokens_per_second": 12170.464
    },
    {
      "epoch": 1.7085241044977106,
      "grad_norm": 0.7612968275476691,
      "learning_rate": 5.152075791729327e-07,
      "loss": 0.32192938327789306,
      "num_input_tokens_seen": 808801456,
      "step": 25375,
      "train_runtime": 66455.2734,
      "train_tokens_per_second": 12170.614
    },
    {
      "epoch": 1.7088607594936709,
      "grad_norm": 0.7431298964019118,
      "learning_rate": 5.140392185090165e-07,
      "loss": 0.28952252864837646,
      "num_input_tokens_seen": 808957096,
      "step": 25380,
      "train_runtime": 66469.1381,
      "train_tokens_per_second": 12170.417
    },
    {
      "epoch": 1.7091974144896311,
      "grad_norm": 1.0990000083010292,
      "learning_rate": 5.128721123297265e-07,
      "loss": 0.30175418853759767,
      "num_input_tokens_seen": 809120936,
      "step": 25385,
      "train_runtime": 66481.5743,
      "train_tokens_per_second": 12170.604
    },
    {
      "epoch": 1.7095340694855912,
      "grad_norm": 0.8230642088537091,
      "learning_rate": 5.117062609614438e-07,
      "loss": 0.2878246784210205,
      "num_input_tokens_seen": 809280272,
      "step": 25390,
      "train_runtime": 66495.3832,
      "train_tokens_per_second": 12170.473
    },
    {
      "epoch": 1.7098707244815512,
      "grad_norm": 0.8900825345581529,
      "learning_rate": 5.105416647301909e-07,
      "loss": 0.2841442584991455,
      "num_input_tokens_seen": 809441840,
      "step": 25395,
      "train_runtime": 66507.9221,
      "train_tokens_per_second": 12170.608
    },
    {
      "epoch": 1.7102073794775114,
      "grad_norm": 0.791905554161877,
      "learning_rate": 5.093783239616468e-07,
      "loss": 0.2772679328918457,
      "num_input_tokens_seen": 809598632,
      "step": 25400,
      "train_runtime": 66521.2847,
      "train_tokens_per_second": 12170.52
    },
    {
      "epoch": 1.7105440344734717,
      "grad_norm": 0.9200165243376989,
      "learning_rate": 5.082162389811352e-07,
      "loss": 0.27681775093078614,
      "num_input_tokens_seen": 809753656,
      "step": 25405,
      "train_runtime": 66534.1817,
      "train_tokens_per_second": 12170.491
    },
    {
      "epoch": 1.7108806894694317,
      "grad_norm": 1.5782738080737202,
      "learning_rate": 5.070554101136288e-07,
      "loss": 0.2743373394012451,
      "num_input_tokens_seen": 809906648,
      "step": 25410,
      "train_runtime": 66548.2509,
      "train_tokens_per_second": 12170.217
    },
    {
      "epoch": 1.7112173444653918,
      "grad_norm": 0.7513323227337197,
      "learning_rate": 5.058958376837497e-07,
      "loss": 0.2892336845397949,
      "num_input_tokens_seen": 810070488,
      "step": 25415,
      "train_runtime": 66560.6932,
      "train_tokens_per_second": 12170.403
    },
    {
      "epoch": 1.711553999461352,
      "grad_norm": 0.8138054579967886,
      "learning_rate": 5.047375220157691e-07,
      "loss": 0.2978404998779297,
      "num_input_tokens_seen": 810232992,
      "step": 25420,
      "train_runtime": 66574.0436,
      "train_tokens_per_second": 12170.404
    },
    {
      "epoch": 1.7118906544573123,
      "grad_norm": 0.7693054586269306,
      "learning_rate": 5.035804634336056e-07,
      "loss": 0.309088921546936,
      "num_input_tokens_seen": 810396688,
      "step": 25425,
      "train_runtime": 66587.1051,
      "train_tokens_per_second": 12170.475
    },
    {
      "epoch": 1.7122273094532723,
      "grad_norm": 0.7561704469377175,
      "learning_rate": 5.024246622608292e-07,
      "loss": 0.28566803932189944,
      "num_input_tokens_seen": 810560528,
      "step": 25430,
      "train_runtime": 66599.5478,
      "train_tokens_per_second": 12170.661
    },
    {
      "epoch": 1.7125639644492323,
      "grad_norm": 0.8608063236563862,
      "learning_rate": 5.01270118820652e-07,
      "loss": 0.3105528116226196,
      "num_input_tokens_seen": 810722896,
      "step": 25435,
      "train_runtime": 66612.1159,
      "train_tokens_per_second": 12170.802
    },
    {
      "epoch": 1.7129006194451926,
      "grad_norm": 0.8956766546025756,
      "learning_rate": 5.001168334359429e-07,
      "loss": 0.3243201971054077,
      "num_input_tokens_seen": 810883976,
      "step": 25440,
      "train_runtime": 66625.0262,
      "train_tokens_per_second": 12170.862
    },
    {
      "epoch": 1.7132372744411528,
      "grad_norm": 0.7532341570616985,
      "learning_rate": 4.989648064292102e-07,
      "loss": 0.2967890024185181,
      "num_input_tokens_seen": 811044848,
      "step": 25445,
      "train_runtime": 66637.9197,
      "train_tokens_per_second": 12170.921
    },
    {
      "epoch": 1.7135739294371128,
      "grad_norm": 0.873970233887233,
      "learning_rate": 4.978140381226177e-07,
      "loss": 0.3010232925415039,
      "num_input_tokens_seen": 811204952,
      "step": 25450,
      "train_runtime": 66650.793,
      "train_tokens_per_second": 12170.972
    },
    {
      "epoch": 1.7139105844330729,
      "grad_norm": 0.8014722503143785,
      "learning_rate": 4.966645288379729e-07,
      "loss": 0.27742342948913573,
      "num_input_tokens_seen": 811367616,
      "step": 25455,
      "train_runtime": 66664.1954,
      "train_tokens_per_second": 12170.965
    },
    {
      "epoch": 1.7142472394290331,
      "grad_norm": 0.8750279199355134,
      "learning_rate": 4.955162788967327e-07,
      "loss": 0.31605224609375,
      "num_input_tokens_seen": 811516008,
      "step": 25460,
      "train_runtime": 66678.2276,
      "train_tokens_per_second": 12170.63
    },
    {
      "epoch": 1.7145838944249934,
      "grad_norm": 0.721234063527085,
      "learning_rate": 4.943692886200013e-07,
      "loss": 0.3078624248504639,
      "num_input_tokens_seen": 811675352,
      "step": 25465,
      "train_runtime": 66691.488,
      "train_tokens_per_second": 12170.599
    },
    {
      "epoch": 1.7149205494209534,
      "grad_norm": 0.7563294229928634,
      "learning_rate": 4.932235583285305e-07,
      "loss": 0.2949864149093628,
      "num_input_tokens_seen": 811838784,
      "step": 25470,
      "train_runtime": 66704.4948,
      "train_tokens_per_second": 12170.676
    },
    {
      "epoch": 1.7152572044169134,
      "grad_norm": 0.8039529212651666,
      "learning_rate": 4.920790883427201e-07,
      "loss": 0.2986870288848877,
      "num_input_tokens_seen": 811996664,
      "step": 25475,
      "train_runtime": 66717.6808,
      "train_tokens_per_second": 12170.637
    },
    {
      "epoch": 1.7155938594128737,
      "grad_norm": 0.7985136356593623,
      "learning_rate": 4.909358789826196e-07,
      "loss": 0.30323760509490966,
      "num_input_tokens_seen": 812152672,
      "step": 25480,
      "train_runtime": 66731.6068,
      "train_tokens_per_second": 12170.435
    },
    {
      "epoch": 1.715930514408834,
      "grad_norm": 0.8711067618686908,
      "learning_rate": 4.897939305679206e-07,
      "loss": 0.3026529550552368,
      "num_input_tokens_seen": 812313928,
      "step": 25485,
      "train_runtime": 66744.0954,
      "train_tokens_per_second": 12170.574
    },
    {
      "epoch": 1.716267169404794,
      "grad_norm": 0.7375686367190549,
      "learning_rate": 4.886532434179686e-07,
      "loss": 0.29832355976104735,
      "num_input_tokens_seen": 812475168,
      "step": 25490,
      "train_runtime": 66757.0145,
      "train_tokens_per_second": 12170.634
    },
    {
      "epoch": 1.716603824400754,
      "grad_norm": 0.7764938970656083,
      "learning_rate": 4.875138178517497e-07,
      "loss": 0.25198540687561033,
      "num_input_tokens_seen": 812632704,
      "step": 25495,
      "train_runtime": 66769.8871,
      "train_tokens_per_second": 12170.647
    },
    {
      "epoch": 1.7169404793967142,
      "grad_norm": 0.841551135718732,
      "learning_rate": 4.863756541879028e-07,
      "loss": 0.315032958984375,
      "num_input_tokens_seen": 812796136,
      "step": 25500,
      "train_runtime": 66782.8999,
      "train_tokens_per_second": 12170.722
    },
    {
      "epoch": 1.7172771343926745,
      "grad_norm": 0.7432246312031906,
      "learning_rate": 4.852387527447117e-07,
      "loss": 0.28313241004943845,
      "num_input_tokens_seen": 812959976,
      "step": 25505,
      "train_runtime": 66795.3476,
      "train_tokens_per_second": 12170.907
    },
    {
      "epoch": 1.7176137893886345,
      "grad_norm": 0.7577631824245612,
      "learning_rate": 4.84103113840107e-07,
      "loss": 0.2972623348236084,
      "num_input_tokens_seen": 813122776,
      "step": 25510,
      "train_runtime": 66808.1767,
      "train_tokens_per_second": 12171.007
    },
    {
      "epoch": 1.7179504443845945,
      "grad_norm": 0.7883606567919665,
      "learning_rate": 4.829687377916664e-07,
      "loss": 0.29777188301086427,
      "num_input_tokens_seen": 813276304,
      "step": 25515,
      "train_runtime": 66821.563,
      "train_tokens_per_second": 12170.866
    },
    {
      "epoch": 1.7182870993805548,
      "grad_norm": 0.8435313346864439,
      "learning_rate": 4.81835624916614e-07,
      "loss": 0.27108373641967776,
      "num_input_tokens_seen": 813434048,
      "step": 25520,
      "train_runtime": 66834.3843,
      "train_tokens_per_second": 12170.892
    },
    {
      "epoch": 1.718623754376515,
      "grad_norm": 0.9132789679859118,
      "learning_rate": 4.807037755318211e-07,
      "loss": 0.29337828159332274,
      "num_input_tokens_seen": 813597104,
      "step": 25525,
      "train_runtime": 66847.4783,
      "train_tokens_per_second": 12170.947
    },
    {
      "epoch": 1.718960409372475,
      "grad_norm": 0.7058141725616477,
      "learning_rate": 4.795731899538086e-07,
      "loss": 0.30762457847595215,
      "num_input_tokens_seen": 813755728,
      "step": 25530,
      "train_runtime": 66860.8611,
      "train_tokens_per_second": 12170.883
    },
    {
      "epoch": 1.719297064368435,
      "grad_norm": 0.7380084515143684,
      "learning_rate": 4.78443868498737e-07,
      "loss": 0.26756463050842283,
      "num_input_tokens_seen": 813917880,
      "step": 25535,
      "train_runtime": 66873.4495,
      "train_tokens_per_second": 12171.017
    },
    {
      "epoch": 1.7196337193643954,
      "grad_norm": 0.8355686413790112,
      "learning_rate": 4.773158114824206e-07,
      "loss": 0.262506365776062,
      "num_input_tokens_seen": 814078216,
      "step": 25540,
      "train_runtime": 66887.1137,
      "train_tokens_per_second": 12170.928
    },
    {
      "epoch": 1.7199703743603556,
      "grad_norm": 0.8208269891522171,
      "learning_rate": 4.7618901922031445e-07,
      "loss": 0.2658960819244385,
      "num_input_tokens_seen": 814241224,
      "step": 25545,
      "train_runtime": 66901.3357,
      "train_tokens_per_second": 12170.777
    },
    {
      "epoch": 1.7203070293563156,
      "grad_norm": 0.8324669074118888,
      "learning_rate": 4.750634920275243e-07,
      "loss": 0.29631292819976807,
      "num_input_tokens_seen": 814399216,
      "step": 25550,
      "train_runtime": 66914.2267,
      "train_tokens_per_second": 12170.793
    },
    {
      "epoch": 1.7206436843522757,
      "grad_norm": 0.8552150436160211,
      "learning_rate": 4.739392302188001e-07,
      "loss": 0.31460275650024416,
      "num_input_tokens_seen": 814561632,
      "step": 25555,
      "train_runtime": 66927.1167,
      "train_tokens_per_second": 12170.876
    },
    {
      "epoch": 1.720980339348236,
      "grad_norm": 0.7958254794931157,
      "learning_rate": 4.728162341085368e-07,
      "loss": 0.299389386177063,
      "num_input_tokens_seen": 814723608,
      "step": 25560,
      "train_runtime": 66940.9859,
      "train_tokens_per_second": 12170.774
    },
    {
      "epoch": 1.7213169943441962,
      "grad_norm": 0.7394302930375213,
      "learning_rate": 4.7169450401077843e-07,
      "loss": 0.2992805004119873,
      "num_input_tokens_seen": 814885680,
      "step": 25565,
      "train_runtime": 66953.5686,
      "train_tokens_per_second": 12170.907
    },
    {
      "epoch": 1.7216536493401562,
      "grad_norm": 0.827383990258824,
      "learning_rate": 4.7057404023921174e-07,
      "loss": 0.2992172956466675,
      "num_input_tokens_seen": 815048136,
      "step": 25570,
      "train_runtime": 66966.8292,
      "train_tokens_per_second": 12170.923
    },
    {
      "epoch": 1.7219903043361162,
      "grad_norm": 0.8310427015417835,
      "learning_rate": 4.694548431071705e-07,
      "loss": 0.29741454124450684,
      "num_input_tokens_seen": 815205896,
      "step": 25575,
      "train_runtime": 66979.6326,
      "train_tokens_per_second": 12170.952
    },
    {
      "epoch": 1.7223269593320765,
      "grad_norm": 0.8117229033236854,
      "learning_rate": 4.683369129276377e-07,
      "loss": 0.29701454639434816,
      "num_input_tokens_seen": 815364968,
      "step": 25580,
      "train_runtime": 66992.5503,
      "train_tokens_per_second": 12170.98
    },
    {
      "epoch": 1.7226636143280367,
      "grad_norm": 0.7097050482395375,
      "learning_rate": 4.6722025001323444e-07,
      "loss": 0.2651503562927246,
      "num_input_tokens_seen": 815520632,
      "step": 25585,
      "train_runtime": 67006.0712,
      "train_tokens_per_second": 12170.847
    },
    {
      "epoch": 1.7230002693239967,
      "grad_norm": 0.8267504674703496,
      "learning_rate": 4.6610485467623643e-07,
      "loss": 0.2984716176986694,
      "num_input_tokens_seen": 815675360,
      "step": 25590,
      "train_runtime": 67019.013,
      "train_tokens_per_second": 12170.805
    },
    {
      "epoch": 1.7233369243199568,
      "grad_norm": 0.7963204100521549,
      "learning_rate": 4.649907272285564e-07,
      "loss": 0.27908778190612793,
      "num_input_tokens_seen": 815836232,
      "step": 25595,
      "train_runtime": 67031.8556,
      "train_tokens_per_second": 12170.873
    },
    {
      "epoch": 1.723673579315917,
      "grad_norm": 0.7782328087253992,
      "learning_rate": 4.638778679817596e-07,
      "loss": 0.28272247314453125,
      "num_input_tokens_seen": 815999424,
      "step": 25600,
      "train_runtime": 67044.741,
      "train_tokens_per_second": 12170.968
    },
    {
      "epoch": 1.7240102343118773,
      "grad_norm": 0.7704604555854352,
      "learning_rate": 4.6276627724705204e-07,
      "loss": 0.30708799362182615,
      "num_input_tokens_seen": 816159296,
      "step": 25605,
      "train_runtime": 67057.7622,
      "train_tokens_per_second": 12170.989
    },
    {
      "epoch": 1.7243468893078373,
      "grad_norm": 0.7417429083057651,
      "learning_rate": 4.616559553352873e-07,
      "loss": 0.2608830213546753,
      "num_input_tokens_seen": 816317360,
      "step": 25610,
      "train_runtime": 67070.5189,
      "train_tokens_per_second": 12171.031
    },
    {
      "epoch": 1.7246835443037973,
      "grad_norm": 0.8246333214622045,
      "learning_rate": 4.6054690255696256e-07,
      "loss": 0.31869897842407224,
      "num_input_tokens_seen": 816480144,
      "step": 25615,
      "train_runtime": 67083.2476,
      "train_tokens_per_second": 12171.148
    },
    {
      "epoch": 1.7250201992997576,
      "grad_norm": 1.004632224475536,
      "learning_rate": 4.594391192222214e-07,
      "loss": 0.30283589363098146,
      "num_input_tokens_seen": 816638352,
      "step": 25620,
      "train_runtime": 67095.9416,
      "train_tokens_per_second": 12171.203
    },
    {
      "epoch": 1.7253568542957178,
      "grad_norm": 0.8824549256015667,
      "learning_rate": 4.583326056408505e-07,
      "loss": 0.27205395698547363,
      "num_input_tokens_seen": 816797808,
      "step": 25625,
      "train_runtime": 67109.0322,
      "train_tokens_per_second": 12171.205
    },
    {
      "epoch": 1.7256935092916779,
      "grad_norm": 0.9383321744551475,
      "learning_rate": 4.5722736212228613e-07,
      "loss": 0.3043744802474976,
      "num_input_tokens_seen": 816952200,
      "step": 25630,
      "train_runtime": 67123.0612,
      "train_tokens_per_second": 12170.962
    },
    {
      "epoch": 1.726030164287638,
      "grad_norm": 0.8515586862345872,
      "learning_rate": 4.5612338897560227e-07,
      "loss": 0.2992269039154053,
      "num_input_tokens_seen": 817114416,
      "step": 25635,
      "train_runtime": 67135.4917,
      "train_tokens_per_second": 12171.124
    },
    {
      "epoch": 1.7263668192835981,
      "grad_norm": 0.9510024916157301,
      "learning_rate": 4.5502068650952527e-07,
      "loss": 0.3066097259521484,
      "num_input_tokens_seen": 817270872,
      "step": 25640,
      "train_runtime": 67148.1475,
      "train_tokens_per_second": 12171.16
    },
    {
      "epoch": 1.7267034742795584,
      "grad_norm": 0.9014889193500454,
      "learning_rate": 4.5391925503241894e-07,
      "loss": 0.30767111778259276,
      "num_input_tokens_seen": 817433176,
      "step": 25645,
      "train_runtime": 67160.7108,
      "train_tokens_per_second": 12171.3
    },
    {
      "epoch": 1.7270401292755184,
      "grad_norm": 0.8593759445809093,
      "learning_rate": 4.5281909485229704e-07,
      "loss": 0.3000320911407471,
      "num_input_tokens_seen": 817591160,
      "step": 25650,
      "train_runtime": 67174.1501,
      "train_tokens_per_second": 12171.217
    },
    {
      "epoch": 1.7273767842714784,
      "grad_norm": 0.8592630782730303,
      "learning_rate": 4.51720206276815e-07,
      "loss": 0.26432290077209475,
      "num_input_tokens_seen": 817752544,
      "step": 25655,
      "train_runtime": 67187.4401,
      "train_tokens_per_second": 12171.211
    },
    {
      "epoch": 1.7277134392674387,
      "grad_norm": 0.8465124939822836,
      "learning_rate": 4.506225896132743e-07,
      "loss": 0.2770785570144653,
      "num_input_tokens_seen": 817911752,
      "step": 25660,
      "train_runtime": 67201.01,
      "train_tokens_per_second": 12171.123
    },
    {
      "epoch": 1.728050094263399,
      "grad_norm": 0.8548950233877984,
      "learning_rate": 4.4952624516861944e-07,
      "loss": 0.29408793449401854,
      "num_input_tokens_seen": 818065904,
      "step": 25665,
      "train_runtime": 67214.5879,
      "train_tokens_per_second": 12170.958
    },
    {
      "epoch": 1.728386749259359,
      "grad_norm": 0.7420863432531281,
      "learning_rate": 4.4843117324944005e-07,
      "loss": 0.24960010051727294,
      "num_input_tokens_seen": 818227848,
      "step": 25670,
      "train_runtime": 67227.1533,
      "train_tokens_per_second": 12171.092
    },
    {
      "epoch": 1.728723404255319,
      "grad_norm": 0.8375266988424531,
      "learning_rate": 4.4733737416196745e-07,
      "loss": 0.29548690319061277,
      "num_input_tokens_seen": 818373512,
      "step": 25675,
      "train_runtime": 67241.2185,
      "train_tokens_per_second": 12170.712
    },
    {
      "epoch": 1.7290600592512793,
      "grad_norm": 0.710513739553924,
      "learning_rate": 4.4624484821208346e-07,
      "loss": 0.27407007217407225,
      "num_input_tokens_seen": 818531256,
      "step": 25680,
      "train_runtime": 67255.011,
      "train_tokens_per_second": 12170.562
    },
    {
      "epoch": 1.7293967142472395,
      "grad_norm": 0.8486943347589415,
      "learning_rate": 4.451535957053044e-07,
      "loss": 0.26765587329864504,
      "num_input_tokens_seen": 818693656,
      "step": 25685,
      "train_runtime": 67268.4462,
      "train_tokens_per_second": 12170.545
    },
    {
      "epoch": 1.7297333692431995,
      "grad_norm": 0.8727696584447258,
      "learning_rate": 4.440636169467999e-07,
      "loss": 0.28300814628601073,
      "num_input_tokens_seen": 818852608,
      "step": 25690,
      "train_runtime": 67282.3951,
      "train_tokens_per_second": 12170.384
    },
    {
      "epoch": 1.7300700242391596,
      "grad_norm": 0.7513397510474625,
      "learning_rate": 4.429749122413757e-07,
      "loss": 0.2883636951446533,
      "num_input_tokens_seen": 819007600,
      "step": 25695,
      "train_runtime": 67295.5058,
      "train_tokens_per_second": 12170.316
    },
    {
      "epoch": 1.7304066792351198,
      "grad_norm": 0.8928870890971538,
      "learning_rate": 4.418874818934865e-07,
      "loss": 0.2849260807037354,
      "num_input_tokens_seen": 819169264,
      "step": 25700,
      "train_runtime": 67307.9851,
      "train_tokens_per_second": 12170.462
    },
    {
      "epoch": 1.73074333423108,
      "grad_norm": 0.8686688202366895,
      "learning_rate": 4.4080132620722806e-07,
      "loss": 0.3069554328918457,
      "num_input_tokens_seen": 819324672,
      "step": 25705,
      "train_runtime": 67322.2376,
      "train_tokens_per_second": 12170.194
    },
    {
      "epoch": 1.73107998922704,
      "grad_norm": 0.8427747531873772,
      "learning_rate": 4.3971644548634116e-07,
      "loss": 0.29826138019561765,
      "num_input_tokens_seen": 819482064,
      "step": 25710,
      "train_runtime": 67335.303,
      "train_tokens_per_second": 12170.17
    },
    {
      "epoch": 1.7314166442230001,
      "grad_norm": 0.9161377771271683,
      "learning_rate": 4.3863284003420826e-07,
      "loss": 0.31550326347351076,
      "num_input_tokens_seen": 819642976,
      "step": 25715,
      "train_runtime": 67348.3844,
      "train_tokens_per_second": 12170.195
    },
    {
      "epoch": 1.7317532992189604,
      "grad_norm": 0.777152507045464,
      "learning_rate": 4.375505101538563e-07,
      "loss": 0.2769683837890625,
      "num_input_tokens_seen": 819800136,
      "step": 25720,
      "train_runtime": 67361.1567,
      "train_tokens_per_second": 12170.221
    },
    {
      "epoch": 1.7320899542149206,
      "grad_norm": 0.8745893098275087,
      "learning_rate": 4.3646945614795487e-07,
      "loss": 0.30621986389160155,
      "num_input_tokens_seen": 819958688,
      "step": 25725,
      "train_runtime": 67373.6435,
      "train_tokens_per_second": 12170.318
    },
    {
      "epoch": 1.7324266092108807,
      "grad_norm": 0.8171291736706696,
      "learning_rate": 4.3538967831881995e-07,
      "loss": 0.3169607400894165,
      "num_input_tokens_seen": 820119296,
      "step": 25730,
      "train_runtime": 67386.5387,
      "train_tokens_per_second": 12170.373
    },
    {
      "epoch": 1.7327632642068407,
      "grad_norm": 0.8059567919709636,
      "learning_rate": 4.343111769684039e-07,
      "loss": 0.29121320247650145,
      "num_input_tokens_seen": 820272776,
      "step": 25735,
      "train_runtime": 67399.5266,
      "train_tokens_per_second": 12170.305
    },
    {
      "epoch": 1.733099919202801,
      "grad_norm": 0.791332015733899,
      "learning_rate": 4.332339523983103e-07,
      "loss": 0.2831947565078735,
      "num_input_tokens_seen": 820431768,
      "step": 25740,
      "train_runtime": 67412.1074,
      "train_tokens_per_second": 12170.392
    },
    {
      "epoch": 1.7334365741987612,
      "grad_norm": 0.8662824553085909,
      "learning_rate": 4.321580049097773e-07,
      "loss": 0.2866279125213623,
      "num_input_tokens_seen": 820587136,
      "step": 25745,
      "train_runtime": 67424.5104,
      "train_tokens_per_second": 12170.457
    },
    {
      "epoch": 1.7337732291947212,
      "grad_norm": 0.7858805360166149,
      "learning_rate": 4.310833348036941e-07,
      "loss": 0.27548966407775877,
      "num_input_tokens_seen": 820744656,
      "step": 25750,
      "train_runtime": 67436.9527,
      "train_tokens_per_second": 12170.548
    },
    {
      "epoch": 1.7341098841906812,
      "grad_norm": 0.8410071802132785,
      "learning_rate": 4.300099423805865e-07,
      "loss": 0.29560658931732176,
      "num_input_tokens_seen": 820901496,
      "step": 25755,
      "train_runtime": 67450.3545,
      "train_tokens_per_second": 12170.455
    },
    {
      "epoch": 1.7344465391866415,
      "grad_norm": 0.8135535352753432,
      "learning_rate": 4.289378279406264e-07,
      "loss": 0.2907859802246094,
      "num_input_tokens_seen": 821059128,
      "step": 25760,
      "train_runtime": 67462.8024,
      "train_tokens_per_second": 12170.546
    },
    {
      "epoch": 1.7347831941826017,
      "grad_norm": 0.8049621834232527,
      "learning_rate": 4.2786699178362647e-07,
      "loss": 0.2981605052947998,
      "num_input_tokens_seen": 821219216,
      "step": 25765,
      "train_runtime": 67476.1306,
      "train_tokens_per_second": 12170.514
    },
    {
      "epoch": 1.7351198491785618,
      "grad_norm": 0.846637612581395,
      "learning_rate": 4.2679743420904254e-07,
      "loss": 0.31073014736175536,
      "num_input_tokens_seen": 821379432,
      "step": 25770,
      "train_runtime": 67488.5688,
      "train_tokens_per_second": 12170.645
    },
    {
      "epoch": 1.7354565041745218,
      "grad_norm": 0.8686555687179883,
      "learning_rate": 4.257291555159726e-07,
      "loss": 0.26355729103088377,
      "num_input_tokens_seen": 821538192,
      "step": 25775,
      "train_runtime": 67501.7748,
      "train_tokens_per_second": 12170.616
    },
    {
      "epoch": 1.735793159170482,
      "grad_norm": 0.8296878955671245,
      "learning_rate": 4.246621560031594e-07,
      "loss": 0.28075995445251467,
      "num_input_tokens_seen": 821695936,
      "step": 25780,
      "train_runtime": 67515.5376,
      "train_tokens_per_second": 12170.472
    },
    {
      "epoch": 1.7361298141664423,
      "grad_norm": 0.8299509869445777,
      "learning_rate": 4.235964359689837e-07,
      "loss": 0.26936726570129393,
      "num_input_tokens_seen": 821856120,
      "step": 25785,
      "train_runtime": 67527.9666,
      "train_tokens_per_second": 12170.604
    },
    {
      "epoch": 1.7364664691624023,
      "grad_norm": 1.0026215744981537,
      "learning_rate": 4.225319957114726e-07,
      "loss": 0.305324387550354,
      "num_input_tokens_seen": 822012704,
      "step": 25790,
      "train_runtime": 67541.4757,
      "train_tokens_per_second": 12170.488
    },
    {
      "epoch": 1.7368031241583624,
      "grad_norm": 0.7833615752061763,
      "learning_rate": 4.214688355282909e-07,
      "loss": 0.27938241958618165,
      "num_input_tokens_seen": 822174648,
      "step": 25795,
      "train_runtime": 67555.4724,
      "train_tokens_per_second": 12170.363
    },
    {
      "epoch": 1.7371397791543226,
      "grad_norm": 0.9927631937715004,
      "learning_rate": 4.204069557167506e-07,
      "loss": 0.282896089553833,
      "num_input_tokens_seen": 822331984,
      "step": 25800,
      "train_runtime": 67568.6635,
      "train_tokens_per_second": 12170.316
    },
    {
      "epoch": 1.7374764341502829,
      "grad_norm": 0.9051055240597761,
      "learning_rate": 4.1934635657380153e-07,
      "loss": 0.30646042823791503,
      "num_input_tokens_seen": 822490280,
      "step": 25805,
      "train_runtime": 67581.555,
      "train_tokens_per_second": 12170.337
    },
    {
      "epoch": 1.7378130891462429,
      "grad_norm": 0.8259070454977737,
      "learning_rate": 4.1828703839603637e-07,
      "loss": 0.26773734092712403,
      "num_input_tokens_seen": 822651272,
      "step": 25810,
      "train_runtime": 67594.4213,
      "train_tokens_per_second": 12170.402
    },
    {
      "epoch": 1.738149744142203,
      "grad_norm": 0.8860702806949541,
      "learning_rate": 4.172290014796926e-07,
      "loss": 0.31717891693115235,
      "num_input_tokens_seen": 822812392,
      "step": 25815,
      "train_runtime": 67608.1076,
      "train_tokens_per_second": 12170.321
    },
    {
      "epoch": 1.7384863991381632,
      "grad_norm": 0.8253235960927757,
      "learning_rate": 4.1617224612064353e-07,
      "loss": 0.30143697261810304,
      "num_input_tokens_seen": 822968992,
      "step": 25820,
      "train_runtime": 67620.7873,
      "train_tokens_per_second": 12170.355
    },
    {
      "epoch": 1.7388230541341234,
      "grad_norm": 0.8469032430386131,
      "learning_rate": 4.151167726144101e-07,
      "loss": 0.28963942527770997,
      "num_input_tokens_seen": 823131232,
      "step": 25825,
      "train_runtime": 67633.3369,
      "train_tokens_per_second": 12170.496
    },
    {
      "epoch": 1.7391597091300834,
      "grad_norm": 0.9506559237709057,
      "learning_rate": 4.1406258125615106e-07,
      "loss": 0.31564314365386964,
      "num_input_tokens_seen": 823293424,
      "step": 25830,
      "train_runtime": 67646.0705,
      "train_tokens_per_second": 12170.602
    },
    {
      "epoch": 1.7394963641260435,
      "grad_norm": 0.8448783604334964,
      "learning_rate": 4.130096723406674e-07,
      "loss": 0.28644189834594724,
      "num_input_tokens_seen": 823454864,
      "step": 25835,
      "train_runtime": 67658.6114,
      "train_tokens_per_second": 12170.733
    },
    {
      "epoch": 1.7398330191220037,
      "grad_norm": 0.8009349505368849,
      "learning_rate": 4.119580461624023e-07,
      "loss": 0.29476051330566405,
      "num_input_tokens_seen": 823609808,
      "step": 25840,
      "train_runtime": 67671.6484,
      "train_tokens_per_second": 12170.677
    },
    {
      "epoch": 1.740169674117964,
      "grad_norm": 0.8097328828865649,
      "learning_rate": 4.109077030154396e-07,
      "loss": 0.2973953723907471,
      "num_input_tokens_seen": 823768760,
      "step": 25845,
      "train_runtime": 67684.8412,
      "train_tokens_per_second": 12170.654
    },
    {
      "epoch": 1.740506329113924,
      "grad_norm": 0.7735847424402714,
      "learning_rate": 4.0985864319350366e-07,
      "loss": 0.3041586399078369,
      "num_input_tokens_seen": 823928096,
      "step": 25850,
      "train_runtime": 67697.3387,
      "train_tokens_per_second": 12170.761
    },
    {
      "epoch": 1.740842984109884,
      "grad_norm": 0.8000685810483197,
      "learning_rate": 4.0881086698996154e-07,
      "loss": 0.2893008470535278,
      "num_input_tokens_seen": 824088728,
      "step": 25855,
      "train_runtime": 67710.3763,
      "train_tokens_per_second": 12170.789
    },
    {
      "epoch": 1.7411796391058443,
      "grad_norm": 0.7103107132674822,
      "learning_rate": 4.077643746978194e-07,
      "loss": 0.2892823457717896,
      "num_input_tokens_seen": 824247056,
      "step": 25860,
      "train_runtime": 67723.4134,
      "train_tokens_per_second": 12170.784
    },
    {
      "epoch": 1.7415162941018045,
      "grad_norm": 0.7254913796660607,
      "learning_rate": 4.067191666097281e-07,
      "loss": 0.2916910171508789,
      "num_input_tokens_seen": 824405528,
      "step": 25865,
      "train_runtime": 67735.9465,
      "train_tokens_per_second": 12170.872
    },
    {
      "epoch": 1.7418529490977646,
      "grad_norm": 0.7478447812368314,
      "learning_rate": 4.056752430179728e-07,
      "loss": 0.2701808214187622,
      "num_input_tokens_seen": 824566288,
      "step": 25870,
      "train_runtime": 67749.1007,
      "train_tokens_per_second": 12170.882
    },
    {
      "epoch": 1.7421896040937246,
      "grad_norm": 0.8082182337262482,
      "learning_rate": 4.046326042144866e-07,
      "loss": 0.2708259344100952,
      "num_input_tokens_seen": 824725688,
      "step": 25875,
      "train_runtime": 67762.1114,
      "train_tokens_per_second": 12170.897
    },
    {
      "epoch": 1.7425262590896848,
      "grad_norm": 0.7136393166697113,
      "learning_rate": 4.0359125049083956e-07,
      "loss": 0.26763110160827636,
      "num_input_tokens_seen": 824885296,
      "step": 25880,
      "train_runtime": 67775.2683,
      "train_tokens_per_second": 12170.889
    },
    {
      "epoch": 1.742862914085645,
      "grad_norm": 0.7808109555891378,
      "learning_rate": 4.025511821382422e-07,
      "loss": 0.28880484104156495,
      "num_input_tokens_seen": 825044464,
      "step": 25885,
      "train_runtime": 67788.3239,
      "train_tokens_per_second": 12170.893
    },
    {
      "epoch": 1.7431995690816051,
      "grad_norm": 0.7700625647874363,
      "learning_rate": 4.015123994475462e-07,
      "loss": 0.2705874443054199,
      "num_input_tokens_seen": 825205648,
      "step": 25890,
      "train_runtime": 67801.2345,
      "train_tokens_per_second": 12170.953
    },
    {
      "epoch": 1.7435362240775654,
      "grad_norm": 0.8753905789152233,
      "learning_rate": 4.0047490270924417e-07,
      "loss": 0.2902699947357178,
      "num_input_tokens_seen": 825369488,
      "step": 25895,
      "train_runtime": 67813.6777,
      "train_tokens_per_second": 12171.136
    },
    {
      "epoch": 1.7438728790735256,
      "grad_norm": 0.7599121855863348,
      "learning_rate": 3.994386922134691e-07,
      "loss": 0.29365177154541017,
      "num_input_tokens_seen": 825531344,
      "step": 25900,
      "train_runtime": 67826.4489,
      "train_tokens_per_second": 12171.231
    },
    {
      "epoch": 1.7442095340694856,
      "grad_norm": 0.7599242902897211,
      "learning_rate": 3.984037682499936e-07,
      "loss": 0.29141740798950194,
      "num_input_tokens_seen": 825695104,
      "step": 25905,
      "train_runtime": 67839.5429,
      "train_tokens_per_second": 12171.295
    },
    {
      "epoch": 1.7445461890654457,
      "grad_norm": 0.8078913076789045,
      "learning_rate": 3.973701311082301e-07,
      "loss": 0.30506291389465334,
      "num_input_tokens_seen": 825855064,
      "step": 25910,
      "train_runtime": 67853.4133,
      "train_tokens_per_second": 12171.165
    },
    {
      "epoch": 1.744882844061406,
      "grad_norm": 0.8250400055358497,
      "learning_rate": 3.9633778107723454e-07,
      "loss": 0.28368372917175294,
      "num_input_tokens_seen": 826014464,
      "step": 25915,
      "train_runtime": 67866.966,
      "train_tokens_per_second": 12171.083
    },
    {
      "epoch": 1.7452194990573662,
      "grad_norm": 0.7950938350677481,
      "learning_rate": 3.953067184456966e-07,
      "loss": 0.303162670135498,
      "num_input_tokens_seen": 826177464,
      "step": 25920,
      "train_runtime": 67881.2696,
      "train_tokens_per_second": 12170.919
    },
    {
      "epoch": 1.7455561540533262,
      "grad_norm": 0.8179246516177795,
      "learning_rate": 3.9427694350195224e-07,
      "loss": 0.29900636672973635,
      "num_input_tokens_seen": 826339368,
      "step": 25925,
      "train_runtime": 67894.4983,
      "train_tokens_per_second": 12170.933
    },
    {
      "epoch": 1.7458928090492862,
      "grad_norm": 0.8591215166716707,
      "learning_rate": 3.9324845653397446e-07,
      "loss": 0.30592570304870603,
      "num_input_tokens_seen": 826497864,
      "step": 25930,
      "train_runtime": 67906.9102,
      "train_tokens_per_second": 12171.042
    },
    {
      "epoch": 1.7462294640452465,
      "grad_norm": 0.8280538889502835,
      "learning_rate": 3.922212578293755e-07,
      "loss": 0.2747958660125732,
      "num_input_tokens_seen": 826653080,
      "step": 25935,
      "train_runtime": 67919.6671,
      "train_tokens_per_second": 12171.041
    },
    {
      "epoch": 1.7465661190412067,
      "grad_norm": 0.8072086912341124,
      "learning_rate": 3.911953476754088e-07,
      "loss": 0.27523221969604494,
      "num_input_tokens_seen": 826812528,
      "step": 25940,
      "train_runtime": 67932.7064,
      "train_tokens_per_second": 12171.052
    },
    {
      "epoch": 1.7469027740371668,
      "grad_norm": 0.8527988954133155,
      "learning_rate": 3.9017072635896716e-07,
      "loss": 0.3086763143539429,
      "num_input_tokens_seen": 826976368,
      "step": 25945,
      "train_runtime": 67945.1314,
      "train_tokens_per_second": 12171.238
    },
    {
      "epoch": 1.7472394290331268,
      "grad_norm": 0.8207133553929679,
      "learning_rate": 3.891473941665819e-07,
      "loss": 0.309571647644043,
      "num_input_tokens_seen": 827139120,
      "step": 25950,
      "train_runtime": 67958.576,
      "train_tokens_per_second": 12171.225
    },
    {
      "epoch": 1.747576084029087,
      "grad_norm": 0.822180435531457,
      "learning_rate": 3.881253513844246e-07,
      "loss": 0.284910774230957,
      "num_input_tokens_seen": 827294512,
      "step": 25955,
      "train_runtime": 67972.2687,
      "train_tokens_per_second": 12171.059
    },
    {
      "epoch": 1.7479127390250473,
      "grad_norm": 0.8049814467717832,
      "learning_rate": 3.871045982983063e-07,
      "loss": 0.29324338436126707,
      "num_input_tokens_seen": 827456960,
      "step": 25960,
      "train_runtime": 67984.9214,
      "train_tokens_per_second": 12171.184
    },
    {
      "epoch": 1.7482493940210073,
      "grad_norm": 0.7793104658189561,
      "learning_rate": 3.8608513519367897e-07,
      "loss": 0.25891363620758057,
      "num_input_tokens_seen": 827620000,
      "step": 25965,
      "train_runtime": 67997.7645,
      "train_tokens_per_second": 12171.282
    },
    {
      "epoch": 1.7485860490169673,
      "grad_norm": 0.9563367614426816,
      "learning_rate": 3.8506696235562915e-07,
      "loss": 0.29891912937164306,
      "num_input_tokens_seen": 827779664,
      "step": 25970,
      "train_runtime": 68010.9037,
      "train_tokens_per_second": 12171.279
    },
    {
      "epoch": 1.7489227040129276,
      "grad_norm": 0.8005269478719645,
      "learning_rate": 3.840500800688884e-07,
      "loss": 0.2913177490234375,
      "num_input_tokens_seen": 827943328,
      "step": 25975,
      "train_runtime": 68023.9444,
      "train_tokens_per_second": 12171.351
    },
    {
      "epoch": 1.7492593590088878,
      "grad_norm": 0.7911652828329712,
      "learning_rate": 3.8303448861782333e-07,
      "loss": 0.29439411163330076,
      "num_input_tokens_seen": 828099856,
      "step": 25980,
      "train_runtime": 68038.2685,
      "train_tokens_per_second": 12171.09
    },
    {
      "epoch": 1.7495960140048479,
      "grad_norm": 0.8557106367441378,
      "learning_rate": 3.8202018828644105e-07,
      "loss": 0.2996375322341919,
      "num_input_tokens_seen": 828260176,
      "step": 25985,
      "train_runtime": 68050.9975,
      "train_tokens_per_second": 12171.169
    },
    {
      "epoch": 1.749932669000808,
      "grad_norm": 1.3179167487126955,
      "learning_rate": 3.8100717935838804e-07,
      "loss": 0.30783030986785886,
      "num_input_tokens_seen": 828420016,
      "step": 25990,
      "train_runtime": 68064.4122,
      "train_tokens_per_second": 12171.118
    },
    {
      "epoch": 1.7502693239967682,
      "grad_norm": 0.7986354102465841,
      "learning_rate": 3.799954621169477e-07,
      "loss": 0.2694204807281494,
      "num_input_tokens_seen": 828574960,
      "step": 25995,
      "train_runtime": 68076.8509,
      "train_tokens_per_second": 12171.171
    },
    {
      "epoch": 1.7506059789927284,
      "grad_norm": 0.8353484628297734,
      "learning_rate": 3.7898503684504483e-07,
      "loss": 0.28495326042175295,
      "num_input_tokens_seen": 828729904,
      "step": 26000,
      "train_runtime": 68090.1167,
      "train_tokens_per_second": 12171.075
    },
    {
      "epoch": 1.7509426339886884,
      "grad_norm": 0.7994364132069013,
      "learning_rate": 3.7797590382524066e-07,
      "loss": 0.28560824394226075,
      "num_input_tokens_seen": 828883616,
      "step": 26005,
      "train_runtime": 68103.5639,
      "train_tokens_per_second": 12170.929
    },
    {
      "epoch": 1.7512792889846485,
      "grad_norm": 0.7861424831465385,
      "learning_rate": 3.76968063339736e-07,
      "loss": 0.29622807502746584,
      "num_input_tokens_seen": 829046816,
      "step": 26010,
      "train_runtime": 68117.155,
      "train_tokens_per_second": 12170.896
    },
    {
      "epoch": 1.7516159439806087,
      "grad_norm": 0.8568620090538059,
      "learning_rate": 3.759615156703722e-07,
      "loss": 0.28516526222229005,
      "num_input_tokens_seen": 829206736,
      "step": 26015,
      "train_runtime": 68130.4184,
      "train_tokens_per_second": 12170.874
    },
    {
      "epoch": 1.751952598976569,
      "grad_norm": 0.8105062700021983,
      "learning_rate": 3.749562610986235e-07,
      "loss": 0.3117044448852539,
      "num_input_tokens_seen": 829365456,
      "step": 26020,
      "train_runtime": 68143.4044,
      "train_tokens_per_second": 12170.884
    },
    {
      "epoch": 1.752289253972529,
      "grad_norm": 0.8457838730293944,
      "learning_rate": 3.739522999056094e-07,
      "loss": 0.2834632396697998,
      "num_input_tokens_seen": 829519520,
      "step": 26025,
      "train_runtime": 68156.2977,
      "train_tokens_per_second": 12170.842
    },
    {
      "epoch": 1.752625908968489,
      "grad_norm": 0.7966888624256843,
      "learning_rate": 3.7294963237208326e-07,
      "loss": 0.2870750665664673,
      "num_input_tokens_seen": 829681256,
      "step": 26030,
      "train_runtime": 68168.8305,
      "train_tokens_per_second": 12170.977
    },
    {
      "epoch": 1.7529625639644493,
      "grad_norm": 0.7851451419305832,
      "learning_rate": 3.7194825877843786e-07,
      "loss": 0.27614150047302244,
      "num_input_tokens_seen": 829843536,
      "step": 26035,
      "train_runtime": 68182.1082,
      "train_tokens_per_second": 12170.987
    },
    {
      "epoch": 1.7532992189604095,
      "grad_norm": 0.7373579735495371,
      "learning_rate": 3.7094817940470375e-07,
      "loss": 0.30138297080993653,
      "num_input_tokens_seen": 830000384,
      "step": 26040,
      "train_runtime": 68195.6072,
      "train_tokens_per_second": 12170.878
    },
    {
      "epoch": 1.7536358739563696,
      "grad_norm": 0.7720647820238346,
      "learning_rate": 3.6994939453055044e-07,
      "loss": 0.3058760404586792,
      "num_input_tokens_seen": 830162568,
      "step": 26045,
      "train_runtime": 68208.0551,
      "train_tokens_per_second": 12171.034
    },
    {
      "epoch": 1.7539725289523296,
      "grad_norm": 0.863624455936118,
      "learning_rate": 3.6895190443528515e-07,
      "loss": 0.2941605091094971,
      "num_input_tokens_seen": 830325784,
      "step": 26050,
      "train_runtime": 68221.5634,
      "train_tokens_per_second": 12171.017
    },
    {
      "epoch": 1.7543091839482898,
      "grad_norm": 0.7516782898069383,
      "learning_rate": 3.679557093978525e-07,
      "loss": 0.3141140937805176,
      "num_input_tokens_seen": 830485960,
      "step": 26055,
      "train_runtime": 68235.1696,
      "train_tokens_per_second": 12170.937
    },
    {
      "epoch": 1.75464583894425,
      "grad_norm": 0.8657816006187536,
      "learning_rate": 3.669608096968341e-07,
      "loss": 0.29130532741546633,
      "num_input_tokens_seen": 830642672,
      "step": 26060,
      "train_runtime": 68248.1366,
      "train_tokens_per_second": 12170.921
    },
    {
      "epoch": 1.75498249394021,
      "grad_norm": 0.8440661121969699,
      "learning_rate": 3.6596720561045295e-07,
      "loss": 0.27753758430480957,
      "num_input_tokens_seen": 830805288,
      "step": 26065,
      "train_runtime": 68261.5345,
      "train_tokens_per_second": 12170.914
    },
    {
      "epoch": 1.7553191489361701,
      "grad_norm": 0.8284615120113076,
      "learning_rate": 3.6497489741656455e-07,
      "loss": 0.30378992557525636,
      "num_input_tokens_seen": 830966136,
      "step": 26070,
      "train_runtime": 68274.6076,
      "train_tokens_per_second": 12170.94
    },
    {
      "epoch": 1.7556558039321304,
      "grad_norm": 0.7518668615361924,
      "learning_rate": 3.639838853926669e-07,
      "loss": 0.2975461006164551,
      "num_input_tokens_seen": 831129832,
      "step": 26075,
      "train_runtime": 68287.6435,
      "train_tokens_per_second": 12171.014
    },
    {
      "epoch": 1.7559924589280906,
      "grad_norm": 0.8750457070563752,
      "learning_rate": 3.629941698158923e-07,
      "loss": 0.33007795810699464,
      "num_input_tokens_seen": 831287536,
      "step": 26080,
      "train_runtime": 68300.1319,
      "train_tokens_per_second": 12171.097
    },
    {
      "epoch": 1.7563291139240507,
      "grad_norm": 0.8406191793330116,
      "learning_rate": 3.6200575096301207e-07,
      "loss": 0.2685899257659912,
      "num_input_tokens_seen": 831447480,
      "step": 26085,
      "train_runtime": 68312.6264,
      "train_tokens_per_second": 12171.212
    },
    {
      "epoch": 1.7566657689200107,
      "grad_norm": 0.8406631617474014,
      "learning_rate": 3.61018629110434e-07,
      "loss": 0.30750741958618166,
      "num_input_tokens_seen": 831608376,
      "step": 26090,
      "train_runtime": 68325.4583,
      "train_tokens_per_second": 12171.281
    },
    {
      "epoch": 1.757002423915971,
      "grad_norm": 0.7740435196887874,
      "learning_rate": 3.6003280453420386e-07,
      "loss": 0.2779327630996704,
      "num_input_tokens_seen": 831771016,
      "step": 26095,
      "train_runtime": 68339.0841,
      "train_tokens_per_second": 12171.234
    },
    {
      "epoch": 1.7573390789119312,
      "grad_norm": 0.8472729438857713,
      "learning_rate": 3.59048277510004e-07,
      "loss": 0.27883272171020507,
      "num_input_tokens_seen": 831930344,
      "step": 26100,
      "train_runtime": 68352.1277,
      "train_tokens_per_second": 12171.243
    },
    {
      "epoch": 1.7576757339078912,
      "grad_norm": 0.8978613058697938,
      "learning_rate": 3.5806504831315423e-07,
      "loss": 0.27771453857421874,
      "num_input_tokens_seen": 832091472,
      "step": 26105,
      "train_runtime": 68365.2024,
      "train_tokens_per_second": 12171.272
    },
    {
      "epoch": 1.7580123889038513,
      "grad_norm": 0.8187986171834535,
      "learning_rate": 3.570831172186112e-07,
      "loss": 0.28376994132995603,
      "num_input_tokens_seen": 832254864,
      "step": 26110,
      "train_runtime": 68378.8612,
      "train_tokens_per_second": 12171.23
    },
    {
      "epoch": 1.7583490438998115,
      "grad_norm": 0.8138659095683347,
      "learning_rate": 3.5610248450097084e-07,
      "loss": 0.2963928937911987,
      "num_input_tokens_seen": 832415560,
      "step": 26115,
      "train_runtime": 68391.6667,
      "train_tokens_per_second": 12171.301
    },
    {
      "epoch": 1.7586856988957718,
      "grad_norm": 0.7578083758134397,
      "learning_rate": 3.551231504344604e-07,
      "loss": 0.2807722806930542,
      "num_input_tokens_seen": 832573352,
      "step": 26120,
      "train_runtime": 68404.1396,
      "train_tokens_per_second": 12171.388
    },
    {
      "epoch": 1.7590223538917318,
      "grad_norm": 0.7789650086734451,
      "learning_rate": 3.5414511529295036e-07,
      "loss": 0.30507802963256836,
      "num_input_tokens_seen": 832728200,
      "step": 26125,
      "train_runtime": 68417.4133,
      "train_tokens_per_second": 12171.29
    },
    {
      "epoch": 1.7593590088876918,
      "grad_norm": 0.9526643775849574,
      "learning_rate": 3.5316837934994465e-07,
      "loss": 0.3145780563354492,
      "num_input_tokens_seen": 832885256,
      "step": 26130,
      "train_runtime": 68430.1543,
      "train_tokens_per_second": 12171.319
    },
    {
      "epoch": 1.759695663883652,
      "grad_norm": 0.709408052662059,
      "learning_rate": 3.521929428785836e-07,
      "loss": 0.28622188568115237,
      "num_input_tokens_seen": 833047232,
      "step": 26135,
      "train_runtime": 68443.3345,
      "train_tokens_per_second": 12171.342
    },
    {
      "epoch": 1.7600323188796123,
      "grad_norm": 0.7418910698613626,
      "learning_rate": 3.5121880615164584e-07,
      "loss": 0.3126822471618652,
      "num_input_tokens_seen": 833206032,
      "step": 26140,
      "train_runtime": 68456.2397,
      "train_tokens_per_second": 12171.367
    },
    {
      "epoch": 1.7603689738755723,
      "grad_norm": 0.7834278032485381,
      "learning_rate": 3.5024596944154445e-07,
      "loss": 0.30379271507263184,
      "num_input_tokens_seen": 833362264,
      "step": 26145,
      "train_runtime": 68469.3929,
      "train_tokens_per_second": 12171.311
    },
    {
      "epoch": 1.7607056288715324,
      "grad_norm": 0.8651300463911419,
      "learning_rate": 3.4927443302033127e-07,
      "loss": 0.2798401117324829,
      "num_input_tokens_seen": 833522792,
      "step": 26150,
      "train_runtime": 68481.8012,
      "train_tokens_per_second": 12171.45
    },
    {
      "epoch": 1.7610422838674926,
      "grad_norm": 0.8041563410212744,
      "learning_rate": 3.4830419715969233e-07,
      "loss": 0.2808249473571777,
      "num_input_tokens_seen": 833682240,
      "step": 26155,
      "train_runtime": 68495.3045,
      "train_tokens_per_second": 12171.378
    },
    {
      "epoch": 1.7613789388634529,
      "grad_norm": 0.8873948348689605,
      "learning_rate": 3.4733526213095114e-07,
      "loss": 0.29953155517578123,
      "num_input_tokens_seen": 833844440,
      "step": 26160,
      "train_runtime": 68507.8935,
      "train_tokens_per_second": 12171.509
    },
    {
      "epoch": 1.761715593859413,
      "grad_norm": 0.8487255523831553,
      "learning_rate": 3.4636762820506876e-07,
      "loss": 0.29048471450805663,
      "num_input_tokens_seen": 834007064,
      "step": 26165,
      "train_runtime": 68522.0354,
      "train_tokens_per_second": 12171.37
    },
    {
      "epoch": 1.762052248855373,
      "grad_norm": 0.9712930232759466,
      "learning_rate": 3.4540129565263867e-07,
      "loss": 0.3151499032974243,
      "num_input_tokens_seen": 834167144,
      "step": 26170,
      "train_runtime": 68534.8183,
      "train_tokens_per_second": 12171.436
    },
    {
      "epoch": 1.7623889038513332,
      "grad_norm": 0.8271596968123386,
      "learning_rate": 3.444362647438942e-07,
      "loss": 0.29479053020477297,
      "num_input_tokens_seen": 834329040,
      "step": 26175,
      "train_runtime": 68547.361,
      "train_tokens_per_second": 12171.571
    },
    {
      "epoch": 1.7627255588472934,
      "grad_norm": 0.8318965925249644,
      "learning_rate": 3.434725357487029e-07,
      "loss": 0.30252196788787844,
      "num_input_tokens_seen": 834492256,
      "step": 26180,
      "train_runtime": 68560.2413,
      "train_tokens_per_second": 12171.665
    },
    {
      "epoch": 1.7630622138432535,
      "grad_norm": 0.8731690536144744,
      "learning_rate": 3.4251010893656844e-07,
      "loss": 0.3206812381744385,
      "num_input_tokens_seen": 834650152,
      "step": 26185,
      "train_runtime": 68572.6762,
      "train_tokens_per_second": 12171.76
    },
    {
      "epoch": 1.7633988688392135,
      "grad_norm": 0.9388429486263101,
      "learning_rate": 3.4154898457663065e-07,
      "loss": 0.2851539373397827,
      "num_input_tokens_seen": 834804648,
      "step": 26190,
      "train_runtime": 68585.7187,
      "train_tokens_per_second": 12171.698
    },
    {
      "epoch": 1.7637355238351737,
      "grad_norm": 0.7323318521974965,
      "learning_rate": 3.40589162937664e-07,
      "loss": 0.27820625305175783,
      "num_input_tokens_seen": 834960120,
      "step": 26195,
      "train_runtime": 68599.1278,
      "train_tokens_per_second": 12171.585
    },
    {
      "epoch": 1.764072178831134,
      "grad_norm": 0.81224541549757,
      "learning_rate": 3.3963064428808034e-07,
      "loss": 0.273958683013916,
      "num_input_tokens_seen": 835119608,
      "step": 26200,
      "train_runtime": 68612.3652,
      "train_tokens_per_second": 12171.561
    },
    {
      "epoch": 1.764408833827094,
      "grad_norm": 0.7423470566975758,
      "learning_rate": 3.386734288959265e-07,
      "loss": 0.29813809394836427,
      "num_input_tokens_seen": 835275592,
      "step": 26205,
      "train_runtime": 68626.2912,
      "train_tokens_per_second": 12171.364
    },
    {
      "epoch": 1.764745488823054,
      "grad_norm": 0.8000263585349465,
      "learning_rate": 3.377175170288838e-07,
      "loss": 0.2990430355072021,
      "num_input_tokens_seen": 835436328,
      "step": 26210,
      "train_runtime": 68639.382,
      "train_tokens_per_second": 12171.385
    },
    {
      "epoch": 1.7650821438190143,
      "grad_norm": 0.8424547830672207,
      "learning_rate": 3.367629089542712e-07,
      "loss": 0.27294411659240725,
      "num_input_tokens_seen": 835600168,
      "step": 26215,
      "train_runtime": 68651.8029,
      "train_tokens_per_second": 12171.569
    },
    {
      "epoch": 1.7654187988149745,
      "grad_norm": 0.8310820624953507,
      "learning_rate": 3.3580960493904024e-07,
      "loss": 0.3068228244781494,
      "num_input_tokens_seen": 835760416,
      "step": 26220,
      "train_runtime": 68664.853,
      "train_tokens_per_second": 12171.59
    },
    {
      "epoch": 1.7657554538109346,
      "grad_norm": 0.7836867062670843,
      "learning_rate": 3.3485760524978093e-07,
      "loss": 0.2934136867523193,
      "num_input_tokens_seen": 835915984,
      "step": 26225,
      "train_runtime": 68678.22,
      "train_tokens_per_second": 12171.486
    },
    {
      "epoch": 1.7660921088068946,
      "grad_norm": 0.8118792612966774,
      "learning_rate": 3.3390691015271646e-07,
      "loss": 0.27722289562225344,
      "num_input_tokens_seen": 836075000,
      "step": 26230,
      "train_runtime": 68691.2459,
      "train_tokens_per_second": 12171.493
    },
    {
      "epoch": 1.7664287638028549,
      "grad_norm": 0.784134256028318,
      "learning_rate": 3.329575199137053e-07,
      "loss": 0.2810239315032959,
      "num_input_tokens_seen": 836236240,
      "step": 26235,
      "train_runtime": 68704.8053,
      "train_tokens_per_second": 12171.437
    },
    {
      "epoch": 1.766765418798815,
      "grad_norm": 0.8146416939158622,
      "learning_rate": 3.3200943479824113e-07,
      "loss": 0.3166502952575684,
      "num_input_tokens_seen": 836393896,
      "step": 26240,
      "train_runtime": 68718.6067,
      "train_tokens_per_second": 12171.287
    },
    {
      "epoch": 1.7671020737947751,
      "grad_norm": 0.7143019389419657,
      "learning_rate": 3.3106265507145406e-07,
      "loss": 0.24813783168792725,
      "num_input_tokens_seen": 836550104,
      "step": 26245,
      "train_runtime": 68731.6147,
      "train_tokens_per_second": 12171.256
    },
    {
      "epoch": 1.7674387287907352,
      "grad_norm": 0.7938698004270511,
      "learning_rate": 3.3011718099810687e-07,
      "loss": 0.27550978660583497,
      "num_input_tokens_seen": 836704096,
      "step": 26250,
      "train_runtime": 68744.0983,
      "train_tokens_per_second": 12171.286
    },
    {
      "epoch": 1.7677753837866954,
      "grad_norm": 0.8644887944098784,
      "learning_rate": 3.2917301284259904e-07,
      "loss": 0.29041543006896975,
      "num_input_tokens_seen": 836864144,
      "step": 26255,
      "train_runtime": 68756.5172,
      "train_tokens_per_second": 12171.416
    },
    {
      "epoch": 1.7681120387826557,
      "grad_norm": 0.8290599040203592,
      "learning_rate": 3.2823015086896335e-07,
      "loss": 0.30643649101257325,
      "num_input_tokens_seen": 837019848,
      "step": 26260,
      "train_runtime": 68769.0561,
      "train_tokens_per_second": 12171.46
    },
    {
      "epoch": 1.7684486937786157,
      "grad_norm": 0.778524774155057,
      "learning_rate": 3.2728859534087e-07,
      "loss": 0.2897134065628052,
      "num_input_tokens_seen": 837181992,
      "step": 26265,
      "train_runtime": 68782.9625,
      "train_tokens_per_second": 12171.357
    },
    {
      "epoch": 1.7687853487745757,
      "grad_norm": 0.7934363973745343,
      "learning_rate": 3.263483465216194e-07,
      "loss": 0.2934154987335205,
      "num_input_tokens_seen": 837344960,
      "step": 26270,
      "train_runtime": 68796.4616,
      "train_tokens_per_second": 12171.338
    },
    {
      "epoch": 1.769122003770536,
      "grad_norm": 0.7411787010396509,
      "learning_rate": 3.2540940467415126e-07,
      "loss": 0.278179407119751,
      "num_input_tokens_seen": 837503672,
      "step": 26275,
      "train_runtime": 68810.1784,
      "train_tokens_per_second": 12171.218
    },
    {
      "epoch": 1.7694586587664962,
      "grad_norm": 0.7441210240292669,
      "learning_rate": 3.2447177006103726e-07,
      "loss": 0.2868351936340332,
      "num_input_tokens_seen": 837660168,
      "step": 26280,
      "train_runtime": 68822.9762,
      "train_tokens_per_second": 12171.228
    },
    {
      "epoch": 1.7697953137624562,
      "grad_norm": 0.7288735938442041,
      "learning_rate": 3.235354429444831e-07,
      "loss": 0.28961236476898194,
      "num_input_tokens_seen": 837815408,
      "step": 26285,
      "train_runtime": 68836.6762,
      "train_tokens_per_second": 12171.061
    },
    {
      "epoch": 1.7701319687584163,
      "grad_norm": 0.830222078252063,
      "learning_rate": 3.226004235863306e-07,
      "loss": 0.3047652244567871,
      "num_input_tokens_seen": 837978176,
      "step": 26290,
      "train_runtime": 68849.4121,
      "train_tokens_per_second": 12171.174
    },
    {
      "epoch": 1.7704686237543765,
      "grad_norm": 0.8074273294046584,
      "learning_rate": 3.2166671224805437e-07,
      "loss": 0.2839834690093994,
      "num_input_tokens_seen": 838138664,
      "step": 26295,
      "train_runtime": 68862.1879,
      "train_tokens_per_second": 12171.247
    },
    {
      "epoch": 1.7708052787503368,
      "grad_norm": 0.7779803608716466,
      "learning_rate": 3.207343091907633e-07,
      "loss": 0.28159394264221194,
      "num_input_tokens_seen": 838294120,
      "step": 26300,
      "train_runtime": 68875.3881,
      "train_tokens_per_second": 12171.171
    },
    {
      "epoch": 1.7711419337462968,
      "grad_norm": 0.8394080431693071,
      "learning_rate": 3.1980321467520226e-07,
      "loss": 0.29482412338256836,
      "num_input_tokens_seen": 838445272,
      "step": 26305,
      "train_runtime": 68888.7817,
      "train_tokens_per_second": 12170.999
    },
    {
      "epoch": 1.7714785887422568,
      "grad_norm": 1.0024168854164968,
      "learning_rate": 3.1887342896174735e-07,
      "loss": 0.31112070083618165,
      "num_input_tokens_seen": 838601096,
      "step": 26310,
      "train_runtime": 68901.8674,
      "train_tokens_per_second": 12170.949
    },
    {
      "epoch": 1.771815243738217,
      "grad_norm": 0.7712599981166565,
      "learning_rate": 3.1794495231041276e-07,
      "loss": 0.2907053709030151,
      "num_input_tokens_seen": 838763824,
      "step": 26315,
      "train_runtime": 68914.5879,
      "train_tokens_per_second": 12171.063
    },
    {
      "epoch": 1.7721518987341773,
      "grad_norm": 0.7886069376926764,
      "learning_rate": 3.1701778498084035e-07,
      "loss": 0.3035216569900513,
      "num_input_tokens_seen": 838917928,
      "step": 26320,
      "train_runtime": 68927.9492,
      "train_tokens_per_second": 12170.94
    },
    {
      "epoch": 1.7724885537301374,
      "grad_norm": 0.8122901571560102,
      "learning_rate": 3.160919272323121e-07,
      "loss": 0.30497236251831056,
      "num_input_tokens_seen": 839081768,
      "step": 26325,
      "train_runtime": 68940.3833,
      "train_tokens_per_second": 12171.121
    },
    {
      "epoch": 1.7728252087260974,
      "grad_norm": 0.7139299546880953,
      "learning_rate": 3.1516737932374087e-07,
      "loss": 0.3071502447128296,
      "num_input_tokens_seen": 839236904,
      "step": 26330,
      "train_runtime": 68953.5961,
      "train_tokens_per_second": 12171.039
    },
    {
      "epoch": 1.7731618637220576,
      "grad_norm": 0.7649674316862415,
      "learning_rate": 3.142441415136727e-07,
      "loss": 0.2909811496734619,
      "num_input_tokens_seen": 839397824,
      "step": 26335,
      "train_runtime": 68966.3011,
      "train_tokens_per_second": 12171.13
    },
    {
      "epoch": 1.773498518718018,
      "grad_norm": 0.7717244992605078,
      "learning_rate": 3.133222140602893e-07,
      "loss": 0.2872778415679932,
      "num_input_tokens_seen": 839551816,
      "step": 26340,
      "train_runtime": 68980.2827,
      "train_tokens_per_second": 12170.896
    },
    {
      "epoch": 1.773835173713978,
      "grad_norm": 0.7768202715292135,
      "learning_rate": 3.124015972214034e-07,
      "loss": 0.2989944934844971,
      "num_input_tokens_seen": 839709280,
      "step": 26345,
      "train_runtime": 68993.4136,
      "train_tokens_per_second": 12170.861
    },
    {
      "epoch": 1.774171828709938,
      "grad_norm": 0.8654011796888754,
      "learning_rate": 3.114822912544635e-07,
      "loss": 0.30635437965393064,
      "num_input_tokens_seen": 839866096,
      "step": 26350,
      "train_runtime": 69005.8644,
      "train_tokens_per_second": 12170.938
    },
    {
      "epoch": 1.7745084837058982,
      "grad_norm": 0.8214804899610394,
      "learning_rate": 3.1056429641655007e-07,
      "loss": 0.28124299049377444,
      "num_input_tokens_seen": 840024112,
      "step": 26355,
      "train_runtime": 69019.1233,
      "train_tokens_per_second": 12170.889
    },
    {
      "epoch": 1.7748451387018584,
      "grad_norm": 0.8413278397004615,
      "learning_rate": 3.0964761296437665e-07,
      "loss": 0.28792123794555663,
      "num_input_tokens_seen": 840185136,
      "step": 26360,
      "train_runtime": 69032.3353,
      "train_tokens_per_second": 12170.893
    },
    {
      "epoch": 1.7751817936978185,
      "grad_norm": 0.8216510706756607,
      "learning_rate": 3.087322411542937e-07,
      "loss": 0.29073960781097413,
      "num_input_tokens_seen": 840347024,
      "step": 26365,
      "train_runtime": 69044.8832,
      "train_tokens_per_second": 12171.025
    },
    {
      "epoch": 1.7755184486937785,
      "grad_norm": 0.8013352594745409,
      "learning_rate": 3.0781818124227867e-07,
      "loss": 0.27419252395629884,
      "num_input_tokens_seen": 840504648,
      "step": 26370,
      "train_runtime": 69058.6353,
      "train_tokens_per_second": 12170.884
    },
    {
      "epoch": 1.7758551036897388,
      "grad_norm": 0.8202755964720164,
      "learning_rate": 3.0690543348394776e-07,
      "loss": 0.28411192893981935,
      "num_input_tokens_seen": 840667976,
      "step": 26375,
      "train_runtime": 69072.2673,
      "train_tokens_per_second": 12170.847
    },
    {
      "epoch": 1.776191758685699,
      "grad_norm": 0.8912154372593172,
      "learning_rate": 3.059939981345467e-07,
      "loss": 0.27138538360595704,
      "num_input_tokens_seen": 840822112,
      "step": 26380,
      "train_runtime": 69085.4308,
      "train_tokens_per_second": 12170.759
    },
    {
      "epoch": 1.776528413681659,
      "grad_norm": 0.8501836326542918,
      "learning_rate": 3.050838754489566e-07,
      "loss": 0.2908538818359375,
      "num_input_tokens_seen": 840981976,
      "step": 26385,
      "train_runtime": 69098.4911,
      "train_tokens_per_second": 12170.772
    },
    {
      "epoch": 1.776865068677619,
      "grad_norm": 0.7550063820444615,
      "learning_rate": 3.041750656816894e-07,
      "loss": 0.28323516845703123,
      "num_input_tokens_seen": 841141640,
      "step": 26390,
      "train_runtime": 69111.5914,
      "train_tokens_per_second": 12170.775
    },
    {
      "epoch": 1.7772017236735793,
      "grad_norm": 0.8579419856571887,
      "learning_rate": 3.0326756908689135e-07,
      "loss": 0.3049760818481445,
      "num_input_tokens_seen": 841302576,
      "step": 26395,
      "train_runtime": 69124.9971,
      "train_tokens_per_second": 12170.743
    },
    {
      "epoch": 1.7775383786695396,
      "grad_norm": 0.7977291916700361,
      "learning_rate": 3.0236138591833995e-07,
      "loss": 0.2616129875183105,
      "num_input_tokens_seen": 841460184,
      "step": 26400,
      "train_runtime": 69138.0468,
      "train_tokens_per_second": 12170.725
    },
    {
      "epoch": 1.7778750336654996,
      "grad_norm": 0.7683885665471022,
      "learning_rate": 3.0145651642944863e-07,
      "loss": 0.3078320503234863,
      "num_input_tokens_seen": 841622592,
      "step": 26405,
      "train_runtime": 69150.6958,
      "train_tokens_per_second": 12170.848
    },
    {
      "epoch": 1.7782116886614596,
      "grad_norm": 0.8113822853931928,
      "learning_rate": 3.005529608732588e-07,
      "loss": 0.2570197582244873,
      "num_input_tokens_seen": 841781400,
      "step": 26410,
      "train_runtime": 69163.7673,
      "train_tokens_per_second": 12170.844
    },
    {
      "epoch": 1.7785483436574199,
      "grad_norm": 0.8022122693523719,
      "learning_rate": 2.996507195024495e-07,
      "loss": 0.3254894495010376,
      "num_input_tokens_seen": 841945240,
      "step": 26415,
      "train_runtime": 69176.2114,
      "train_tokens_per_second": 12171.023
    },
    {
      "epoch": 1.7788849986533801,
      "grad_norm": 0.8486430712695983,
      "learning_rate": 2.9874979256932614e-07,
      "loss": 0.2866269588470459,
      "num_input_tokens_seen": 842102160,
      "step": 26420,
      "train_runtime": 69188.6619,
      "train_tokens_per_second": 12171.101
    },
    {
      "epoch": 1.7792216536493402,
      "grad_norm": 0.8584621658467494,
      "learning_rate": 2.9785018032583325e-07,
      "loss": 0.30195250511169436,
      "num_input_tokens_seen": 842264576,
      "step": 26425,
      "train_runtime": 69201.2978,
      "train_tokens_per_second": 12171.225
    },
    {
      "epoch": 1.7795583086453002,
      "grad_norm": 0.9538845435241674,
      "learning_rate": 2.969518830235435e-07,
      "loss": 0.29738245010375974,
      "num_input_tokens_seen": 842419520,
      "step": 26430,
      "train_runtime": 69214.352,
      "train_tokens_per_second": 12171.168
    },
    {
      "epoch": 1.7798949636412604,
      "grad_norm": 0.8158518019941738,
      "learning_rate": 2.960549009136626e-07,
      "loss": 0.28472921848297117,
      "num_input_tokens_seen": 842581320,
      "step": 26435,
      "train_runtime": 69227.5765,
      "train_tokens_per_second": 12171.18
    },
    {
      "epoch": 1.7802316186372207,
      "grad_norm": 0.8009863560088409,
      "learning_rate": 2.951592342470289e-07,
      "loss": 0.3249552249908447,
      "num_input_tokens_seen": 842738432,
      "step": 26440,
      "train_runtime": 69240.0797,
      "train_tokens_per_second": 12171.252
    },
    {
      "epoch": 1.7805682736331807,
      "grad_norm": 0.8623727914416444,
      "learning_rate": 2.942648832741124e-07,
      "loss": 0.2858766555786133,
      "num_input_tokens_seen": 842896992,
      "step": 26445,
      "train_runtime": 69253.5758,
      "train_tokens_per_second": 12171.169
    },
    {
      "epoch": 1.7809049286291407,
      "grad_norm": 0.7653244663419082,
      "learning_rate": 2.933718482450154e-07,
      "loss": 0.2949942111968994,
      "num_input_tokens_seen": 843057480,
      "step": 26450,
      "train_runtime": 69266.1621,
      "train_tokens_per_second": 12171.275
    },
    {
      "epoch": 1.781241583625101,
      "grad_norm": 0.9713830903690602,
      "learning_rate": 2.9248012940947423e-07,
      "loss": 0.3111457109451294,
      "num_input_tokens_seen": 843219160,
      "step": 26455,
      "train_runtime": 69278.8527,
      "train_tokens_per_second": 12171.379
    },
    {
      "epoch": 1.7815782386210612,
      "grad_norm": 0.839682377014265,
      "learning_rate": 2.915897270168522e-07,
      "loss": 0.30369384288787843,
      "num_input_tokens_seen": 843376336,
      "step": 26460,
      "train_runtime": 69291.9963,
      "train_tokens_per_second": 12171.338
    },
    {
      "epoch": 1.7819148936170213,
      "grad_norm": 0.8515163723669086,
      "learning_rate": 2.9070064131615004e-07,
      "loss": 0.3054384708404541,
      "num_input_tokens_seen": 843533048,
      "step": 26465,
      "train_runtime": 69304.6387,
      "train_tokens_per_second": 12171.379
    },
    {
      "epoch": 1.7822515486129813,
      "grad_norm": 0.8684719584492041,
      "learning_rate": 2.8981287255599567e-07,
      "loss": 0.2868688106536865,
      "num_input_tokens_seen": 843686896,
      "step": 26470,
      "train_runtime": 69318.2983,
      "train_tokens_per_second": 12171.2
    },
    {
      "epoch": 1.7825882036089415,
      "grad_norm": 0.7817718611075751,
      "learning_rate": 2.8892642098465205e-07,
      "loss": 0.2904038906097412,
      "num_input_tokens_seen": 843847192,
      "step": 26475,
      "train_runtime": 69331.7107,
      "train_tokens_per_second": 12171.158
    },
    {
      "epoch": 1.7829248586049018,
      "grad_norm": 0.7985097989223785,
      "learning_rate": 2.8804128685001263e-07,
      "loss": 0.278177547454834,
      "num_input_tokens_seen": 844008120,
      "step": 26480,
      "train_runtime": 69344.7488,
      "train_tokens_per_second": 12171.19
    },
    {
      "epoch": 1.7832615136008618,
      "grad_norm": 0.8259976727480475,
      "learning_rate": 2.8715747039960216e-07,
      "loss": 0.26516387462615965,
      "num_input_tokens_seen": 844165168,
      "step": 26485,
      "train_runtime": 69358.1688,
      "train_tokens_per_second": 12171.099
    },
    {
      "epoch": 1.7835981685968219,
      "grad_norm": 0.8790284237270245,
      "learning_rate": 2.8627497188057684e-07,
      "loss": 0.29796299934387205,
      "num_input_tokens_seen": 844324064,
      "step": 26490,
      "train_runtime": 69371.4057,
      "train_tokens_per_second": 12171.068
    },
    {
      "epoch": 1.783934823592782,
      "grad_norm": 0.7942859710881532,
      "learning_rate": 2.853937915397248e-07,
      "loss": 0.32318692207336425,
      "num_input_tokens_seen": 844485232,
      "step": 26495,
      "train_runtime": 69385.072,
      "train_tokens_per_second": 12170.993
    },
    {
      "epoch": 1.7842714785887424,
      "grad_norm": 0.7780740517604148,
      "learning_rate": 2.845139296234639e-07,
      "loss": 0.28280413150787354,
      "num_input_tokens_seen": 844649072,
      "step": 26500,
      "train_runtime": 69397.5151,
      "train_tokens_per_second": 12171.172
    },
    {
      "epoch": 1.7846081335847024,
      "grad_norm": 0.7190213314704383,
      "learning_rate": 2.836353863778479e-07,
      "loss": 0.2819064617156982,
      "num_input_tokens_seen": 844810744,
      "step": 26505,
      "train_runtime": 69410.7384,
      "train_tokens_per_second": 12171.182
    },
    {
      "epoch": 1.7849447885806624,
      "grad_norm": 0.7697888179222947,
      "learning_rate": 2.8275816204855534e-07,
      "loss": 0.3100119590759277,
      "num_input_tokens_seen": 844971208,
      "step": 26510,
      "train_runtime": 69423.6948,
      "train_tokens_per_second": 12171.222
    },
    {
      "epoch": 1.7852814435766227,
      "grad_norm": 0.9136180012287308,
      "learning_rate": 2.818822568809015e-07,
      "loss": 0.3110760450363159,
      "num_input_tokens_seen": 845125936,
      "step": 26515,
      "train_runtime": 69436.5256,
      "train_tokens_per_second": 12171.201
    },
    {
      "epoch": 1.785618098572583,
      "grad_norm": 0.7611624020400739,
      "learning_rate": 2.810076711198284e-07,
      "loss": 0.27563962936401365,
      "num_input_tokens_seen": 845283576,
      "step": 26520,
      "train_runtime": 69449.4226,
      "train_tokens_per_second": 12171.211
    },
    {
      "epoch": 1.785954753568543,
      "grad_norm": 0.9659565279866518,
      "learning_rate": 2.801344050099131e-07,
      "loss": 0.3011836767196655,
      "num_input_tokens_seen": 845447416,
      "step": 26525,
      "train_runtime": 69461.8692,
      "train_tokens_per_second": 12171.389
    },
    {
      "epoch": 1.786291408564503,
      "grad_norm": 0.7948474156573586,
      "learning_rate": 2.7926245879536076e-07,
      "loss": 0.2969949245452881,
      "num_input_tokens_seen": 845606752,
      "step": 26530,
      "train_runtime": 69475.34,
      "train_tokens_per_second": 12171.322
    },
    {
      "epoch": 1.7866280635604632,
      "grad_norm": 0.744289892948673,
      "learning_rate": 2.7839183272000857e-07,
      "loss": 0.27946319580078127,
      "num_input_tokens_seen": 845769944,
      "step": 26535,
      "train_runtime": 69488.9681,
      "train_tokens_per_second": 12171.284
    },
    {
      "epoch": 1.7869647185564235,
      "grad_norm": 0.8897919507115974,
      "learning_rate": 2.775225270273241e-07,
      "loss": 0.2998188972473145,
      "num_input_tokens_seen": 845932008,
      "step": 26540,
      "train_runtime": 69502.1031,
      "train_tokens_per_second": 12171.315
    },
    {
      "epoch": 1.7873013735523835,
      "grad_norm": 0.7775340306162459,
      "learning_rate": 2.7665454196040665e-07,
      "loss": 0.26177880764007566,
      "num_input_tokens_seen": 846093840,
      "step": 26545,
      "train_runtime": 69515.3271,
      "train_tokens_per_second": 12171.328
    },
    {
      "epoch": 1.7876380285483435,
      "grad_norm": 0.8306001769269159,
      "learning_rate": 2.7578787776198325e-07,
      "loss": 0.30067741870880127,
      "num_input_tokens_seen": 846248448,
      "step": 26550,
      "train_runtime": 69528.9022,
      "train_tokens_per_second": 12171.175
    },
    {
      "epoch": 1.7879746835443038,
      "grad_norm": 0.9156496712006085,
      "learning_rate": 2.7492253467441775e-07,
      "loss": 0.2776355504989624,
      "num_input_tokens_seen": 846409776,
      "step": 26555,
      "train_runtime": 69541.8786,
      "train_tokens_per_second": 12171.224
    },
    {
      "epoch": 1.788311338540264,
      "grad_norm": 0.8179280601311999,
      "learning_rate": 2.740585129396961e-07,
      "loss": 0.29197301864624026,
      "num_input_tokens_seen": 846569896,
      "step": 26560,
      "train_runtime": 69554.5982,
      "train_tokens_per_second": 12171.3
    },
    {
      "epoch": 1.788647993536224,
      "grad_norm": 0.8103438662941587,
      "learning_rate": 2.731958127994433e-07,
      "loss": 0.27991909980773927,
      "num_input_tokens_seen": 846732024,
      "step": 26565,
      "train_runtime": 69567.4803,
      "train_tokens_per_second": 12171.377
    },
    {
      "epoch": 1.788984648532184,
      "grad_norm": 1.0943768422021962,
      "learning_rate": 2.7233443449490695e-07,
      "loss": 0.31302409172058104,
      "num_input_tokens_seen": 846887488,
      "step": 26570,
      "train_runtime": 69579.9311,
      "train_tokens_per_second": 12171.433
    },
    {
      "epoch": 1.7893213035281443,
      "grad_norm": 0.7808940685293475,
      "learning_rate": 2.7147437826697153e-07,
      "loss": 0.2940995693206787,
      "num_input_tokens_seen": 847049192,
      "step": 26575,
      "train_runtime": 69593.3163,
      "train_tokens_per_second": 12171.416
    },
    {
      "epoch": 1.7896579585241046,
      "grad_norm": 0.9172771589261917,
      "learning_rate": 2.706156443561481e-07,
      "loss": 0.2893664836883545,
      "num_input_tokens_seen": 847211304,
      "step": 26580,
      "train_runtime": 69606.5284,
      "train_tokens_per_second": 12171.435
    },
    {
      "epoch": 1.7899946135200646,
      "grad_norm": 0.7460807742213781,
      "learning_rate": 2.6975823300257884e-07,
      "loss": 0.29347786903381345,
      "num_input_tokens_seen": 847367928,
      "step": 26585,
      "train_runtime": 69619.7416,
      "train_tokens_per_second": 12171.374
    },
    {
      "epoch": 1.7903312685160246,
      "grad_norm": 0.7473205973219429,
      "learning_rate": 2.6890214444603604e-07,
      "loss": 0.2912222623825073,
      "num_input_tokens_seen": 847523624,
      "step": 26590,
      "train_runtime": 69632.4937,
      "train_tokens_per_second": 12171.381
    },
    {
      "epoch": 1.790667923511985,
      "grad_norm": 0.7965104210228536,
      "learning_rate": 2.6804737892592193e-07,
      "loss": 0.3100412368774414,
      "num_input_tokens_seen": 847686256,
      "step": 26595,
      "train_runtime": 69645.8614,
      "train_tokens_per_second": 12171.38
    },
    {
      "epoch": 1.7910045785079451,
      "grad_norm": 0.9732187312324612,
      "learning_rate": 2.671939366812687e-07,
      "loss": 0.28713240623474123,
      "num_input_tokens_seen": 847847328,
      "step": 26600,
      "train_runtime": 69659.1927,
      "train_tokens_per_second": 12171.363
    },
    {
      "epoch": 1.7913412335039052,
      "grad_norm": 0.8453433863728608,
      "learning_rate": 2.66341817950741e-07,
      "loss": 0.28653693199157715,
      "num_input_tokens_seen": 848010512,
      "step": 26605,
      "train_runtime": 69672.0872,
      "train_tokens_per_second": 12171.453
    },
    {
      "epoch": 1.7916778884998652,
      "grad_norm": 0.7798753280996197,
      "learning_rate": 2.6549102297262764e-07,
      "loss": 0.28526284694671633,
      "num_input_tokens_seen": 848168784,
      "step": 26610,
      "train_runtime": 69685.1881,
      "train_tokens_per_second": 12171.436
    },
    {
      "epoch": 1.7920145434958255,
      "grad_norm": 0.893826422158948,
      "learning_rate": 2.6464155198485387e-07,
      "loss": 0.304105281829834,
      "num_input_tokens_seen": 848328712,
      "step": 26615,
      "train_runtime": 69697.6191,
      "train_tokens_per_second": 12171.559
    },
    {
      "epoch": 1.7923511984917857,
      "grad_norm": 0.8597403663330097,
      "learning_rate": 2.6379340522496853e-07,
      "loss": 0.2891183376312256,
      "num_input_tokens_seen": 848485704,
      "step": 26620,
      "train_runtime": 69710.3551,
      "train_tokens_per_second": 12171.588
    },
    {
      "epoch": 1.7926878534877457,
      "grad_norm": 0.8495991106187967,
      "learning_rate": 2.6294658293015584e-07,
      "loss": 0.3196293115615845,
      "num_input_tokens_seen": 848649200,
      "step": 26625,
      "train_runtime": 69723.4012,
      "train_tokens_per_second": 12171.655
    },
    {
      "epoch": 1.7930245084837058,
      "grad_norm": 0.9631711133287381,
      "learning_rate": 2.621010853372258e-07,
      "loss": 0.3268672227859497,
      "num_input_tokens_seen": 848806304,
      "step": 26630,
      "train_runtime": 69736.6707,
      "train_tokens_per_second": 12171.592
    },
    {
      "epoch": 1.793361163479666,
      "grad_norm": 0.9387957795925109,
      "learning_rate": 2.612569126826187e-07,
      "loss": 0.31911966800689695,
      "num_input_tokens_seen": 848962744,
      "step": 26635,
      "train_runtime": 69749.7901,
      "train_tokens_per_second": 12171.545
    },
    {
      "epoch": 1.7936978184756263,
      "grad_norm": 0.8178304682726176,
      "learning_rate": 2.6041406520240563e-07,
      "loss": 0.31730635166168214,
      "num_input_tokens_seen": 849124456,
      "step": 26640,
      "train_runtime": 69762.8175,
      "train_tokens_per_second": 12171.591
    },
    {
      "epoch": 1.7940344734715863,
      "grad_norm": 0.9103141172288868,
      "learning_rate": 2.5957254313228595e-07,
      "loss": 0.30668325424194337,
      "num_input_tokens_seen": 849278744,
      "step": 26645,
      "train_runtime": 69775.7198,
      "train_tokens_per_second": 12171.551
    },
    {
      "epoch": 1.7943711284675463,
      "grad_norm": 0.7533510882360192,
      "learning_rate": 2.5873234670758753e-07,
      "loss": 0.26509573459625246,
      "num_input_tokens_seen": 849442584,
      "step": 26650,
      "train_runtime": 69788.1565,
      "train_tokens_per_second": 12171.73
    },
    {
      "epoch": 1.7947077834635066,
      "grad_norm": 0.7524662014517115,
      "learning_rate": 2.578934761632712e-07,
      "loss": 0.2707791566848755,
      "num_input_tokens_seen": 849605800,
      "step": 26655,
      "train_runtime": 69801.0623,
      "train_tokens_per_second": 12171.818
    },
    {
      "epoch": 1.7950444384594668,
      "grad_norm": 0.8772132485207392,
      "learning_rate": 2.570559317339211e-07,
      "loss": 0.30501859188079833,
      "num_input_tokens_seen": 849768344,
      "step": 26660,
      "train_runtime": 69814.5716,
      "train_tokens_per_second": 12171.791
    },
    {
      "epoch": 1.7953810934554268,
      "grad_norm": 0.7625004168181143,
      "learning_rate": 2.56219713653757e-07,
      "loss": 0.28802788257598877,
      "num_input_tokens_seen": 849932088,
      "step": 26665,
      "train_runtime": 69827.0049,
      "train_tokens_per_second": 12171.968
    },
    {
      "epoch": 1.7957177484513869,
      "grad_norm": 0.8772056415708562,
      "learning_rate": 2.5538482215662187e-07,
      "loss": 0.30212740898132323,
      "num_input_tokens_seen": 850090472,
      "step": 26670,
      "train_runtime": 69840.2489,
      "train_tokens_per_second": 12171.928
    },
    {
      "epoch": 1.7960544034473471,
      "grad_norm": 0.8855667963808826,
      "learning_rate": 2.545512574759923e-07,
      "loss": 0.2796017169952393,
      "num_input_tokens_seen": 850252184,
      "step": 26675,
      "train_runtime": 69853.9973,
      "train_tokens_per_second": 12171.847
    },
    {
      "epoch": 1.7963910584433074,
      "grad_norm": 0.8290050021208254,
      "learning_rate": 2.537190198449713e-07,
      "loss": 0.2940730094909668,
      "num_input_tokens_seen": 850408768,
      "step": 26680,
      "train_runtime": 69867.3703,
      "train_tokens_per_second": 12171.759
    },
    {
      "epoch": 1.7967277134392674,
      "grad_norm": 0.8240447804713282,
      "learning_rate": 2.5288810949629207e-07,
      "loss": 0.2990156650543213,
      "num_input_tokens_seen": 850570472,
      "step": 26685,
      "train_runtime": 69880.3496,
      "train_tokens_per_second": 12171.812
    },
    {
      "epoch": 1.7970643684352274,
      "grad_norm": 0.7609781275274295,
      "learning_rate": 2.5205852666231544e-07,
      "loss": 0.27978370189666746,
      "num_input_tokens_seen": 850728848,
      "step": 26690,
      "train_runtime": 69893.6487,
      "train_tokens_per_second": 12171.762
    },
    {
      "epoch": 1.7974010234311877,
      "grad_norm": 0.8347665405452467,
      "learning_rate": 2.5123027157503144e-07,
      "loss": 0.2759535312652588,
      "num_input_tokens_seen": 850885792,
      "step": 26695,
      "train_runtime": 69907.7705,
      "train_tokens_per_second": 12171.548
    },
    {
      "epoch": 1.797737678427148,
      "grad_norm": 0.746905643368537,
      "learning_rate": 2.504033444660592e-07,
      "loss": 0.2952668905258179,
      "num_input_tokens_seen": 851049424,
      "step": 26700,
      "train_runtime": 69920.8115,
      "train_tokens_per_second": 12171.618
    },
    {
      "epoch": 1.798074333423108,
      "grad_norm": 0.7734686036517736,
      "learning_rate": 2.4957774556664706e-07,
      "loss": 0.2850339412689209,
      "num_input_tokens_seen": 851212688,
      "step": 26705,
      "train_runtime": 69933.8341,
      "train_tokens_per_second": 12171.686
    },
    {
      "epoch": 1.798410988419068,
      "grad_norm": 0.8431556528138787,
      "learning_rate": 2.487534751076692e-07,
      "loss": 0.29331364631652834,
      "num_input_tokens_seen": 851374424,
      "step": 26710,
      "train_runtime": 69946.3702,
      "train_tokens_per_second": 12171.817
    },
    {
      "epoch": 1.7987476434150282,
      "grad_norm": 0.834886919573142,
      "learning_rate": 2.47930533319633e-07,
      "loss": 0.2681942701339722,
      "num_input_tokens_seen": 851528336,
      "step": 26715,
      "train_runtime": 69958.8773,
      "train_tokens_per_second": 12171.841
    },
    {
      "epoch": 1.7990842984109885,
      "grad_norm": 0.8432093017256403,
      "learning_rate": 2.4710892043266823e-07,
      "loss": 0.32567903995513914,
      "num_input_tokens_seen": 851690496,
      "step": 26720,
      "train_runtime": 69971.4842,
      "train_tokens_per_second": 12171.966
    },
    {
      "epoch": 1.7994209534069485,
      "grad_norm": 0.8489220961008915,
      "learning_rate": 2.462886366765388e-07,
      "loss": 0.2852949857711792,
      "num_input_tokens_seen": 851852832,
      "step": 26725,
      "train_runtime": 69984.0955,
      "train_tokens_per_second": 12172.092
    },
    {
      "epoch": 1.7997576084029085,
      "grad_norm": 0.8852310269786221,
      "learning_rate": 2.454696822806341e-07,
      "loss": 0.2998690128326416,
      "num_input_tokens_seen": 852010520,
      "step": 26730,
      "train_runtime": 69997.8805,
      "train_tokens_per_second": 12171.947
    },
    {
      "epoch": 1.8000942633988688,
      "grad_norm": 0.7549663852554792,
      "learning_rate": 2.446520574739708e-07,
      "loss": 0.27947320938110354,
      "num_input_tokens_seen": 852171632,
      "step": 26735,
      "train_runtime": 70010.9127,
      "train_tokens_per_second": 12171.983
    },
    {
      "epoch": 1.800430918394829,
      "grad_norm": 0.8860537681445387,
      "learning_rate": 2.438357624851967e-07,
      "loss": 0.26890971660614016,
      "num_input_tokens_seen": 852331712,
      "step": 26740,
      "train_runtime": 70024.192,
      "train_tokens_per_second": 12171.961
    },
    {
      "epoch": 1.800767573390789,
      "grad_norm": 0.7879392209855716,
      "learning_rate": 2.430207975425847e-07,
      "loss": 0.29252939224243163,
      "num_input_tokens_seen": 852492552,
      "step": 26745,
      "train_runtime": 70037.0574,
      "train_tokens_per_second": 12172.021
    },
    {
      "epoch": 1.801104228386749,
      "grad_norm": 0.8344264937951852,
      "learning_rate": 2.4220716287403744e-07,
      "loss": 0.30969891548156736,
      "num_input_tokens_seen": 852656392,
      "step": 26750,
      "train_runtime": 70049.4946,
      "train_tokens_per_second": 12172.199
    },
    {
      "epoch": 1.8014408833827094,
      "grad_norm": 0.7862081879847111,
      "learning_rate": 2.413948587070869e-07,
      "loss": 0.3006144523620605,
      "num_input_tokens_seen": 852807936,
      "step": 26755,
      "train_runtime": 70062.944,
      "train_tokens_per_second": 12172.025
    },
    {
      "epoch": 1.8017775383786696,
      "grad_norm": 0.8554405836798711,
      "learning_rate": 2.405838852688891e-07,
      "loss": 0.3035117149353027,
      "num_input_tokens_seen": 852968656,
      "step": 26760,
      "train_runtime": 70075.3923,
      "train_tokens_per_second": 12172.157
    },
    {
      "epoch": 1.8021141933746296,
      "grad_norm": 0.8904804851752237,
      "learning_rate": 2.397742427862315e-07,
      "loss": 0.2719240427017212,
      "num_input_tokens_seen": 853128136,
      "step": 26765,
      "train_runtime": 70088.3924,
      "train_tokens_per_second": 12172.174
    },
    {
      "epoch": 1.8024508483705897,
      "grad_norm": 0.8915945111092533,
      "learning_rate": 2.3896593148552695e-07,
      "loss": 0.2951150417327881,
      "num_input_tokens_seen": 853280664,
      "step": 26770,
      "train_runtime": 70101.1917,
      "train_tokens_per_second": 12172.128
    },
    {
      "epoch": 1.80278750336655,
      "grad_norm": 0.8245057323119559,
      "learning_rate": 2.3815895159281844e-07,
      "loss": 0.28969459533691405,
      "num_input_tokens_seen": 853442976,
      "step": 26775,
      "train_runtime": 70113.8058,
      "train_tokens_per_second": 12172.253
    },
    {
      "epoch": 1.8031241583625102,
      "grad_norm": 0.8556535117509269,
      "learning_rate": 2.3735330333377493e-07,
      "loss": 0.2846656322479248,
      "num_input_tokens_seen": 853604824,
      "step": 26780,
      "train_runtime": 70126.3535,
      "train_tokens_per_second": 12172.383
    },
    {
      "epoch": 1.8034608133584702,
      "grad_norm": 0.772273224390241,
      "learning_rate": 2.3654898693369232e-07,
      "loss": 0.292094087600708,
      "num_input_tokens_seen": 853764232,
      "step": 26785,
      "train_runtime": 70139.8357,
      "train_tokens_per_second": 12172.316
    },
    {
      "epoch": 1.8037974683544302,
      "grad_norm": 0.8577426238771159,
      "learning_rate": 2.3574600261749735e-07,
      "loss": 0.28982200622558596,
      "num_input_tokens_seen": 853926832,
      "step": 26790,
      "train_runtime": 70154.5461,
      "train_tokens_per_second": 12172.081
    },
    {
      "epoch": 1.8041341233503905,
      "grad_norm": 0.8266505341005959,
      "learning_rate": 2.3494435060973875e-07,
      "loss": 0.2977802038192749,
      "num_input_tokens_seen": 854085640,
      "step": 26795,
      "train_runtime": 70167.098,
      "train_tokens_per_second": 12172.167
    },
    {
      "epoch": 1.8044707783463507,
      "grad_norm": 0.92197930972693,
      "learning_rate": 2.3414403113459937e-07,
      "loss": 0.30182411670684817,
      "num_input_tokens_seen": 854242840,
      "step": 26800,
      "train_runtime": 70180.241,
      "train_tokens_per_second": 12172.127
    },
    {
      "epoch": 1.8048074333423108,
      "grad_norm": 0.8122168619509869,
      "learning_rate": 2.3334504441588413e-07,
      "loss": 0.2700166702270508,
      "num_input_tokens_seen": 854401624,
      "step": 26805,
      "train_runtime": 70192.8185,
      "train_tokens_per_second": 12172.209
    },
    {
      "epoch": 1.8051440883382708,
      "grad_norm": 0.7848662164994366,
      "learning_rate": 2.3254739067702704e-07,
      "loss": 0.28065104484558107,
      "num_input_tokens_seen": 854563432,
      "step": 26810,
      "train_runtime": 70205.3599,
      "train_tokens_per_second": 12172.339
    },
    {
      "epoch": 1.805480743334231,
      "grad_norm": 0.8519509662108244,
      "learning_rate": 2.3175107014109077e-07,
      "loss": 0.26915946006774905,
      "num_input_tokens_seen": 854721024,
      "step": 26815,
      "train_runtime": 70218.0283,
      "train_tokens_per_second": 12172.387
    },
    {
      "epoch": 1.8058173983301913,
      "grad_norm": 0.8552962025029581,
      "learning_rate": 2.3095608303076223e-07,
      "loss": 0.2965372562408447,
      "num_input_tokens_seen": 854882704,
      "step": 26820,
      "train_runtime": 70231.0499,
      "train_tokens_per_second": 12172.432
    },
    {
      "epoch": 1.8061540533261513,
      "grad_norm": 0.8146027501272736,
      "learning_rate": 2.3016242956835798e-07,
      "loss": 0.32484922409057615,
      "num_input_tokens_seen": 855044992,
      "step": 26825,
      "train_runtime": 70243.6594,
      "train_tokens_per_second": 12172.558
    },
    {
      "epoch": 1.8064907083221113,
      "grad_norm": 0.7712852727858308,
      "learning_rate": 2.2937010997581999e-07,
      "loss": 0.2744196891784668,
      "num_input_tokens_seen": 855208256,
      "step": 26830,
      "train_runtime": 70257.3118,
      "train_tokens_per_second": 12172.516
    },
    {
      "epoch": 1.8068273633180716,
      "grad_norm": 0.8824969294482463,
      "learning_rate": 2.2857912447471818e-07,
      "loss": 0.2860698938369751,
      "num_input_tokens_seen": 855360912,
      "step": 26835,
      "train_runtime": 70270.3805,
      "train_tokens_per_second": 12172.425
    },
    {
      "epoch": 1.8071640183140318,
      "grad_norm": 0.7538160364834378,
      "learning_rate": 2.2778947328625012e-07,
      "loss": 0.2795755386352539,
      "num_input_tokens_seen": 855515216,
      "step": 26840,
      "train_runtime": 70283.6717,
      "train_tokens_per_second": 12172.318
    },
    {
      "epoch": 1.8075006733099919,
      "grad_norm": 0.9374565537426578,
      "learning_rate": 2.2700115663123745e-07,
      "loss": 0.3118173360824585,
      "num_input_tokens_seen": 855673800,
      "step": 26845,
      "train_runtime": 70297.1075,
      "train_tokens_per_second": 12172.248
    },
    {
      "epoch": 1.807837328305952,
      "grad_norm": 0.8595840250080625,
      "learning_rate": 2.2621417473013164e-07,
      "loss": 0.27746379375457764,
      "num_input_tokens_seen": 855835848,
      "step": 26850,
      "train_runtime": 70310.2749,
      "train_tokens_per_second": 12172.273
    },
    {
      "epoch": 1.8081739833019121,
      "grad_norm": 0.9068871418435703,
      "learning_rate": 2.254285278030094e-07,
      "loss": 0.28034520149230957,
      "num_input_tokens_seen": 855995944,
      "step": 26855,
      "train_runtime": 70323.6377,
      "train_tokens_per_second": 12172.236
    },
    {
      "epoch": 1.8085106382978724,
      "grad_norm": 0.8317324688611203,
      "learning_rate": 2.24644216069575e-07,
      "loss": 0.29037060737609866,
      "num_input_tokens_seen": 856148896,
      "step": 26860,
      "train_runtime": 70336.3185,
      "train_tokens_per_second": 12172.216
    },
    {
      "epoch": 1.8088472932938324,
      "grad_norm": 0.7757104198167448,
      "learning_rate": 2.238612397491574e-07,
      "loss": 0.2824215888977051,
      "num_input_tokens_seen": 856310856,
      "step": 26865,
      "train_runtime": 70348.8927,
      "train_tokens_per_second": 12172.343
    },
    {
      "epoch": 1.8091839482897925,
      "grad_norm": 0.83949279553892,
      "learning_rate": 2.230795990607143e-07,
      "loss": 0.292067289352417,
      "num_input_tokens_seen": 856473144,
      "step": 26870,
      "train_runtime": 70361.3229,
      "train_tokens_per_second": 12172.499
    },
    {
      "epoch": 1.809520603285753,
      "grad_norm": 0.9379394170247293,
      "learning_rate": 2.2229929422282915e-07,
      "loss": 0.28386387825012205,
      "num_input_tokens_seen": 856630024,
      "step": 26875,
      "train_runtime": 70374.9725,
      "train_tokens_per_second": 12172.367
    },
    {
      "epoch": 1.809857258281713,
      "grad_norm": 0.8766440839778201,
      "learning_rate": 2.215203254537107e-07,
      "loss": 0.3175637245178223,
      "num_input_tokens_seen": 856790704,
      "step": 26880,
      "train_runtime": 70388.26,
      "train_tokens_per_second": 12172.352
    },
    {
      "epoch": 1.810193913277673,
      "grad_norm": 0.7867602307389877,
      "learning_rate": 2.2074269297119588e-07,
      "loss": 0.3082930088043213,
      "num_input_tokens_seen": 856952152,
      "step": 26885,
      "train_runtime": 70401.8576,
      "train_tokens_per_second": 12172.295
    },
    {
      "epoch": 1.8105305682736332,
      "grad_norm": 0.8274595863706828,
      "learning_rate": 2.199663969927479e-07,
      "loss": 0.2812648296356201,
      "num_input_tokens_seen": 857111688,
      "step": 26890,
      "train_runtime": 70415.3642,
      "train_tokens_per_second": 12172.225
    },
    {
      "epoch": 1.8108672232695935,
      "grad_norm": 1.0489949372173941,
      "learning_rate": 2.1919143773545316e-07,
      "loss": 0.2942085266113281,
      "num_input_tokens_seen": 857273320,
      "step": 26895,
      "train_runtime": 70429.0447,
      "train_tokens_per_second": 12172.156
    },
    {
      "epoch": 1.8112038782655535,
      "grad_norm": 1.6175586818236811,
      "learning_rate": 2.184178154160288e-07,
      "loss": 0.3006558418273926,
      "num_input_tokens_seen": 857427400,
      "step": 26900,
      "train_runtime": 70441.6398,
      "train_tokens_per_second": 12172.167
    },
    {
      "epoch": 1.8115405332615135,
      "grad_norm": 0.7812805752391684,
      "learning_rate": 2.1764553025081514e-07,
      "loss": 0.2858154773712158,
      "num_input_tokens_seen": 857589792,
      "step": 26905,
      "train_runtime": 70456.1148,
      "train_tokens_per_second": 12171.971
    },
    {
      "epoch": 1.8118771882574738,
      "grad_norm": 0.963713689849728,
      "learning_rate": 2.168745824557794e-07,
      "loss": 0.2870452642440796,
      "num_input_tokens_seen": 857747496,
      "step": 26910,
      "train_runtime": 70469.1924,
      "train_tokens_per_second": 12171.95
    },
    {
      "epoch": 1.812213843253434,
      "grad_norm": 0.8174060605646627,
      "learning_rate": 2.161049722465136e-07,
      "loss": 0.2772712230682373,
      "num_input_tokens_seen": 857908232,
      "step": 26915,
      "train_runtime": 70481.6191,
      "train_tokens_per_second": 12172.085
    },
    {
      "epoch": 1.812550498249394,
      "grad_norm": 0.8383726227129716,
      "learning_rate": 2.1533669983823835e-07,
      "loss": 0.29603776931762693,
      "num_input_tokens_seen": 858072072,
      "step": 26920,
      "train_runtime": 70494.0458,
      "train_tokens_per_second": 12172.263
    },
    {
      "epoch": 1.812887153245354,
      "grad_norm": 0.7308651503164498,
      "learning_rate": 2.1456976544579732e-07,
      "loss": 0.27979373931884766,
      "num_input_tokens_seen": 858235912,
      "step": 26925,
      "train_runtime": 70506.4791,
      "train_tokens_per_second": 12172.44
    },
    {
      "epoch": 1.8132238082413143,
      "grad_norm": 0.767349085594718,
      "learning_rate": 2.138041692836623e-07,
      "loss": 0.295100736618042,
      "num_input_tokens_seen": 858394888,
      "step": 26930,
      "train_runtime": 70519.6022,
      "train_tokens_per_second": 12172.43
    },
    {
      "epoch": 1.8135604632372746,
      "grad_norm": 0.9164780369529396,
      "learning_rate": 2.1303991156592818e-07,
      "loss": 0.3028141975402832,
      "num_input_tokens_seen": 858553216,
      "step": 26935,
      "train_runtime": 70532.7465,
      "train_tokens_per_second": 12172.406
    },
    {
      "epoch": 1.8138971182332346,
      "grad_norm": 0.8682771103575873,
      "learning_rate": 2.122769925063195e-07,
      "loss": 0.2896735191345215,
      "num_input_tokens_seen": 858712496,
      "step": 26940,
      "train_runtime": 70545.6425,
      "train_tokens_per_second": 12172.439
    },
    {
      "epoch": 1.8142337732291947,
      "grad_norm": 0.8867369246284782,
      "learning_rate": 2.1151541231818174e-07,
      "loss": 0.28209409713745115,
      "num_input_tokens_seen": 858873528,
      "step": 26945,
      "train_runtime": 70558.8065,
      "train_tokens_per_second": 12172.45
    },
    {
      "epoch": 1.814570428225155,
      "grad_norm": 0.854163509027148,
      "learning_rate": 2.1075517121449063e-07,
      "loss": 0.3170738220214844,
      "num_input_tokens_seen": 859031216,
      "step": 26950,
      "train_runtime": 70571.5282,
      "train_tokens_per_second": 12172.49
    },
    {
      "epoch": 1.8149070832211152,
      "grad_norm": 0.908505366459971,
      "learning_rate": 2.0999626940784334e-07,
      "loss": 0.26206698417663576,
      "num_input_tokens_seen": 859191872,
      "step": 26955,
      "train_runtime": 70584.9612,
      "train_tokens_per_second": 12172.449
    },
    {
      "epoch": 1.8152437382170752,
      "grad_norm": 0.9832339802011741,
      "learning_rate": 2.0923870711046568e-07,
      "loss": 0.297837495803833,
      "num_input_tokens_seen": 859353216,
      "step": 26960,
      "train_runtime": 70598.6674,
      "train_tokens_per_second": 12172.372
    },
    {
      "epoch": 1.8155803932130352,
      "grad_norm": 0.7283856366433215,
      "learning_rate": 2.0848248453420704e-07,
      "loss": 0.27922711372375486,
      "num_input_tokens_seen": 859514096,
      "step": 26965,
      "train_runtime": 70611.5434,
      "train_tokens_per_second": 12172.43
    },
    {
      "epoch": 1.8159170482089955,
      "grad_norm": 0.8231284588693126,
      "learning_rate": 2.077276018905422e-07,
      "loss": 0.2874908924102783,
      "num_input_tokens_seen": 859675656,
      "step": 26970,
      "train_runtime": 70624.6661,
      "train_tokens_per_second": 12172.456
    },
    {
      "epoch": 1.8162537032049557,
      "grad_norm": 0.8546622526273433,
      "learning_rate": 2.0697405939057225e-07,
      "loss": 0.2974217176437378,
      "num_input_tokens_seen": 859827240,
      "step": 26975,
      "train_runtime": 70638.0449,
      "train_tokens_per_second": 12172.297
    },
    {
      "epoch": 1.8165903582009157,
      "grad_norm": 0.7666640566728249,
      "learning_rate": 2.0622185724502253e-07,
      "loss": 0.2839369535446167,
      "num_input_tokens_seen": 859985024,
      "step": 26980,
      "train_runtime": 70651.3617,
      "train_tokens_per_second": 12172.236
    },
    {
      "epoch": 1.8169270131968758,
      "grad_norm": 0.7860437404352247,
      "learning_rate": 2.0547099566424367e-07,
      "loss": 0.2720181465148926,
      "num_input_tokens_seen": 860148560,
      "step": 26985,
      "train_runtime": 70664.3772,
      "train_tokens_per_second": 12172.308
    },
    {
      "epoch": 1.817263668192836,
      "grad_norm": 0.8569469413834617,
      "learning_rate": 2.0472147485821325e-07,
      "loss": 0.2900655269622803,
      "num_input_tokens_seen": 860306064,
      "step": 26990,
      "train_runtime": 70678.8148,
      "train_tokens_per_second": 12172.05
    },
    {
      "epoch": 1.8176003231887963,
      "grad_norm": 0.9471079522298579,
      "learning_rate": 2.0397329503653084e-07,
      "loss": 0.2749172687530518,
      "num_input_tokens_seen": 860464344,
      "step": 26995,
      "train_runtime": 70691.7561,
      "train_tokens_per_second": 12172.061
    },
    {
      "epoch": 1.8179369781847563,
      "grad_norm": 0.8197566014626608,
      "learning_rate": 2.03226456408423e-07,
      "loss": 0.3060157299041748,
      "num_input_tokens_seen": 860616656,
      "step": 27000,
      "train_runtime": 70705.2216,
      "train_tokens_per_second": 12171.897
    },
    {
      "epoch": 1.8182736331807163,
      "grad_norm": 0.7698219813025884,
      "learning_rate": 2.024809591827409e-07,
      "loss": 0.29669058322906494,
      "num_input_tokens_seen": 860779640,
      "step": 27005,
      "train_runtime": 70718.7237,
      "train_tokens_per_second": 12171.877
    },
    {
      "epoch": 1.8186102881766766,
      "grad_norm": 0.8284426152034386,
      "learning_rate": 2.0173680356796066e-07,
      "loss": 0.27449684143066405,
      "num_input_tokens_seen": 860935504,
      "step": 27010,
      "train_runtime": 70732.6553,
      "train_tokens_per_second": 12171.684
    },
    {
      "epoch": 1.8189469431726368,
      "grad_norm": 0.7858228479675196,
      "learning_rate": 2.00993989772183e-07,
      "loss": 0.29461596012115476,
      "num_input_tokens_seen": 861094576,
      "step": 27015,
      "train_runtime": 70745.8136,
      "train_tokens_per_second": 12171.668
    },
    {
      "epoch": 1.8192835981685969,
      "grad_norm": 0.7830966772800374,
      "learning_rate": 2.0025251800313284e-07,
      "loss": 0.31585776805877686,
      "num_input_tokens_seen": 861252496,
      "step": 27020,
      "train_runtime": 70759.0538,
      "train_tokens_per_second": 12171.623
    },
    {
      "epoch": 1.8196202531645569,
      "grad_norm": 0.7890415898640881,
      "learning_rate": 1.9951238846816045e-07,
      "loss": 0.30937507152557375,
      "num_input_tokens_seen": 861416216,
      "step": 27025,
      "train_runtime": 70771.4763,
      "train_tokens_per_second": 12171.799
    },
    {
      "epoch": 1.8199569081605171,
      "grad_norm": 0.7877864798480069,
      "learning_rate": 1.987736013742414e-07,
      "loss": 0.29053387641906736,
      "num_input_tokens_seen": 861571560,
      "step": 27030,
      "train_runtime": 70784.5743,
      "train_tokens_per_second": 12171.742
    },
    {
      "epoch": 1.8202935631564774,
      "grad_norm": 0.8178219373525091,
      "learning_rate": 1.9803615692797375e-07,
      "loss": 0.27255997657775877,
      "num_input_tokens_seen": 861731472,
      "step": 27035,
      "train_runtime": 70797.2511,
      "train_tokens_per_second": 12171.821
    },
    {
      "epoch": 1.8206302181524374,
      "grad_norm": 0.7285418331981861,
      "learning_rate": 1.9730005533558417e-07,
      "loss": 0.27718803882598875,
      "num_input_tokens_seen": 861892720,
      "step": 27040,
      "train_runtime": 70809.7353,
      "train_tokens_per_second": 12171.952
    },
    {
      "epoch": 1.8209668731483974,
      "grad_norm": 0.72801899341444,
      "learning_rate": 1.9656529680291802e-07,
      "loss": 0.2734086036682129,
      "num_input_tokens_seen": 862053848,
      "step": 27045,
      "train_runtime": 70823.3548,
      "train_tokens_per_second": 12171.887
    },
    {
      "epoch": 1.8213035281443577,
      "grad_norm": 0.8027209417211447,
      "learning_rate": 1.9583188153544986e-07,
      "loss": 0.2741126775741577,
      "num_input_tokens_seen": 862207736,
      "step": 27050,
      "train_runtime": 70836.0261,
      "train_tokens_per_second": 12171.882
    },
    {
      "epoch": 1.821640183140318,
      "grad_norm": 0.8287585620063663,
      "learning_rate": 1.9509980973827724e-07,
      "loss": 0.2951030969619751,
      "num_input_tokens_seen": 862369176,
      "step": 27055,
      "train_runtime": 70849.7478,
      "train_tokens_per_second": 12171.803
    },
    {
      "epoch": 1.821976838136278,
      "grad_norm": 0.8445685786220319,
      "learning_rate": 1.943690816161209e-07,
      "loss": 0.30646162033081054,
      "num_input_tokens_seen": 862529032,
      "step": 27060,
      "train_runtime": 70862.4304,
      "train_tokens_per_second": 12171.88
    },
    {
      "epoch": 1.822313493132238,
      "grad_norm": 0.8443710592610861,
      "learning_rate": 1.9363969737332677e-07,
      "loss": 0.2744742870330811,
      "num_input_tokens_seen": 862687952,
      "step": 27065,
      "train_runtime": 70874.9777,
      "train_tokens_per_second": 12171.968
    },
    {
      "epoch": 1.8226501481281983,
      "grad_norm": 0.6975098821958919,
      "learning_rate": 1.9291165721386507e-07,
      "loss": 0.2670555591583252,
      "num_input_tokens_seen": 862842960,
      "step": 27070,
      "train_runtime": 70888.5189,
      "train_tokens_per_second": 12171.829
    },
    {
      "epoch": 1.8229868031241585,
      "grad_norm": 0.7878061834787141,
      "learning_rate": 1.9218496134132957e-07,
      "loss": 0.3179195404052734,
      "num_input_tokens_seen": 863002136,
      "step": 27075,
      "train_runtime": 70901.5271,
      "train_tokens_per_second": 12171.841
    },
    {
      "epoch": 1.8233234581201185,
      "grad_norm": 0.7916837964782856,
      "learning_rate": 1.9145960995893831e-07,
      "loss": 0.2753622055053711,
      "num_input_tokens_seen": 863158112,
      "step": 27080,
      "train_runtime": 70914.0261,
      "train_tokens_per_second": 12171.895
    },
    {
      "epoch": 1.8236601131160786,
      "grad_norm": 0.8246354755762004,
      "learning_rate": 1.9073560326953344e-07,
      "loss": 0.2880223274230957,
      "num_input_tokens_seen": 863318592,
      "step": 27085,
      "train_runtime": 70926.6211,
      "train_tokens_per_second": 12171.997
    },
    {
      "epoch": 1.8239967681120388,
      "grad_norm": 0.8622359354830219,
      "learning_rate": 1.9001294147558303e-07,
      "loss": 0.30566768646240233,
      "num_input_tokens_seen": 863474168,
      "step": 27090,
      "train_runtime": 70939.7161,
      "train_tokens_per_second": 12171.943
    },
    {
      "epoch": 1.824333423107999,
      "grad_norm": 0.8378015478811262,
      "learning_rate": 1.892916247791743e-07,
      "loss": 0.30236384868621824,
      "num_input_tokens_seen": 863631352,
      "step": 27095,
      "train_runtime": 70952.6254,
      "train_tokens_per_second": 12171.944
    },
    {
      "epoch": 1.824670078103959,
      "grad_norm": 0.8060336163800181,
      "learning_rate": 1.885716533820231e-07,
      "loss": 0.28802154064178465,
      "num_input_tokens_seen": 863792936,
      "step": 27100,
      "train_runtime": 70965.3796,
      "train_tokens_per_second": 12172.033
    },
    {
      "epoch": 1.8250067330999191,
      "grad_norm": 0.8323050353956888,
      "learning_rate": 1.878530274854662e-07,
      "loss": 0.29305567741394045,
      "num_input_tokens_seen": 863953968,
      "step": 27105,
      "train_runtime": 70978.3294,
      "train_tokens_per_second": 12172.081
    },
    {
      "epoch": 1.8253433880958794,
      "grad_norm": 0.7824473206246785,
      "learning_rate": 1.8713574729046557e-07,
      "loss": 0.30150279998779295,
      "num_input_tokens_seen": 864114440,
      "step": 27110,
      "train_runtime": 70990.7776,
      "train_tokens_per_second": 12172.207
    },
    {
      "epoch": 1.8256800430918396,
      "grad_norm": 0.8605467319914241,
      "learning_rate": 1.8641981299760582e-07,
      "loss": 0.2926782608032227,
      "num_input_tokens_seen": 864269000,
      "step": 27115,
      "train_runtime": 71004.3929,
      "train_tokens_per_second": 12172.05
    },
    {
      "epoch": 1.8260166980877997,
      "grad_norm": 0.838649902523344,
      "learning_rate": 1.857052248070962e-07,
      "loss": 0.29631705284118653,
      "num_input_tokens_seen": 864429408,
      "step": 27120,
      "train_runtime": 71017.7441,
      "train_tokens_per_second": 12172.02
    },
    {
      "epoch": 1.8263533530837597,
      "grad_norm": 0.9731379689988796,
      "learning_rate": 1.8499198291876907e-07,
      "loss": 0.27010622024536135,
      "num_input_tokens_seen": 864588032,
      "step": 27125,
      "train_runtime": 71030.9605,
      "train_tokens_per_second": 12171.988
    },
    {
      "epoch": 1.82669000807972,
      "grad_norm": 0.713343469981639,
      "learning_rate": 1.8428008753207937e-07,
      "loss": 0.30358912944793703,
      "num_input_tokens_seen": 864750528,
      "step": 27130,
      "train_runtime": 71044.7812,
      "train_tokens_per_second": 12171.908
    },
    {
      "epoch": 1.8270266630756802,
      "grad_norm": 0.8260824296237773,
      "learning_rate": 1.8356953884610673e-07,
      "loss": 0.30570759773254397,
      "num_input_tokens_seen": 864914120,
      "step": 27135,
      "train_runtime": 71058.475,
      "train_tokens_per_second": 12171.864
    },
    {
      "epoch": 1.8273633180716402,
      "grad_norm": 0.8570184211488898,
      "learning_rate": 1.8286033705955554e-07,
      "loss": 0.28254094123840334,
      "num_input_tokens_seen": 865076880,
      "step": 27140,
      "train_runtime": 71071.5496,
      "train_tokens_per_second": 12171.915
    },
    {
      "epoch": 1.8276999730676002,
      "grad_norm": 0.8343303100347506,
      "learning_rate": 1.8215248237074879e-07,
      "loss": 0.29530584812164307,
      "num_input_tokens_seen": 865239400,
      "step": 27145,
      "train_runtime": 71085.7976,
      "train_tokens_per_second": 12171.762
    },
    {
      "epoch": 1.8280366280635605,
      "grad_norm": 0.7009581422536433,
      "learning_rate": 1.8144597497763816e-07,
      "loss": 0.2859818696975708,
      "num_input_tokens_seen": 865390984,
      "step": 27150,
      "train_runtime": 71098.207,
      "train_tokens_per_second": 12171.769
    },
    {
      "epoch": 1.8283732830595207,
      "grad_norm": 0.7674281128069027,
      "learning_rate": 1.8074081507779506e-07,
      "loss": 0.27884206771850584,
      "num_input_tokens_seen": 865554824,
      "step": 27155,
      "train_runtime": 71110.6173,
      "train_tokens_per_second": 12171.949
    },
    {
      "epoch": 1.8287099380554808,
      "grad_norm": 0.772589477624742,
      "learning_rate": 1.8003700286841563e-07,
      "loss": 0.3010308504104614,
      "num_input_tokens_seen": 865717416,
      "step": 27160,
      "train_runtime": 71123.9511,
      "train_tokens_per_second": 12171.953
    },
    {
      "epoch": 1.8290465930514408,
      "grad_norm": 0.8344049154720434,
      "learning_rate": 1.793345385463191e-07,
      "loss": 0.30673604011535643,
      "num_input_tokens_seen": 865878976,
      "step": 27165,
      "train_runtime": 71137.0557,
      "train_tokens_per_second": 12171.982
    },
    {
      "epoch": 1.829383248047401,
      "grad_norm": 0.8267761204159912,
      "learning_rate": 1.7863342230794667e-07,
      "loss": 0.31325130462646483,
      "num_input_tokens_seen": 866037416,
      "step": 27170,
      "train_runtime": 71150.231,
      "train_tokens_per_second": 12171.955
    },
    {
      "epoch": 1.8297199030433613,
      "grad_norm": 0.7932352032205876,
      "learning_rate": 1.7793365434936316e-07,
      "loss": 0.2942698001861572,
      "num_input_tokens_seen": 866200624,
      "step": 27175,
      "train_runtime": 71163.1174,
      "train_tokens_per_second": 12172.044
    },
    {
      "epoch": 1.8300565580393213,
      "grad_norm": 0.84363682521872,
      "learning_rate": 1.7723523486625704e-07,
      "loss": 0.2991994857788086,
      "num_input_tokens_seen": 866355792,
      "step": 27180,
      "train_runtime": 71176.7025,
      "train_tokens_per_second": 12171.901
    },
    {
      "epoch": 1.8303932130352814,
      "grad_norm": 0.8049487824964443,
      "learning_rate": 1.7653816405393766e-07,
      "loss": 0.3009037017822266,
      "num_input_tokens_seen": 866514768,
      "step": 27185,
      "train_runtime": 71189.9325,
      "train_tokens_per_second": 12171.872
    },
    {
      "epoch": 1.8307298680312416,
      "grad_norm": 0.8832984858639972,
      "learning_rate": 1.758424421073418e-07,
      "loss": 0.29663715362548826,
      "num_input_tokens_seen": 866676288,
      "step": 27190,
      "train_runtime": 71202.4249,
      "train_tokens_per_second": 12172.005
    },
    {
      "epoch": 1.8310665230272019,
      "grad_norm": 0.7787264884003384,
      "learning_rate": 1.7514806922102222e-07,
      "loss": 0.28052749633789065,
      "num_input_tokens_seen": 866830216,
      "step": 27195,
      "train_runtime": 71216.3233,
      "train_tokens_per_second": 12171.791
    },
    {
      "epoch": 1.8314031780231619,
      "grad_norm": 0.8618031279864221,
      "learning_rate": 1.7445504558916027e-07,
      "loss": 0.27704639434814454,
      "num_input_tokens_seen": 866988416,
      "step": 27200,
      "train_runtime": 71230.1039,
      "train_tokens_per_second": 12171.657
    },
    {
      "epoch": 1.831739833019122,
      "grad_norm": 0.8402533145130692,
      "learning_rate": 1.7376337140555754e-07,
      "loss": 0.2793675661087036,
      "num_input_tokens_seen": 867147880,
      "step": 27205,
      "train_runtime": 71243.5687,
      "train_tokens_per_second": 12171.595
    },
    {
      "epoch": 1.8320764880150822,
      "grad_norm": 0.8377721991737594,
      "learning_rate": 1.7307304686363825e-07,
      "loss": 0.28644700050354005,
      "num_input_tokens_seen": 867307248,
      "step": 27210,
      "train_runtime": 71257.0334,
      "train_tokens_per_second": 12171.532
    },
    {
      "epoch": 1.8324131430110424,
      "grad_norm": 0.7719629430588928,
      "learning_rate": 1.723840721564496e-07,
      "loss": 0.2700688362121582,
      "num_input_tokens_seen": 867470088,
      "step": 27215,
      "train_runtime": 71269.4752,
      "train_tokens_per_second": 12171.692
    },
    {
      "epoch": 1.8327497980070024,
      "grad_norm": 0.7966758936086434,
      "learning_rate": 1.7169644747666081e-07,
      "loss": 0.27162020206451415,
      "num_input_tokens_seen": 867630056,
      "step": 27220,
      "train_runtime": 71282.4986,
      "train_tokens_per_second": 12171.712
    },
    {
      "epoch": 1.8330864530029625,
      "grad_norm": 0.8698336350133569,
      "learning_rate": 1.710101730165642e-07,
      "loss": 0.28916010856628416,
      "num_input_tokens_seen": 867791136,
      "step": 27225,
      "train_runtime": 71295.0016,
      "train_tokens_per_second": 12171.837
    },
    {
      "epoch": 1.8334231079989227,
      "grad_norm": 0.7624470016255673,
      "learning_rate": 1.7032524896807402e-07,
      "loss": 0.26928303241729734,
      "num_input_tokens_seen": 867948760,
      "step": 27230,
      "train_runtime": 71309.272,
      "train_tokens_per_second": 12171.612
    },
    {
      "epoch": 1.833759762994883,
      "grad_norm": 0.8232484234653391,
      "learning_rate": 1.6964167552272702e-07,
      "loss": 0.30181159973144533,
      "num_input_tokens_seen": 868107912,
      "step": 27235,
      "train_runtime": 71322.8292,
      "train_tokens_per_second": 12171.529
    },
    {
      "epoch": 1.834096417990843,
      "grad_norm": 0.8877054460294689,
      "learning_rate": 1.6895945287168314e-07,
      "loss": 0.310361385345459,
      "num_input_tokens_seen": 868267312,
      "step": 27240,
      "train_runtime": 71336.861,
      "train_tokens_per_second": 12171.37
    },
    {
      "epoch": 1.834433072986803,
      "grad_norm": 0.921462805687509,
      "learning_rate": 1.682785812057225e-07,
      "loss": 0.30920722484588625,
      "num_input_tokens_seen": 868418360,
      "step": 27245,
      "train_runtime": 71349.6333,
      "train_tokens_per_second": 12171.308
    },
    {
      "epoch": 1.8347697279827633,
      "grad_norm": 0.846711529164498,
      "learning_rate": 1.6759906071525e-07,
      "loss": 0.2906870126724243,
      "num_input_tokens_seen": 868579528,
      "step": 27250,
      "train_runtime": 71363.4556,
      "train_tokens_per_second": 12171.209
    },
    {
      "epoch": 1.8351063829787235,
      "grad_norm": 0.922495521749728,
      "learning_rate": 1.669208915902909e-07,
      "loss": 0.33587601184844973,
      "num_input_tokens_seen": 868743368,
      "step": 27255,
      "train_runtime": 71375.881,
      "train_tokens_per_second": 12171.386
    },
    {
      "epoch": 1.8354430379746836,
      "grad_norm": 0.7516951150734633,
      "learning_rate": 1.6624407402049237e-07,
      "loss": 0.27452590465545657,
      "num_input_tokens_seen": 868902072,
      "step": 27260,
      "train_runtime": 71389.8614,
      "train_tokens_per_second": 12171.225
    },
    {
      "epoch": 1.8357796929706436,
      "grad_norm": 0.8144726842322143,
      "learning_rate": 1.655686081951252e-07,
      "loss": 0.2877694606781006,
      "num_input_tokens_seen": 869058184,
      "step": 27265,
      "train_runtime": 71402.4853,
      "train_tokens_per_second": 12171.26
    },
    {
      "epoch": 1.8361163479666038,
      "grad_norm": 0.9652416625952254,
      "learning_rate": 1.6489449430308058e-07,
      "loss": 0.3203382968902588,
      "num_input_tokens_seen": 869219888,
      "step": 27270,
      "train_runtime": 71415.6469,
      "train_tokens_per_second": 12171.281
    },
    {
      "epoch": 1.836453002962564,
      "grad_norm": 0.8147954320205363,
      "learning_rate": 1.6422173253287323e-07,
      "loss": 0.3085162162780762,
      "num_input_tokens_seen": 869383728,
      "step": 27275,
      "train_runtime": 71428.0705,
      "train_tokens_per_second": 12171.458
    },
    {
      "epoch": 1.8367896579585241,
      "grad_norm": 1.0029804878846063,
      "learning_rate": 1.6355032307263764e-07,
      "loss": 0.2888693332672119,
      "num_input_tokens_seen": 869543632,
      "step": 27280,
      "train_runtime": 71440.7871,
      "train_tokens_per_second": 12171.529
    },
    {
      "epoch": 1.8371263129544841,
      "grad_norm": 0.8176198685320356,
      "learning_rate": 1.628802661101314e-07,
      "loss": 0.2625081539154053,
      "num_input_tokens_seen": 869699504,
      "step": 27285,
      "train_runtime": 71453.7781,
      "train_tokens_per_second": 12171.498
    },
    {
      "epoch": 1.8374629679504444,
      "grad_norm": 0.753889105424598,
      "learning_rate": 1.6221156183273524e-07,
      "loss": 0.2896974802017212,
      "num_input_tokens_seen": 869862688,
      "step": 27290,
      "train_runtime": 71466.6475,
      "train_tokens_per_second": 12171.589
    },
    {
      "epoch": 1.8377996229464046,
      "grad_norm": 0.8992659563319446,
      "learning_rate": 1.615442104274484e-07,
      "loss": 0.30241103172302247,
      "num_input_tokens_seen": 870023304,
      "step": 27295,
      "train_runtime": 71479.0831,
      "train_tokens_per_second": 12171.719
    },
    {
      "epoch": 1.8381362779423647,
      "grad_norm": 0.9794266627502588,
      "learning_rate": 1.6087821208089492e-07,
      "loss": 0.3027640819549561,
      "num_input_tokens_seen": 870184600,
      "step": 27300,
      "train_runtime": 71491.4919,
      "train_tokens_per_second": 12171.862
    },
    {
      "epoch": 1.8384729329383247,
      "grad_norm": 0.7482425865694278,
      "learning_rate": 1.6021356697931866e-07,
      "loss": 0.2621957778930664,
      "num_input_tokens_seen": 870347520,
      "step": 27305,
      "train_runtime": 71504.9509,
      "train_tokens_per_second": 12171.85
    },
    {
      "epoch": 1.838809587934285,
      "grad_norm": 0.7781524330733605,
      "learning_rate": 1.5955027530858537e-07,
      "loss": 0.2564433813095093,
      "num_input_tokens_seen": 870507552,
      "step": 27310,
      "train_runtime": 71517.9568,
      "train_tokens_per_second": 12171.874
    },
    {
      "epoch": 1.8391462429302452,
      "grad_norm": 0.7501029699322518,
      "learning_rate": 1.5888833725418275e-07,
      "loss": 0.2613529682159424,
      "num_input_tokens_seen": 870667560,
      "step": 27315,
      "train_runtime": 71531.1477,
      "train_tokens_per_second": 12171.866
    },
    {
      "epoch": 1.8394828979262052,
      "grad_norm": 0.8239663025035118,
      "learning_rate": 1.582277530012194e-07,
      "loss": 0.26694355010986326,
      "num_input_tokens_seen": 870829224,
      "step": 27320,
      "train_runtime": 71543.6367,
      "train_tokens_per_second": 12172.001
    },
    {
      "epoch": 1.8398195529221653,
      "grad_norm": 0.9116417283775063,
      "learning_rate": 1.5756852273442536e-07,
      "loss": 0.29428849220275877,
      "num_input_tokens_seen": 870984408,
      "step": 27325,
      "train_runtime": 71556.6518,
      "train_tokens_per_second": 12171.956
    },
    {
      "epoch": 1.8401562079181255,
      "grad_norm": 0.7737270867876666,
      "learning_rate": 1.569106466381526e-07,
      "loss": 0.3038698673248291,
      "num_input_tokens_seen": 871145784,
      "step": 27330,
      "train_runtime": 71570.2242,
      "train_tokens_per_second": 12171.902
    },
    {
      "epoch": 1.8404928629140858,
      "grad_norm": 0.7782434940868362,
      "learning_rate": 1.5625412489637337e-07,
      "loss": 0.27227582931518557,
      "num_input_tokens_seen": 871305880,
      "step": 27335,
      "train_runtime": 71582.6391,
      "train_tokens_per_second": 12172.028
    },
    {
      "epoch": 1.8408295179100458,
      "grad_norm": 0.8889795593827744,
      "learning_rate": 1.555989576926842e-07,
      "loss": 0.29062957763671876,
      "num_input_tokens_seen": 871469720,
      "step": 27340,
      "train_runtime": 71595.0706,
      "train_tokens_per_second": 12172.203
    },
    {
      "epoch": 1.8411661729060058,
      "grad_norm": 0.8594452637531506,
      "learning_rate": 1.5494514521029736e-07,
      "loss": 0.2943892478942871,
      "num_input_tokens_seen": 871631744,
      "step": 27345,
      "train_runtime": 71607.6637,
      "train_tokens_per_second": 12172.325
    },
    {
      "epoch": 1.841502827901966,
      "grad_norm": 0.7905380407009803,
      "learning_rate": 1.5429268763205108e-07,
      "loss": 0.28862152099609373,
      "num_input_tokens_seen": 871792472,
      "step": 27350,
      "train_runtime": 71620.4294,
      "train_tokens_per_second": 12172.399
    },
    {
      "epoch": 1.8418394828979263,
      "grad_norm": 0.7986330508974063,
      "learning_rate": 1.5364158514040328e-07,
      "loss": 0.2979848623275757,
      "num_input_tokens_seen": 871952536,
      "step": 27355,
      "train_runtime": 71633.171,
      "train_tokens_per_second": 12172.469
    },
    {
      "epoch": 1.8421761378938863,
      "grad_norm": 1.0396200796633621,
      "learning_rate": 1.5299183791743223e-07,
      "loss": 0.2980222225189209,
      "num_input_tokens_seen": 872106840,
      "step": 27360,
      "train_runtime": 71646.6097,
      "train_tokens_per_second": 12172.339
    },
    {
      "epoch": 1.8425127928898464,
      "grad_norm": 0.9223987932403381,
      "learning_rate": 1.523434461448381e-07,
      "loss": 0.31063032150268555,
      "num_input_tokens_seen": 872270680,
      "step": 27365,
      "train_runtime": 71659.0347,
      "train_tokens_per_second": 12172.515
    },
    {
      "epoch": 1.8428494478858066,
      "grad_norm": 0.7906190716627437,
      "learning_rate": 1.5169641000394086e-07,
      "loss": 0.2921443462371826,
      "num_input_tokens_seen": 872434336,
      "step": 27370,
      "train_runtime": 71672.1503,
      "train_tokens_per_second": 12172.571
    },
    {
      "epoch": 1.8431861028817669,
      "grad_norm": 0.8057299439125879,
      "learning_rate": 1.5105072967568246e-07,
      "loss": 0.2900339126586914,
      "num_input_tokens_seen": 872591552,
      "step": 27375,
      "train_runtime": 71684.9227,
      "train_tokens_per_second": 12172.595
    },
    {
      "epoch": 1.843522757877727,
      "grad_norm": 0.7175964867064437,
      "learning_rate": 1.504064053406251e-07,
      "loss": 0.29469482898712157,
      "num_input_tokens_seen": 872753232,
      "step": 27380,
      "train_runtime": 71698.5611,
      "train_tokens_per_second": 12172.535
    },
    {
      "epoch": 1.843859412873687,
      "grad_norm": 0.9335406914726482,
      "learning_rate": 1.4976343717895192e-07,
      "loss": 0.300518274307251,
      "num_input_tokens_seen": 872914872,
      "step": 27385,
      "train_runtime": 71711.5613,
      "train_tokens_per_second": 12172.582
    },
    {
      "epoch": 1.8441960678696472,
      "grad_norm": 0.8167853420517825,
      "learning_rate": 1.491218253704685e-07,
      "loss": 0.2991645097732544,
      "num_input_tokens_seen": 873077648,
      "step": 27390,
      "train_runtime": 71724.2979,
      "train_tokens_per_second": 12172.69
    },
    {
      "epoch": 1.8445327228656074,
      "grad_norm": 0.7686677637617825,
      "learning_rate": 1.484815700945974e-07,
      "loss": 0.29986402988433836,
      "num_input_tokens_seen": 873234432,
      "step": 27395,
      "train_runtime": 71736.9216,
      "train_tokens_per_second": 12172.734
    },
    {
      "epoch": 1.8448693778615675,
      "grad_norm": 0.8054071096197858,
      "learning_rate": 1.478426715303849e-07,
      "loss": 0.2911808013916016,
      "num_input_tokens_seen": 873388616,
      "step": 27400,
      "train_runtime": 71750.7463,
      "train_tokens_per_second": 12172.537
    },
    {
      "epoch": 1.8452060328575275,
      "grad_norm": 0.7755531074048934,
      "learning_rate": 1.4720512985649694e-07,
      "loss": 0.28577733039855957,
      "num_input_tokens_seen": 873548624,
      "step": 27405,
      "train_runtime": 71764.6433,
      "train_tokens_per_second": 12172.409
    },
    {
      "epoch": 1.8455426878534877,
      "grad_norm": 0.9375542783148483,
      "learning_rate": 1.4656894525122034e-07,
      "loss": 0.295374321937561,
      "num_input_tokens_seen": 873706784,
      "step": 27410,
      "train_runtime": 71777.4695,
      "train_tokens_per_second": 12172.438
    },
    {
      "epoch": 1.845879342849448,
      "grad_norm": 0.8289218066428363,
      "learning_rate": 1.459341178924617e-07,
      "loss": 0.2863925933837891,
      "num_input_tokens_seen": 873866424,
      "step": 27415,
      "train_runtime": 71790.1119,
      "train_tokens_per_second": 12172.518
    },
    {
      "epoch": 1.846215997845408,
      "grad_norm": 0.8324948723938166,
      "learning_rate": 1.453006479577479e-07,
      "loss": 0.2693448543548584,
      "num_input_tokens_seen": 874019312,
      "step": 27420,
      "train_runtime": 71802.5536,
      "train_tokens_per_second": 12172.538
    },
    {
      "epoch": 1.846552652841368,
      "grad_norm": 0.8032478503855339,
      "learning_rate": 1.4466853562422778e-07,
      "loss": 0.30429399013519287,
      "num_input_tokens_seen": 874168136,
      "step": 27425,
      "train_runtime": 71815.4404,
      "train_tokens_per_second": 12172.426
    },
    {
      "epoch": 1.8468893078373283,
      "grad_norm": 0.8397375923041422,
      "learning_rate": 1.440377810686694e-07,
      "loss": 0.27332324981689454,
      "num_input_tokens_seen": 874327848,
      "step": 27430,
      "train_runtime": 71829.4099,
      "train_tokens_per_second": 12172.282
    },
    {
      "epoch": 1.8472259628332885,
      "grad_norm": 0.9140841420402113,
      "learning_rate": 1.4340838446746052e-07,
      "loss": 0.293894624710083,
      "num_input_tokens_seen": 874486616,
      "step": 27435,
      "train_runtime": 71842.1057,
      "train_tokens_per_second": 12172.341
    },
    {
      "epoch": 1.8475626178292486,
      "grad_norm": 0.8225732553987689,
      "learning_rate": 1.4278034599661094e-07,
      "loss": 0.2723236083984375,
      "num_input_tokens_seen": 874645128,
      "step": 27440,
      "train_runtime": 71855.2614,
      "train_tokens_per_second": 12172.319
    },
    {
      "epoch": 1.8478992728252086,
      "grad_norm": 1.056085285217674,
      "learning_rate": 1.421536658317485e-07,
      "loss": 0.27869529724121095,
      "num_input_tokens_seen": 874797904,
      "step": 27445,
      "train_runtime": 71868.6581,
      "train_tokens_per_second": 12172.175
    },
    {
      "epoch": 1.8482359278211689,
      "grad_norm": 0.8281075836119839,
      "learning_rate": 1.41528344148123e-07,
      "loss": 0.2799793004989624,
      "num_input_tokens_seen": 874961608,
      "step": 27450,
      "train_runtime": 71881.7164,
      "train_tokens_per_second": 12172.241
    },
    {
      "epoch": 1.848572582817129,
      "grad_norm": 0.9206121065506052,
      "learning_rate": 1.40904381120604e-07,
      "loss": 0.25560123920440675,
      "num_input_tokens_seen": 875118712,
      "step": 27455,
      "train_runtime": 71894.7985,
      "train_tokens_per_second": 12172.212
    },
    {
      "epoch": 1.8489092378130891,
      "grad_norm": 0.8389948818156299,
      "learning_rate": 1.4028177692367973e-07,
      "loss": 0.29099810123443604,
      "num_input_tokens_seen": 875277384,
      "step": 27460,
      "train_runtime": 71907.6928,
      "train_tokens_per_second": 12172.236
    },
    {
      "epoch": 1.8492458928090492,
      "grad_norm": 0.7940698844836693,
      "learning_rate": 1.3966053173146032e-07,
      "loss": 0.29418370723724363,
      "num_input_tokens_seen": 875438088,
      "step": 27465,
      "train_runtime": 71920.6307,
      "train_tokens_per_second": 12172.28
    },
    {
      "epoch": 1.8495825478050094,
      "grad_norm": 0.7213998746705305,
      "learning_rate": 1.3904064571767462e-07,
      "loss": 0.2881680727005005,
      "num_input_tokens_seen": 875601896,
      "step": 27470,
      "train_runtime": 71933.7346,
      "train_tokens_per_second": 12172.34
    },
    {
      "epoch": 1.8499192028009697,
      "grad_norm": 0.7783994836679651,
      "learning_rate": 1.3842211905567227e-07,
      "loss": 0.27355542182922366,
      "num_input_tokens_seen": 875759312,
      "step": 27475,
      "train_runtime": 71946.8358,
      "train_tokens_per_second": 12172.312
    },
    {
      "epoch": 1.8502558577969297,
      "grad_norm": 0.8238684200032111,
      "learning_rate": 1.378049519184216e-07,
      "loss": 0.29291419982910155,
      "num_input_tokens_seen": 875917384,
      "step": 27480,
      "train_runtime": 71960.0374,
      "train_tokens_per_second": 12172.275
    },
    {
      "epoch": 1.8505925127928897,
      "grad_norm": 0.7760330888477129,
      "learning_rate": 1.3718914447851116e-07,
      "loss": 0.27587141990661623,
      "num_input_tokens_seen": 876080352,
      "step": 27485,
      "train_runtime": 71972.8502,
      "train_tokens_per_second": 12172.373
    },
    {
      "epoch": 1.85092916778885,
      "grad_norm": 0.7645014310661358,
      "learning_rate": 1.3657469690815106e-07,
      "loss": 0.28146064281463623,
      "num_input_tokens_seen": 876244040,
      "step": 27490,
      "train_runtime": 71985.9259,
      "train_tokens_per_second": 12172.436
    },
    {
      "epoch": 1.8512658227848102,
      "grad_norm": 0.7520126415959406,
      "learning_rate": 1.359616093791677e-07,
      "loss": 0.3067142486572266,
      "num_input_tokens_seen": 876406488,
      "step": 27495,
      "train_runtime": 71998.5805,
      "train_tokens_per_second": 12172.552
    },
    {
      "epoch": 1.8516024777807703,
      "grad_norm": 0.7197164296566871,
      "learning_rate": 1.353498820630106e-07,
      "loss": 0.28169894218444824,
      "num_input_tokens_seen": 876568728,
      "step": 27500,
      "train_runtime": 72011.7924,
      "train_tokens_per_second": 12172.572
    },
    {
      "epoch": 1.8519391327767303,
      "grad_norm": 0.8231057014036376,
      "learning_rate": 1.3473951513074625e-07,
      "loss": 0.2980319023132324,
      "num_input_tokens_seen": 876728376,
      "step": 27505,
      "train_runtime": 72024.4911,
      "train_tokens_per_second": 12172.642
    },
    {
      "epoch": 1.8522757877726905,
      "grad_norm": 0.8179333184858466,
      "learning_rate": 1.341305087530631e-07,
      "loss": 0.2921454906463623,
      "num_input_tokens_seen": 876886784,
      "step": 27510,
      "train_runtime": 72037.5784,
      "train_tokens_per_second": 12172.63
    },
    {
      "epoch": 1.8526124427686508,
      "grad_norm": 0.9488685337931034,
      "learning_rate": 1.3352286310026663e-07,
      "loss": 0.30609047412872314,
      "num_input_tokens_seen": 877046024,
      "step": 27515,
      "train_runtime": 72050.8642,
      "train_tokens_per_second": 12172.595
    },
    {
      "epoch": 1.8529490977646108,
      "grad_norm": 0.7935453432337211,
      "learning_rate": 1.329165783422831e-07,
      "loss": 0.31588268280029297,
      "num_input_tokens_seen": 877205224,
      "step": 27520,
      "train_runtime": 72064.5287,
      "train_tokens_per_second": 12172.497
    },
    {
      "epoch": 1.8532857527605708,
      "grad_norm": 0.8468160317694157,
      "learning_rate": 1.3231165464865914e-07,
      "loss": 0.2817737579345703,
      "num_input_tokens_seen": 877360056,
      "step": 27525,
      "train_runtime": 72077.2143,
      "train_tokens_per_second": 12172.502
    },
    {
      "epoch": 1.853622407756531,
      "grad_norm": 0.7974732968250614,
      "learning_rate": 1.3170809218855884e-07,
      "loss": 0.2877452850341797,
      "num_input_tokens_seen": 877522680,
      "step": 27530,
      "train_runtime": 72090.5771,
      "train_tokens_per_second": 12172.502
    },
    {
      "epoch": 1.8539590627524913,
      "grad_norm": 0.8331855129202257,
      "learning_rate": 1.3110589113076554e-07,
      "loss": 0.29010734558105467,
      "num_input_tokens_seen": 877686056,
      "step": 27535,
      "train_runtime": 72103.5532,
      "train_tokens_per_second": 12172.577
    },
    {
      "epoch": 1.8542957177484514,
      "grad_norm": 0.8540674606929154,
      "learning_rate": 1.3050505164368566e-07,
      "loss": 0.2872800588607788,
      "num_input_tokens_seen": 877844984,
      "step": 27540,
      "train_runtime": 72116.479,
      "train_tokens_per_second": 12172.599
    },
    {
      "epoch": 1.8546323727444114,
      "grad_norm": 0.8976389332919095,
      "learning_rate": 1.299055738953392e-07,
      "loss": 0.26750071048736573,
      "num_input_tokens_seen": 877999608,
      "step": 27545,
      "train_runtime": 72129.9104,
      "train_tokens_per_second": 12172.476
    },
    {
      "epoch": 1.8549690277403716,
      "grad_norm": 0.775081261180271,
      "learning_rate": 1.293074580533704e-07,
      "loss": 0.28496038913726807,
      "num_input_tokens_seen": 878162664,
      "step": 27550,
      "train_runtime": 72142.9858,
      "train_tokens_per_second": 12172.53
    },
    {
      "epoch": 1.855305682736332,
      "grad_norm": 0.8260434916689043,
      "learning_rate": 1.2871070428503885e-07,
      "loss": 0.30309720039367677,
      "num_input_tokens_seen": 878318168,
      "step": 27555,
      "train_runtime": 72156.2119,
      "train_tokens_per_second": 12172.454
    },
    {
      "epoch": 1.855642337732292,
      "grad_norm": 0.7614330294648849,
      "learning_rate": 1.2811531275722543e-07,
      "loss": 0.31922097206115724,
      "num_input_tokens_seen": 878477896,
      "step": 27560,
      "train_runtime": 72169.2854,
      "train_tokens_per_second": 12172.462
    },
    {
      "epoch": 1.855978992728252,
      "grad_norm": 0.8398075272620146,
      "learning_rate": 1.275212836364298e-07,
      "loss": 0.2937641143798828,
      "num_input_tokens_seen": 878640016,
      "step": 27565,
      "train_runtime": 72182.9039,
      "train_tokens_per_second": 12172.412
    },
    {
      "epoch": 1.8563156477242122,
      "grad_norm": 0.807061108497259,
      "learning_rate": 1.2692861708877014e-07,
      "loss": 0.2721449136734009,
      "num_input_tokens_seen": 878797080,
      "step": 27570,
      "train_runtime": 72195.3304,
      "train_tokens_per_second": 12172.492
    },
    {
      "epoch": 1.8566523027201725,
      "grad_norm": 0.7607230875107283,
      "learning_rate": 1.2633731327998332e-07,
      "loss": 0.28148822784423827,
      "num_input_tokens_seen": 878956872,
      "step": 27575,
      "train_runtime": 72209.6594,
      "train_tokens_per_second": 12172.289
    },
    {
      "epoch": 1.8569889577161325,
      "grad_norm": 0.8350663103725182,
      "learning_rate": 1.25747372375426e-07,
      "loss": 0.29016618728637694,
      "num_input_tokens_seen": 879110848,
      "step": 27580,
      "train_runtime": 72222.951,
      "train_tokens_per_second": 12172.181
    },
    {
      "epoch": 1.8573256127120925,
      "grad_norm": 0.8066071274430157,
      "learning_rate": 1.251587945400723e-07,
      "loss": 0.3028055906295776,
      "num_input_tokens_seen": 879271728,
      "step": 27585,
      "train_runtime": 72236.6227,
      "train_tokens_per_second": 12172.105
    },
    {
      "epoch": 1.8576622677080528,
      "grad_norm": 0.8104939244271857,
      "learning_rate": 1.2457157993851832e-07,
      "loss": 0.29297661781311035,
      "num_input_tokens_seen": 879432752,
      "step": 27590,
      "train_runtime": 72249.7598,
      "train_tokens_per_second": 12172.12
    },
    {
      "epoch": 1.857998922704013,
      "grad_norm": 0.8068816762151182,
      "learning_rate": 1.2398572873497438e-07,
      "loss": 0.28179564476013186,
      "num_input_tokens_seen": 879596584,
      "step": 27595,
      "train_runtime": 72262.8702,
      "train_tokens_per_second": 12172.179
    },
    {
      "epoch": 1.858335577699973,
      "grad_norm": 0.8049826965625225,
      "learning_rate": 1.2340124109327334e-07,
      "loss": 0.294586706161499,
      "num_input_tokens_seen": 879757472,
      "step": 27600,
      "train_runtime": 72275.8495,
      "train_tokens_per_second": 12172.219
    },
    {
      "epoch": 1.858672232695933,
      "grad_norm": 0.7914758802484532,
      "learning_rate": 1.22818117176865e-07,
      "loss": 0.3095369338989258,
      "num_input_tokens_seen": 879918976,
      "step": 27605,
      "train_runtime": 72288.3424,
      "train_tokens_per_second": 12172.35
    },
    {
      "epoch": 1.8590088876918933,
      "grad_norm": 0.8477082099647367,
      "learning_rate": 1.2223635714881777e-07,
      "loss": 0.29800169467926024,
      "num_input_tokens_seen": 880075896,
      "step": 27610,
      "train_runtime": 72301.1057,
      "train_tokens_per_second": 12172.371
    },
    {
      "epoch": 1.8593455426878536,
      "grad_norm": 0.757961306211108,
      "learning_rate": 1.2165596117181931e-07,
      "loss": 0.286836051940918,
      "num_input_tokens_seen": 880236864,
      "step": 27615,
      "train_runtime": 72313.9924,
      "train_tokens_per_second": 12172.428
    },
    {
      "epoch": 1.8596821976838136,
      "grad_norm": 0.7219058696615754,
      "learning_rate": 1.2107692940817594e-07,
      "loss": 0.3009026527404785,
      "num_input_tokens_seen": 880394536,
      "step": 27620,
      "train_runtime": 72326.5347,
      "train_tokens_per_second": 12172.497
    },
    {
      "epoch": 1.8600188526797736,
      "grad_norm": 0.8855547725045079,
      "learning_rate": 1.2049926201981088e-07,
      "loss": 0.29579958915710447,
      "num_input_tokens_seen": 880551648,
      "step": 27625,
      "train_runtime": 72339.6324,
      "train_tokens_per_second": 12172.465
    },
    {
      "epoch": 1.8603555076757339,
      "grad_norm": 0.770433702305638,
      "learning_rate": 1.199229591682677e-07,
      "loss": 0.3103304862976074,
      "num_input_tokens_seen": 880710512,
      "step": 27630,
      "train_runtime": 72352.3334,
      "train_tokens_per_second": 12172.524
    },
    {
      "epoch": 1.8606921626716941,
      "grad_norm": 0.9754527664560025,
      "learning_rate": 1.193480210147069e-07,
      "loss": 0.30147709846496584,
      "num_input_tokens_seen": 880873280,
      "step": 27635,
      "train_runtime": 72365.063,
      "train_tokens_per_second": 12172.632
    },
    {
      "epoch": 1.8610288176676542,
      "grad_norm": 0.7885895776645526,
      "learning_rate": 1.1877444771991042e-07,
      "loss": 0.28636789321899414,
      "num_input_tokens_seen": 881035728,
      "step": 27640,
      "train_runtime": 72377.8927,
      "train_tokens_per_second": 12172.719
    },
    {
      "epoch": 1.8613654726636142,
      "grad_norm": 0.7786126360153375,
      "learning_rate": 1.1820223944427279e-07,
      "loss": 0.27569682598114015,
      "num_input_tokens_seen": 881191824,
      "step": 27645,
      "train_runtime": 72390.8343,
      "train_tokens_per_second": 12172.699
    },
    {
      "epoch": 1.8617021276595744,
      "grad_norm": 0.7199390476857687,
      "learning_rate": 1.1763139634781262e-07,
      "loss": 0.29572110176086425,
      "num_input_tokens_seen": 881353488,
      "step": 27650,
      "train_runtime": 72404.0157,
      "train_tokens_per_second": 12172.716
    },
    {
      "epoch": 1.8620387826555347,
      "grad_norm": 0.781387620909201,
      "learning_rate": 1.1706191859016392e-07,
      "loss": 0.3168900489807129,
      "num_input_tokens_seen": 881513960,
      "step": 27655,
      "train_runtime": 72417.0932,
      "train_tokens_per_second": 12172.733
    },
    {
      "epoch": 1.8623754376514947,
      "grad_norm": 0.8617449163816636,
      "learning_rate": 1.1649380633057927e-07,
      "loss": 0.2769595146179199,
      "num_input_tokens_seen": 881668912,
      "step": 27660,
      "train_runtime": 72429.9984,
      "train_tokens_per_second": 12172.704
    },
    {
      "epoch": 1.8627120926474547,
      "grad_norm": 0.7748137491840571,
      "learning_rate": 1.1592705972792939e-07,
      "loss": 0.27399134635925293,
      "num_input_tokens_seen": 881828072,
      "step": 27665,
      "train_runtime": 72443.7004,
      "train_tokens_per_second": 12172.598
    },
    {
      "epoch": 1.863048747643415,
      "grad_norm": 0.7582729403371697,
      "learning_rate": 1.1536167894070249e-07,
      "loss": 0.30470900535583495,
      "num_input_tokens_seen": 881989632,
      "step": 27670,
      "train_runtime": 72457.2639,
      "train_tokens_per_second": 12172.55
    },
    {
      "epoch": 1.8633854026393752,
      "grad_norm": 0.7719897390691288,
      "learning_rate": 1.1479766412700655e-07,
      "loss": 0.3218668460845947,
      "num_input_tokens_seen": 882153424,
      "step": 27675,
      "train_runtime": 72470.3733,
      "train_tokens_per_second": 12172.608
    },
    {
      "epoch": 1.8637220576353353,
      "grad_norm": 0.7696013560298175,
      "learning_rate": 1.1423501544456595e-07,
      "loss": 0.30865747928619386,
      "num_input_tokens_seen": 882309208,
      "step": 27680,
      "train_runtime": 72482.8419,
      "train_tokens_per_second": 12172.663
    },
    {
      "epoch": 1.8640587126312953,
      "grad_norm": 0.7970654453005196,
      "learning_rate": 1.1367373305072315e-07,
      "loss": 0.27881536483764646,
      "num_input_tokens_seen": 882471712,
      "step": 27685,
      "train_runtime": 72496.1584,
      "train_tokens_per_second": 12172.669
    },
    {
      "epoch": 1.8643953676272556,
      "grad_norm": 0.7205948537495926,
      "learning_rate": 1.1311381710244096e-07,
      "loss": 0.2617627143859863,
      "num_input_tokens_seen": 882633976,
      "step": 27690,
      "train_runtime": 72508.7631,
      "train_tokens_per_second": 12172.79
    },
    {
      "epoch": 1.8647320226232158,
      "grad_norm": 1.0981707922908153,
      "learning_rate": 1.1255526775629521e-07,
      "loss": 0.26013147830963135,
      "num_input_tokens_seen": 882795784,
      "step": 27695,
      "train_runtime": 72522.1962,
      "train_tokens_per_second": 12172.767
    },
    {
      "epoch": 1.8650686776191758,
      "grad_norm": 0.8308397344915243,
      "learning_rate": 1.1199808516848487e-07,
      "loss": 0.30060672760009766,
      "num_input_tokens_seen": 882951624,
      "step": 27700,
      "train_runtime": 72535.1119,
      "train_tokens_per_second": 12172.748
    },
    {
      "epoch": 1.8654053326151359,
      "grad_norm": 0.7614020283838211,
      "learning_rate": 1.1144226949482306e-07,
      "loss": 0.2741485834121704,
      "num_input_tokens_seen": 883109504,
      "step": 27705,
      "train_runtime": 72548.7784,
      "train_tokens_per_second": 12172.631
    },
    {
      "epoch": 1.865741987611096,
      "grad_norm": 0.7267544168293171,
      "learning_rate": 1.1088782089074157e-07,
      "loss": 0.2698942184448242,
      "num_input_tokens_seen": 883266656,
      "step": 27710,
      "train_runtime": 72561.638,
      "train_tokens_per_second": 12172.639
    },
    {
      "epoch": 1.8660786426070564,
      "grad_norm": 0.8558606155696188,
      "learning_rate": 1.1033473951129137e-07,
      "loss": 0.29747495651245115,
      "num_input_tokens_seen": 883426048,
      "step": 27715,
      "train_runtime": 72574.5333,
      "train_tokens_per_second": 12172.673
    },
    {
      "epoch": 1.8664152976030164,
      "grad_norm": 0.822951280738268,
      "learning_rate": 1.0978302551113929e-07,
      "loss": 0.30676970481872556,
      "num_input_tokens_seen": 883588984,
      "step": 27720,
      "train_runtime": 72587.2856,
      "train_tokens_per_second": 12172.779
    },
    {
      "epoch": 1.8667519525989764,
      "grad_norm": 0.8538920617511101,
      "learning_rate": 1.0923267904457025e-07,
      "loss": 0.3054649353027344,
      "num_input_tokens_seen": 883750376,
      "step": 27725,
      "train_runtime": 72600.4562,
      "train_tokens_per_second": 12172.794
    },
    {
      "epoch": 1.8670886075949367,
      "grad_norm": 0.7917885023548918,
      "learning_rate": 1.086837002654867e-07,
      "loss": 0.2991494655609131,
      "num_input_tokens_seen": 883907056,
      "step": 27730,
      "train_runtime": 72613.288,
      "train_tokens_per_second": 12172.8
    },
    {
      "epoch": 1.867425262590897,
      "grad_norm": 0.8429094191556364,
      "learning_rate": 1.0813608932740915e-07,
      "loss": 0.2981977939605713,
      "num_input_tokens_seen": 884070896,
      "step": 27735,
      "train_runtime": 72625.7173,
      "train_tokens_per_second": 12172.973
    },
    {
      "epoch": 1.867761917586857,
      "grad_norm": 0.8632924956538994,
      "learning_rate": 1.0758984638347569e-07,
      "loss": 0.29195926189422605,
      "num_input_tokens_seen": 884231936,
      "step": 27740,
      "train_runtime": 72638.1958,
      "train_tokens_per_second": 12173.099
    },
    {
      "epoch": 1.868098572582817,
      "grad_norm": 0.9102952049022162,
      "learning_rate": 1.0704497158644022e-07,
      "loss": 0.32872300148010253,
      "num_input_tokens_seen": 884386520,
      "step": 27745,
      "train_runtime": 72650.6623,
      "train_tokens_per_second": 12173.138
    },
    {
      "epoch": 1.8684352275787772,
      "grad_norm": 0.7497864074848651,
      "learning_rate": 1.0650146508867642e-07,
      "loss": 0.28098583221435547,
      "num_input_tokens_seen": 884544528,
      "step": 27750,
      "train_runtime": 72663.752,
      "train_tokens_per_second": 12173.119
    },
    {
      "epoch": 1.8687718825747375,
      "grad_norm": 0.8301490276644983,
      "learning_rate": 1.0595932704217327e-07,
      "loss": 0.3004870891571045,
      "num_input_tokens_seen": 884706520,
      "step": 27755,
      "train_runtime": 72676.898,
      "train_tokens_per_second": 12173.146
    },
    {
      "epoch": 1.8691085375706975,
      "grad_norm": 0.7934361981303704,
      "learning_rate": 1.0541855759853837e-07,
      "loss": 0.28049535751342775,
      "num_input_tokens_seen": 884870280,
      "step": 27760,
      "train_runtime": 72689.3069,
      "train_tokens_per_second": 12173.321
    },
    {
      "epoch": 1.8694451925666575,
      "grad_norm": 0.8005042190087888,
      "learning_rate": 1.0487915690899685e-07,
      "loss": 0.3012115240097046,
      "num_input_tokens_seen": 885032152,
      "step": 27765,
      "train_runtime": 72702.4477,
      "train_tokens_per_second": 12173.347
    },
    {
      "epoch": 1.8697818475626178,
      "grad_norm": 0.7385473938643486,
      "learning_rate": 1.0434112512438921e-07,
      "loss": 0.2842231750488281,
      "num_input_tokens_seen": 885192552,
      "step": 27770,
      "train_runtime": 72715.9269,
      "train_tokens_per_second": 12173.297
    },
    {
      "epoch": 1.870118502558578,
      "grad_norm": 0.8847364400413646,
      "learning_rate": 1.0380446239517506e-07,
      "loss": 0.272721529006958,
      "num_input_tokens_seen": 885345888,
      "step": 27775,
      "train_runtime": 72729.2223,
      "train_tokens_per_second": 12173.18
    },
    {
      "epoch": 1.870455157554538,
      "grad_norm": 0.8821201818777644,
      "learning_rate": 1.032691688714299e-07,
      "loss": 0.28608098030090334,
      "num_input_tokens_seen": 885495360,
      "step": 27780,
      "train_runtime": 72742.1418,
      "train_tokens_per_second": 12173.072
    },
    {
      "epoch": 1.870791812550498,
      "grad_norm": 0.8705735683370164,
      "learning_rate": 1.0273524470284734e-07,
      "loss": 0.3107900142669678,
      "num_input_tokens_seen": 885658848,
      "step": 27785,
      "train_runtime": 72755.1377,
      "train_tokens_per_second": 12173.145
    },
    {
      "epoch": 1.8711284675464583,
      "grad_norm": 0.9010580375654128,
      "learning_rate": 1.0220269003873739e-07,
      "loss": 0.28125360012054446,
      "num_input_tokens_seen": 885815280,
      "step": 27790,
      "train_runtime": 72768.2008,
      "train_tokens_per_second": 12173.11
    },
    {
      "epoch": 1.8714651225424186,
      "grad_norm": 0.8152127847206131,
      "learning_rate": 1.0167150502802703e-07,
      "loss": 0.2652273416519165,
      "num_input_tokens_seen": 885973968,
      "step": 27795,
      "train_runtime": 72781.8532,
      "train_tokens_per_second": 12173.006
    },
    {
      "epoch": 1.8718017775383786,
      "grad_norm": 0.7832099321616367,
      "learning_rate": 1.0114168981926075e-07,
      "loss": 0.30316591262817383,
      "num_input_tokens_seen": 886136128,
      "step": 27800,
      "train_runtime": 72795.6731,
      "train_tokens_per_second": 12172.923
    },
    {
      "epoch": 1.8721384325343386,
      "grad_norm": 0.7796280147400254,
      "learning_rate": 1.006132445605995e-07,
      "loss": 0.2941952466964722,
      "num_input_tokens_seen": 886287960,
      "step": 27805,
      "train_runtime": 72808.8443,
      "train_tokens_per_second": 12172.806
    },
    {
      "epoch": 1.872475087530299,
      "grad_norm": 0.7871098871950551,
      "learning_rate": 1.0008616939982118e-07,
      "loss": 0.28180084228515623,
      "num_input_tokens_seen": 886445032,
      "step": 27810,
      "train_runtime": 72822.2442,
      "train_tokens_per_second": 12172.723
    },
    {
      "epoch": 1.8728117425262591,
      "grad_norm": 0.8251149939105866,
      "learning_rate": 9.956046448432177e-08,
      "loss": 0.2820472478866577,
      "num_input_tokens_seen": 886604480,
      "step": 27815,
      "train_runtime": 72836.1057,
      "train_tokens_per_second": 12172.596
    },
    {
      "epoch": 1.8731483975222192,
      "grad_norm": 0.750269688245949,
      "learning_rate": 9.90361299611109e-08,
      "loss": 0.30066475868225095,
      "num_input_tokens_seen": 886766152,
      "step": 27820,
      "train_runtime": 72849.1102,
      "train_tokens_per_second": 12172.642
    },
    {
      "epoch": 1.8734850525181792,
      "grad_norm": 0.877487740789378,
      "learning_rate": 9.851316597681959e-08,
      "loss": 0.30822277069091797,
      "num_input_tokens_seen": 886925728,
      "step": 27825,
      "train_runtime": 72862.5207,
      "train_tokens_per_second": 12172.592
    },
    {
      "epoch": 1.8738217075141395,
      "grad_norm": 0.8351287076508815,
      "learning_rate": 9.799157267769088e-08,
      "loss": 0.30189104080200196,
      "num_input_tokens_seen": 887083760,
      "step": 27830,
      "train_runtime": 72875.3919,
      "train_tokens_per_second": 12172.61
    },
    {
      "epoch": 1.8741583625100997,
      "grad_norm": 0.803474666309235,
      "learning_rate": 9.747135020958753e-08,
      "loss": 0.29504618644714353,
      "num_input_tokens_seen": 887245288,
      "step": 27835,
      "train_runtime": 72888.314,
      "train_tokens_per_second": 12172.669
    },
    {
      "epoch": 1.8744950175060597,
      "grad_norm": 0.7797560994529075,
      "learning_rate": 9.695249871798818e-08,
      "loss": 0.2727928638458252,
      "num_input_tokens_seen": 887406264,
      "step": 27840,
      "train_runtime": 72902.1108,
      "train_tokens_per_second": 12172.573
    },
    {
      "epoch": 1.8748316725020198,
      "grad_norm": 0.8622211542518032,
      "learning_rate": 9.643501834798786e-08,
      "loss": 0.2960002899169922,
      "num_input_tokens_seen": 887567200,
      "step": 27845,
      "train_runtime": 72915.9309,
      "train_tokens_per_second": 12172.473
    },
    {
      "epoch": 1.8751683274979802,
      "grad_norm": 0.8963240547969548,
      "learning_rate": 9.591890924429858e-08,
      "loss": 0.3056071996688843,
      "num_input_tokens_seen": 887731040,
      "step": 27850,
      "train_runtime": 72928.3555,
      "train_tokens_per_second": 12172.646
    },
    {
      "epoch": 1.8755049824939403,
      "grad_norm": 0.780586472801103,
      "learning_rate": 9.540417155124826e-08,
      "loss": 0.2770434617996216,
      "num_input_tokens_seen": 887886440,
      "step": 27855,
      "train_runtime": 72942.3981,
      "train_tokens_per_second": 12172.433
    },
    {
      "epoch": 1.8758416374899003,
      "grad_norm": 0.7850527896914666,
      "learning_rate": 9.489080541278117e-08,
      "loss": 0.29499149322509766,
      "num_input_tokens_seen": 888044232,
      "step": 27860,
      "train_runtime": 72955.1403,
      "train_tokens_per_second": 12172.47
    },
    {
      "epoch": 1.8761782924858605,
      "grad_norm": 0.7553449602378549,
      "learning_rate": 9.437881097246026e-08,
      "loss": 0.2947093486785889,
      "num_input_tokens_seen": 888207952,
      "step": 27865,
      "train_runtime": 72968.2097,
      "train_tokens_per_second": 12172.533
    },
    {
      "epoch": 1.8765149474818208,
      "grad_norm": 0.786645120767396,
      "learning_rate": 9.3868188373461e-08,
      "loss": 0.29761443138122556,
      "num_input_tokens_seen": 888366648,
      "step": 27870,
      "train_runtime": 72982.0319,
      "train_tokens_per_second": 12172.402
    },
    {
      "epoch": 1.8768516024777808,
      "grad_norm": 0.898577498960538,
      "learning_rate": 9.335893775857918e-08,
      "loss": 0.30380902290344236,
      "num_input_tokens_seen": 888524440,
      "step": 27875,
      "train_runtime": 72996.075,
      "train_tokens_per_second": 12172.222
    },
    {
      "epoch": 1.8771882574737409,
      "grad_norm": 1.1507064642213123,
      "learning_rate": 9.285105927022309e-08,
      "loss": 0.2756707191467285,
      "num_input_tokens_seen": 888683984,
      "step": 27880,
      "train_runtime": 73008.9749,
      "train_tokens_per_second": 12172.257
    },
    {
      "epoch": 1.877524912469701,
      "grad_norm": 0.8484865408624757,
      "learning_rate": 9.234455305042133e-08,
      "loss": 0.28602607250213624,
      "num_input_tokens_seen": 888837904,
      "step": 27885,
      "train_runtime": 73021.9953,
      "train_tokens_per_second": 12172.194
    },
    {
      "epoch": 1.8778615674656614,
      "grad_norm": 0.8267845931243424,
      "learning_rate": 9.183941924081563e-08,
      "loss": 0.2835225105285645,
      "num_input_tokens_seen": 888996872,
      "step": 27890,
      "train_runtime": 73034.9354,
      "train_tokens_per_second": 12172.214
    },
    {
      "epoch": 1.8781982224616214,
      "grad_norm": 0.7959633390473106,
      "learning_rate": 9.133565798266574e-08,
      "loss": 0.2821208477020264,
      "num_input_tokens_seen": 889155368,
      "step": 27895,
      "train_runtime": 73047.6879,
      "train_tokens_per_second": 12172.259
    },
    {
      "epoch": 1.8785348774575814,
      "grad_norm": 0.882191701119769,
      "learning_rate": 9.083326941684622e-08,
      "loss": 0.3249369621276855,
      "num_input_tokens_seen": 889314256,
      "step": 27900,
      "train_runtime": 73060.8167,
      "train_tokens_per_second": 12172.246
    },
    {
      "epoch": 1.8788715324535417,
      "grad_norm": 0.7827373944313485,
      "learning_rate": 9.033225368384912e-08,
      "loss": 0.31013007164001466,
      "num_input_tokens_seen": 889474016,
      "step": 27905,
      "train_runtime": 73074.0017,
      "train_tokens_per_second": 12172.236
    },
    {
      "epoch": 1.879208187449502,
      "grad_norm": 0.7182055697924479,
      "learning_rate": 8.98326109237807e-08,
      "loss": 0.2798023700714111,
      "num_input_tokens_seen": 889633184,
      "step": 27910,
      "train_runtime": 73086.4462,
      "train_tokens_per_second": 12172.342
    },
    {
      "epoch": 1.879544842445462,
      "grad_norm": 0.8014438382240342,
      "learning_rate": 8.933434127636586e-08,
      "loss": 0.3125504732131958,
      "num_input_tokens_seen": 889791848,
      "step": 27915,
      "train_runtime": 73099.2601,
      "train_tokens_per_second": 12172.378
    },
    {
      "epoch": 1.879881497441422,
      "grad_norm": 0.9111176863754624,
      "learning_rate": 8.883744488094314e-08,
      "loss": 0.3022294521331787,
      "num_input_tokens_seen": 889949632,
      "step": 27920,
      "train_runtime": 73112.1967,
      "train_tokens_per_second": 12172.383
    },
    {
      "epoch": 1.8802181524373822,
      "grad_norm": 0.721452676212702,
      "learning_rate": 8.834192187646917e-08,
      "loss": 0.2927600383758545,
      "num_input_tokens_seen": 890113472,
      "step": 27925,
      "train_runtime": 73124.6317,
      "train_tokens_per_second": 12172.553
    },
    {
      "epoch": 1.8805548074333425,
      "grad_norm": 0.8724636394628685,
      "learning_rate": 8.784777240151365e-08,
      "loss": 0.311005163192749,
      "num_input_tokens_seen": 890272752,
      "step": 27930,
      "train_runtime": 73137.8408,
      "train_tokens_per_second": 12172.533
    },
    {
      "epoch": 1.8808914624293025,
      "grad_norm": 0.7803253748907685,
      "learning_rate": 8.73549965942655e-08,
      "loss": 0.2907422542572021,
      "num_input_tokens_seen": 890433480,
      "step": 27935,
      "train_runtime": 73150.5806,
      "train_tokens_per_second": 12172.61
    },
    {
      "epoch": 1.8812281174252625,
      "grad_norm": 0.6945714830876641,
      "learning_rate": 8.686359459252725e-08,
      "loss": 0.28446648120880125,
      "num_input_tokens_seen": 890589720,
      "step": 27940,
      "train_runtime": 73163.3329,
      "train_tokens_per_second": 12172.624
    },
    {
      "epoch": 1.8815647724212228,
      "grad_norm": 0.7098115133308185,
      "learning_rate": 8.637356653371787e-08,
      "loss": 0.2761401176452637,
      "num_input_tokens_seen": 890750944,
      "step": 27945,
      "train_runtime": 73176.6691,
      "train_tokens_per_second": 12172.608
    },
    {
      "epoch": 1.881901427417183,
      "grad_norm": 0.8796978111265379,
      "learning_rate": 8.588491255487163e-08,
      "loss": 0.26764066219329835,
      "num_input_tokens_seen": 890911096,
      "step": 27950,
      "train_runtime": 73190.6225,
      "train_tokens_per_second": 12172.476
    },
    {
      "epoch": 1.882238082413143,
      "grad_norm": 0.8406207447190257,
      "learning_rate": 8.539763279264035e-08,
      "loss": 0.27055528163909914,
      "num_input_tokens_seen": 891071288,
      "step": 27955,
      "train_runtime": 73203.1611,
      "train_tokens_per_second": 12172.579
    },
    {
      "epoch": 1.882574737409103,
      "grad_norm": 0.7910239075909938,
      "learning_rate": 8.49117273832889e-08,
      "loss": 0.3221412658691406,
      "num_input_tokens_seen": 891229752,
      "step": 27960,
      "train_runtime": 73216.1942,
      "train_tokens_per_second": 12172.577
    },
    {
      "epoch": 1.8829113924050633,
      "grad_norm": 0.7409938632377756,
      "learning_rate": 8.442719646270082e-08,
      "loss": 0.2672263622283936,
      "num_input_tokens_seen": 891389520,
      "step": 27965,
      "train_runtime": 73228.8565,
      "train_tokens_per_second": 12172.654
    },
    {
      "epoch": 1.8832480474010236,
      "grad_norm": 0.8110239820585948,
      "learning_rate": 8.394404016637159e-08,
      "loss": 0.2805251836776733,
      "num_input_tokens_seen": 891552336,
      "step": 27970,
      "train_runtime": 73242.3189,
      "train_tokens_per_second": 12172.639
    },
    {
      "epoch": 1.8835847023969836,
      "grad_norm": 0.8218219254639022,
      "learning_rate": 8.346225862941592e-08,
      "loss": 0.2921317100524902,
      "num_input_tokens_seen": 891707480,
      "step": 27975,
      "train_runtime": 73255.6706,
      "train_tokens_per_second": 12172.539
    },
    {
      "epoch": 1.8839213573929436,
      "grad_norm": 0.807117688588943,
      "learning_rate": 8.298185198656161e-08,
      "loss": 0.2985464334487915,
      "num_input_tokens_seen": 891868888,
      "step": 27980,
      "train_runtime": 73268.6137,
      "train_tokens_per_second": 12172.591
    },
    {
      "epoch": 1.884258012388904,
      "grad_norm": 0.8038610046617606,
      "learning_rate": 8.25028203721534e-08,
      "loss": 0.2723139762878418,
      "num_input_tokens_seen": 892029672,
      "step": 27985,
      "train_runtime": 73282.313,
      "train_tokens_per_second": 12172.51
    },
    {
      "epoch": 1.8845946673848641,
      "grad_norm": 0.7800281251705247,
      "learning_rate": 8.202516392015081e-08,
      "loss": 0.297794508934021,
      "num_input_tokens_seen": 892188656,
      "step": 27990,
      "train_runtime": 73295.481,
      "train_tokens_per_second": 12172.492
    },
    {
      "epoch": 1.8849313223808242,
      "grad_norm": 0.8718531534244764,
      "learning_rate": 8.154888276412865e-08,
      "loss": 0.2957782745361328,
      "num_input_tokens_seen": 892344528,
      "step": 27995,
      "train_runtime": 73308.8223,
      "train_tokens_per_second": 12172.403
    },
    {
      "epoch": 1.8852679773767842,
      "grad_norm": 0.7533056061439469,
      "learning_rate": 8.10739770372776e-08,
      "loss": 0.26080372333526614,
      "num_input_tokens_seen": 892506264,
      "step": 28000,
      "train_runtime": 73321.8968,
      "train_tokens_per_second": 12172.438
    },
    {
      "epoch": 1.8856046323727444,
      "grad_norm": 0.740664967469983,
      "learning_rate": 8.060044687240365e-08,
      "loss": 0.26940925121307374,
      "num_input_tokens_seen": 892670104,
      "step": 28005,
      "train_runtime": 73334.3161,
      "train_tokens_per_second": 12172.611
    },
    {
      "epoch": 1.8859412873687047,
      "grad_norm": 0.7150405646488428,
      "learning_rate": 8.012829240192754e-08,
      "loss": 0.2987844705581665,
      "num_input_tokens_seen": 892830728,
      "step": 28010,
      "train_runtime": 73347.0152,
      "train_tokens_per_second": 12172.693
    },
    {
      "epoch": 1.8862779423646647,
      "grad_norm": 0.7692045128709198,
      "learning_rate": 7.9657513757887e-08,
      "loss": 0.2818997859954834,
      "num_input_tokens_seen": 892994200,
      "step": 28015,
      "train_runtime": 73360.7068,
      "train_tokens_per_second": 12172.65
    },
    {
      "epoch": 1.8866145973606248,
      "grad_norm": 0.8062565564343142,
      "learning_rate": 7.918811107193225e-08,
      "loss": 0.26958208084106444,
      "num_input_tokens_seen": 893155776,
      "step": 28020,
      "train_runtime": 73373.2189,
      "train_tokens_per_second": 12172.776
    },
    {
      "epoch": 1.886951252356585,
      "grad_norm": 0.8814467491240532,
      "learning_rate": 7.872008447533109e-08,
      "loss": 0.3207878589630127,
      "num_input_tokens_seen": 893317648,
      "step": 28025,
      "train_runtime": 73386.9338,
      "train_tokens_per_second": 12172.707
    },
    {
      "epoch": 1.8872879073525453,
      "grad_norm": 0.7699173404723976,
      "learning_rate": 7.825343409896547e-08,
      "loss": 0.30684852600097656,
      "num_input_tokens_seen": 893476088,
      "step": 28030,
      "train_runtime": 73399.9028,
      "train_tokens_per_second": 12172.715
    },
    {
      "epoch": 1.8876245623485053,
      "grad_norm": 0.8246030398917025,
      "learning_rate": 7.778816007333268e-08,
      "loss": 0.29779620170593263,
      "num_input_tokens_seen": 893634728,
      "step": 28035,
      "train_runtime": 73412.5214,
      "train_tokens_per_second": 12172.783
    },
    {
      "epoch": 1.8879612173444653,
      "grad_norm": 0.8387324903243494,
      "learning_rate": 7.73242625285453e-08,
      "loss": 0.27918837070465086,
      "num_input_tokens_seen": 893791024,
      "step": 28040,
      "train_runtime": 73425.0311,
      "train_tokens_per_second": 12172.838
    },
    {
      "epoch": 1.8882978723404256,
      "grad_norm": 0.758701521605094,
      "learning_rate": 7.686174159433069e-08,
      "loss": 0.3044358491897583,
      "num_input_tokens_seen": 893953768,
      "step": 28045,
      "train_runtime": 73437.765,
      "train_tokens_per_second": 12172.944
    },
    {
      "epoch": 1.8886345273363858,
      "grad_norm": 0.7765670601924785,
      "learning_rate": 7.640059740003091e-08,
      "loss": 0.27041242122650144,
      "num_input_tokens_seen": 894114256,
      "step": 28050,
      "train_runtime": 73450.6094,
      "train_tokens_per_second": 12173.0
    },
    {
      "epoch": 1.8889711823323458,
      "grad_norm": 0.8746524368276339,
      "learning_rate": 7.594083007460396e-08,
      "loss": 0.29973607063293456,
      "num_input_tokens_seen": 894270216,
      "step": 28055,
      "train_runtime": 73464.1717,
      "train_tokens_per_second": 12172.876
    },
    {
      "epoch": 1.8893078373283059,
      "grad_norm": 0.7495572374933687,
      "learning_rate": 7.54824397466225e-08,
      "loss": 0.29380431175231936,
      "num_input_tokens_seen": 894428280,
      "step": 28060,
      "train_runtime": 73476.59,
      "train_tokens_per_second": 12172.969
    },
    {
      "epoch": 1.8896444923242661,
      "grad_norm": 0.7638209256706875,
      "learning_rate": 7.502542654427347e-08,
      "loss": 0.28548665046691896,
      "num_input_tokens_seen": 894585872,
      "step": 28065,
      "train_runtime": 73489.9202,
      "train_tokens_per_second": 12172.906
    },
    {
      "epoch": 1.8899811473202264,
      "grad_norm": 0.7879950306900086,
      "learning_rate": 7.456979059535907e-08,
      "loss": 0.2989804267883301,
      "num_input_tokens_seen": 894745216,
      "step": 28070,
      "train_runtime": 73502.5665,
      "train_tokens_per_second": 12172.979
    },
    {
      "epoch": 1.8903178023161864,
      "grad_norm": 0.8717905943149454,
      "learning_rate": 7.411553202729682e-08,
      "loss": 0.3004459381103516,
      "num_input_tokens_seen": 894893832,
      "step": 28075,
      "train_runtime": 73515.4863,
      "train_tokens_per_second": 12172.861
    },
    {
      "epoch": 1.8906544573121464,
      "grad_norm": 0.7526816965311344,
      "learning_rate": 7.3662650967119e-08,
      "loss": 0.2719897747039795,
      "num_input_tokens_seen": 895050712,
      "step": 28080,
      "train_runtime": 73528.1004,
      "train_tokens_per_second": 12172.907
    },
    {
      "epoch": 1.8909911123081067,
      "grad_norm": 0.7775656149909072,
      "learning_rate": 7.321114754147151e-08,
      "loss": 0.28068835735321046,
      "num_input_tokens_seen": 895210192,
      "step": 28085,
      "train_runtime": 73540.8782,
      "train_tokens_per_second": 12172.96
    },
    {
      "epoch": 1.891327767304067,
      "grad_norm": 0.9814244959295657,
      "learning_rate": 7.276102187661615e-08,
      "loss": 0.3038950443267822,
      "num_input_tokens_seen": 895369464,
      "step": 28090,
      "train_runtime": 73554.0422,
      "train_tokens_per_second": 12172.947
    },
    {
      "epoch": 1.891664422300027,
      "grad_norm": 0.8089655889903498,
      "learning_rate": 7.231227409842944e-08,
      "loss": 0.304179048538208,
      "num_input_tokens_seen": 895532520,
      "step": 28095,
      "train_runtime": 73567.0785,
      "train_tokens_per_second": 12173.006
    },
    {
      "epoch": 1.892001077295987,
      "grad_norm": 0.8613976202198683,
      "learning_rate": 7.186490433240156e-08,
      "loss": 0.2983412504196167,
      "num_input_tokens_seen": 895687472,
      "step": 28100,
      "train_runtime": 73581.0405,
      "train_tokens_per_second": 12172.802
    },
    {
      "epoch": 1.8923377322919472,
      "grad_norm": 0.9332227754416748,
      "learning_rate": 7.14189127036391e-08,
      "loss": 0.2879823684692383,
      "num_input_tokens_seen": 895848280,
      "step": 28105,
      "train_runtime": 73594.2237,
      "train_tokens_per_second": 12172.807
    },
    {
      "epoch": 1.8926743872879075,
      "grad_norm": 0.7709022203139972,
      "learning_rate": 7.097429933686118e-08,
      "loss": 0.29674487113952636,
      "num_input_tokens_seen": 896006864,
      "step": 28110,
      "train_runtime": 73607.3793,
      "train_tokens_per_second": 12172.786
    },
    {
      "epoch": 1.8930110422838675,
      "grad_norm": 0.9516869616323473,
      "learning_rate": 7.053106435640333e-08,
      "loss": 0.3096130847930908,
      "num_input_tokens_seen": 896170704,
      "step": 28115,
      "train_runtime": 73619.8121,
      "train_tokens_per_second": 12172.956
    },
    {
      "epoch": 1.8933476972798275,
      "grad_norm": 0.7877756701839224,
      "learning_rate": 7.008920788621366e-08,
      "loss": 0.2950739860534668,
      "num_input_tokens_seen": 896329424,
      "step": 28120,
      "train_runtime": 73632.7883,
      "train_tokens_per_second": 12172.966
    },
    {
      "epoch": 1.8936843522757878,
      "grad_norm": 0.732935372895786,
      "learning_rate": 6.964873004985717e-08,
      "loss": 0.2967804431915283,
      "num_input_tokens_seen": 896486808,
      "step": 28125,
      "train_runtime": 73645.2384,
      "train_tokens_per_second": 12173.045
    },
    {
      "epoch": 1.894021007271748,
      "grad_norm": 0.8291071660652243,
      "learning_rate": 6.920963097051148e-08,
      "loss": 0.3088953971862793,
      "num_input_tokens_seen": 896649008,
      "step": 28130,
      "train_runtime": 73658.2597,
      "train_tokens_per_second": 12173.095
    },
    {
      "epoch": 1.894357662267708,
      "grad_norm": 0.8194329257380047,
      "learning_rate": 6.877191077096945e-08,
      "loss": 0.27375426292419436,
      "num_input_tokens_seen": 896803584,
      "step": 28135,
      "train_runtime": 73671.8938,
      "train_tokens_per_second": 12172.941
    },
    {
      "epoch": 1.894694317263668,
      "grad_norm": 0.782473271247263,
      "learning_rate": 6.833556957363818e-08,
      "loss": 0.3107022285461426,
      "num_input_tokens_seen": 896964472,
      "step": 28140,
      "train_runtime": 73685.1733,
      "train_tokens_per_second": 12172.93
    },
    {
      "epoch": 1.8950309722596284,
      "grad_norm": 0.8493701549064354,
      "learning_rate": 6.790060750053895e-08,
      "loss": 0.2936178922653198,
      "num_input_tokens_seen": 897120408,
      "step": 28145,
      "train_runtime": 73698.2501,
      "train_tokens_per_second": 12172.886
    },
    {
      "epoch": 1.8953676272555886,
      "grad_norm": 0.7692199282769623,
      "learning_rate": 6.746702467330723e-08,
      "loss": 0.26618127822875975,
      "num_input_tokens_seen": 897283744,
      "step": 28150,
      "train_runtime": 73711.2212,
      "train_tokens_per_second": 12172.96
    },
    {
      "epoch": 1.8957042822515486,
      "grad_norm": 0.7627440936656119,
      "learning_rate": 6.703482121319383e-08,
      "loss": 0.30237407684326173,
      "num_input_tokens_seen": 897445968,
      "step": 28155,
      "train_runtime": 73723.8346,
      "train_tokens_per_second": 12173.078
    },
    {
      "epoch": 1.8960409372475087,
      "grad_norm": 0.8359778328267923,
      "learning_rate": 6.660399724106204e-08,
      "loss": 0.3182742357254028,
      "num_input_tokens_seen": 897608400,
      "step": 28160,
      "train_runtime": 73736.4154,
      "train_tokens_per_second": 12173.204
    },
    {
      "epoch": 1.896377592243469,
      "grad_norm": 0.79818066085529,
      "learning_rate": 6.617455287739161e-08,
      "loss": 0.2847122669219971,
      "num_input_tokens_seen": 897765384,
      "step": 28165,
      "train_runtime": 73748.9111,
      "train_tokens_per_second": 12173.27
    },
    {
      "epoch": 1.8967142472394292,
      "grad_norm": 0.8447766546393227,
      "learning_rate": 6.574648824227369e-08,
      "loss": 0.2903900623321533,
      "num_input_tokens_seen": 897927832,
      "step": 28170,
      "train_runtime": 73762.242,
      "train_tokens_per_second": 12173.272
    },
    {
      "epoch": 1.8970509022353892,
      "grad_norm": 0.7787624516294193,
      "learning_rate": 6.5319803455417e-08,
      "loss": 0.3023900032043457,
      "num_input_tokens_seen": 898088912,
      "step": 28175,
      "train_runtime": 73775.2078,
      "train_tokens_per_second": 12173.316
    },
    {
      "epoch": 1.8973875572313492,
      "grad_norm": 0.8210472837129141,
      "learning_rate": 6.489449863614106e-08,
      "loss": 0.3084021806716919,
      "num_input_tokens_seen": 898247848,
      "step": 28180,
      "train_runtime": 73788.3178,
      "train_tokens_per_second": 12173.307
    },
    {
      "epoch": 1.8977242122273095,
      "grad_norm": 0.8800119946902392,
      "learning_rate": 6.44705739033824e-08,
      "loss": 0.2934131622314453,
      "num_input_tokens_seen": 898405368,
      "step": 28185,
      "train_runtime": 73801.3516,
      "train_tokens_per_second": 12173.292
    },
    {
      "epoch": 1.8980608672232697,
      "grad_norm": 0.9142698337304438,
      "learning_rate": 6.404802937568899e-08,
      "loss": 0.28001105785369873,
      "num_input_tokens_seen": 898559232,
      "step": 28190,
      "train_runtime": 73815.2767,
      "train_tokens_per_second": 12173.08
    },
    {
      "epoch": 1.8983975222192297,
      "grad_norm": 0.6884178027995702,
      "learning_rate": 6.362686517122463e-08,
      "loss": 0.28214905261993406,
      "num_input_tokens_seen": 898719448,
      "step": 28195,
      "train_runtime": 73827.997,
      "train_tokens_per_second": 12173.152
    },
    {
      "epoch": 1.8987341772151898,
      "grad_norm": 0.8385974836461252,
      "learning_rate": 6.320708140776676e-08,
      "loss": 0.2831768035888672,
      "num_input_tokens_seen": 898878736,
      "step": 28200,
      "train_runtime": 73840.4229,
      "train_tokens_per_second": 12173.261
    },
    {
      "epoch": 1.89907083221115,
      "grad_norm": 0.8653610880790381,
      "learning_rate": 6.278867820270596e-08,
      "loss": 0.28925042152404784,
      "num_input_tokens_seen": 899037200,
      "step": 28205,
      "train_runtime": 73854.0457,
      "train_tokens_per_second": 12173.161
    },
    {
      "epoch": 1.8994074872071103,
      "grad_norm": 0.8720381446700393,
      "learning_rate": 6.237165567304749e-08,
      "loss": 0.29513332843780515,
      "num_input_tokens_seen": 899197256,
      "step": 28210,
      "train_runtime": 73867.7678,
      "train_tokens_per_second": 12173.067
    },
    {
      "epoch": 1.8997441422030703,
      "grad_norm": 0.811306765106638,
      "learning_rate": 6.19560139354114e-08,
      "loss": 0.2988497257232666,
      "num_input_tokens_seen": 899360552,
      "step": 28215,
      "train_runtime": 73880.699,
      "train_tokens_per_second": 12173.146
    },
    {
      "epoch": 1.9000807971990303,
      "grad_norm": 0.8292822932985056,
      "learning_rate": 6.154175310602916e-08,
      "loss": 0.2988564968109131,
      "num_input_tokens_seen": 899520848,
      "step": 28220,
      "train_runtime": 73894.0811,
      "train_tokens_per_second": 12173.111
    },
    {
      "epoch": 1.9004174521949906,
      "grad_norm": 0.7541881695678712,
      "learning_rate": 6.112887330074812e-08,
      "loss": 0.2867105960845947,
      "num_input_tokens_seen": 899678784,
      "step": 28225,
      "train_runtime": 73906.9987,
      "train_tokens_per_second": 12173.12
    },
    {
      "epoch": 1.9007541071909508,
      "grad_norm": 0.8267116817071879,
      "learning_rate": 6.07173746350287e-08,
      "loss": 0.2870633602142334,
      "num_input_tokens_seen": 899840408,
      "step": 28230,
      "train_runtime": 73919.4907,
      "train_tokens_per_second": 12173.25
    },
    {
      "epoch": 1.9010907621869109,
      "grad_norm": 0.7676123475423281,
      "learning_rate": 6.030725722394548e-08,
      "loss": 0.2987704277038574,
      "num_input_tokens_seen": 900002520,
      "step": 28235,
      "train_runtime": 73931.9193,
      "train_tokens_per_second": 12173.396
    },
    {
      "epoch": 1.901427417182871,
      "grad_norm": 0.8790291176505672,
      "learning_rate": 5.98985211821862e-08,
      "loss": 0.3101794242858887,
      "num_input_tokens_seen": 900163456,
      "step": 28240,
      "train_runtime": 73945.8803,
      "train_tokens_per_second": 12173.274
    },
    {
      "epoch": 1.9017640721788311,
      "grad_norm": 1.0266858773191807,
      "learning_rate": 5.9491166624052185e-08,
      "loss": 0.29117119312286377,
      "num_input_tokens_seen": 900325512,
      "step": 28245,
      "train_runtime": 73958.4713,
      "train_tokens_per_second": 12173.393
    },
    {
      "epoch": 1.9021007271747914,
      "grad_norm": 0.789397707385831,
      "learning_rate": 5.908519366345955e-08,
      "loss": 0.28158814907073976,
      "num_input_tokens_seen": 900476712,
      "step": 28250,
      "train_runtime": 73971.6914,
      "train_tokens_per_second": 12173.261
    },
    {
      "epoch": 1.9024373821707514,
      "grad_norm": 0.8854177629166503,
      "learning_rate": 5.868060241393747e-08,
      "loss": 0.30811004638671874,
      "num_input_tokens_seen": 900640552,
      "step": 28255,
      "train_runtime": 73984.1346,
      "train_tokens_per_second": 12173.428
    },
    {
      "epoch": 1.9027740371667115,
      "grad_norm": 0.8370602838821326,
      "learning_rate": 5.8277392988627665e-08,
      "loss": 0.2847187042236328,
      "num_input_tokens_seen": 900798904,
      "step": 28260,
      "train_runtime": 73997.2946,
      "train_tokens_per_second": 12173.403
    },
    {
      "epoch": 1.9031106921626717,
      "grad_norm": 0.8048290156710194,
      "learning_rate": 5.787556550028772e-08,
      "loss": 0.2712104797363281,
      "num_input_tokens_seen": 900958520,
      "step": 28265,
      "train_runtime": 74010.3652,
      "train_tokens_per_second": 12173.41
    },
    {
      "epoch": 1.903447347158632,
      "grad_norm": 0.8516742609514703,
      "learning_rate": 5.747512006128553e-08,
      "loss": 0.3072443962097168,
      "num_input_tokens_seen": 901120152,
      "step": 28270,
      "train_runtime": 74022.8964,
      "train_tokens_per_second": 12173.533
    },
    {
      "epoch": 1.903784002154592,
      "grad_norm": 0.9024429883234057,
      "learning_rate": 5.707605678360595e-08,
      "loss": 0.3129054069519043,
      "num_input_tokens_seen": 901281544,
      "step": 28275,
      "train_runtime": 74035.9174,
      "train_tokens_per_second": 12173.572
    },
    {
      "epoch": 1.904120657150552,
      "grad_norm": 0.6987843881081706,
      "learning_rate": 5.6678375778845276e-08,
      "loss": 0.28444318771362304,
      "num_input_tokens_seen": 901443296,
      "step": 28280,
      "train_runtime": 74048.6893,
      "train_tokens_per_second": 12173.656
    },
    {
      "epoch": 1.9044573121465123,
      "grad_norm": 0.8251804291084669,
      "learning_rate": 5.6282077158213435e-08,
      "loss": 0.27478823661804197,
      "num_input_tokens_seen": 901606808,
      "step": 28285,
      "train_runtime": 74061.7315,
      "train_tokens_per_second": 12173.72
    },
    {
      "epoch": 1.9047939671424725,
      "grad_norm": 0.9621689700239028,
      "learning_rate": 5.588716103253511e-08,
      "loss": 0.3013157844543457,
      "num_input_tokens_seen": 901769256,
      "step": 28290,
      "train_runtime": 74074.3976,
      "train_tokens_per_second": 12173.832
    },
    {
      "epoch": 1.9051306221384325,
      "grad_norm": 0.8814819554953357,
      "learning_rate": 5.549362751224585e-08,
      "loss": 0.3091575622558594,
      "num_input_tokens_seen": 901925816,
      "step": 28295,
      "train_runtime": 74087.1883,
      "train_tokens_per_second": 12173.843
    },
    {
      "epoch": 1.9054672771343926,
      "grad_norm": 0.7852178287167259,
      "learning_rate": 5.5101476707397075e-08,
      "loss": 0.28136377334594725,
      "num_input_tokens_seen": 902086760,
      "step": 28300,
      "train_runtime": 74100.2633,
      "train_tokens_per_second": 12173.867
    },
    {
      "epoch": 1.9058039321303528,
      "grad_norm": 0.7610848008438819,
      "learning_rate": 5.471070872765216e-08,
      "loss": 0.313372540473938,
      "num_input_tokens_seen": 902248544,
      "step": 28305,
      "train_runtime": 74113.747,
      "train_tokens_per_second": 12173.835
    },
    {
      "epoch": 1.906140587126313,
      "grad_norm": 0.7916940591882744,
      "learning_rate": 5.432132368228815e-08,
      "loss": 0.31326045989990237,
      "num_input_tokens_seen": 902408672,
      "step": 28310,
      "train_runtime": 74126.9171,
      "train_tokens_per_second": 12173.832
    },
    {
      "epoch": 1.906477242122273,
      "grad_norm": 0.7808280945266476,
      "learning_rate": 5.393332168019572e-08,
      "loss": 0.2861791610717773,
      "num_input_tokens_seen": 902571800,
      "step": 28315,
      "train_runtime": 74139.7898,
      "train_tokens_per_second": 12173.919
    },
    {
      "epoch": 1.9068138971182331,
      "grad_norm": 0.7657708108083962,
      "learning_rate": 5.354670282987695e-08,
      "loss": 0.3028862476348877,
      "num_input_tokens_seen": 902735640,
      "step": 28320,
      "train_runtime": 74152.2322,
      "train_tokens_per_second": 12174.086
    },
    {
      "epoch": 1.9071505521141934,
      "grad_norm": 0.8783267690119936,
      "learning_rate": 5.316146723945037e-08,
      "loss": 0.3181807041168213,
      "num_input_tokens_seen": 902897328,
      "step": 28325,
      "train_runtime": 74165.2912,
      "train_tokens_per_second": 12174.122
    },
    {
      "epoch": 1.9074872071101536,
      "grad_norm": 0.8886319355882627,
      "learning_rate": 5.277761501664535e-08,
      "loss": 0.2760090589523315,
      "num_input_tokens_seen": 903055960,
      "step": 28330,
      "train_runtime": 74177.977,
      "train_tokens_per_second": 12174.179
    },
    {
      "epoch": 1.9078238621061137,
      "grad_norm": 0.9483963942110969,
      "learning_rate": 5.2395146268804934e-08,
      "loss": 0.2985908031463623,
      "num_input_tokens_seen": 903214912,
      "step": 28335,
      "train_runtime": 74190.5324,
      "train_tokens_per_second": 12174.261
    },
    {
      "epoch": 1.9081605171020737,
      "grad_norm": 0.7896943592919579,
      "learning_rate": 5.201406110288465e-08,
      "loss": 0.2892578125,
      "num_input_tokens_seen": 903374880,
      "step": 28340,
      "train_runtime": 74203.5185,
      "train_tokens_per_second": 12174.286
    },
    {
      "epoch": 1.908497172098034,
      "grad_norm": 0.7524465473696693,
      "learning_rate": 5.1634359625454266e-08,
      "loss": 0.2472975492477417,
      "num_input_tokens_seen": 903536848,
      "step": 28345,
      "train_runtime": 74218.0812,
      "train_tokens_per_second": 12174.08
    },
    {
      "epoch": 1.9088338270939942,
      "grad_norm": 0.8367092223366324,
      "learning_rate": 5.125604194269606e-08,
      "loss": 0.3181473255157471,
      "num_input_tokens_seen": 903694536,
      "step": 28350,
      "train_runtime": 74230.9586,
      "train_tokens_per_second": 12174.092
    },
    {
      "epoch": 1.9091704820899542,
      "grad_norm": 0.9119462557163616,
      "learning_rate": 5.087910816040542e-08,
      "loss": 0.3207047462463379,
      "num_input_tokens_seen": 903857728,
      "step": 28355,
      "train_runtime": 74243.8462,
      "train_tokens_per_second": 12174.177
    },
    {
      "epoch": 1.9095071370859142,
      "grad_norm": 0.8024103287884415,
      "learning_rate": 5.0503558383990235e-08,
      "loss": 0.28914742469787597,
      "num_input_tokens_seen": 904018224,
      "step": 28360,
      "train_runtime": 74256.2757,
      "train_tokens_per_second": 12174.301
    },
    {
      "epoch": 1.9098437920818745,
      "grad_norm": 0.8307908882881928,
      "learning_rate": 5.0129392718472614e-08,
      "loss": 0.29242515563964844,
      "num_input_tokens_seen": 904177264,
      "step": 28365,
      "train_runtime": 74269.4947,
      "train_tokens_per_second": 12174.275
    },
    {
      "epoch": 1.9101804470778347,
      "grad_norm": 0.7647599369130752,
      "learning_rate": 4.9756611268486077e-08,
      "loss": 0.30567944049835205,
      "num_input_tokens_seen": 904339656,
      "step": 28370,
      "train_runtime": 74282.8007,
      "train_tokens_per_second": 12174.281
    },
    {
      "epoch": 1.9105171020737948,
      "grad_norm": 0.8993852462747537,
      "learning_rate": 4.9385214138277795e-08,
      "loss": 0.3096409797668457,
      "num_input_tokens_seen": 904502184,
      "step": 28375,
      "train_runtime": 74295.2507,
      "train_tokens_per_second": 12174.428
    },
    {
      "epoch": 1.9108537570697548,
      "grad_norm": 0.8256392491941635,
      "learning_rate": 4.901520143170857e-08,
      "loss": 0.30844037532806395,
      "num_input_tokens_seen": 904663888,
      "step": 28380,
      "train_runtime": 74308.2605,
      "train_tokens_per_second": 12174.473
    },
    {
      "epoch": 1.911190412065715,
      "grad_norm": 0.805201873224448,
      "learning_rate": 4.864657325225064e-08,
      "loss": 0.29010715484619143,
      "num_input_tokens_seen": 904827728,
      "step": 28385,
      "train_runtime": 74320.696,
      "train_tokens_per_second": 12174.64
    },
    {
      "epoch": 1.9115270670616753,
      "grad_norm": 0.9204060107064614,
      "learning_rate": 4.827932970298932e-08,
      "loss": 0.2752652168273926,
      "num_input_tokens_seen": 904984472,
      "step": 28390,
      "train_runtime": 74333.6273,
      "train_tokens_per_second": 12174.631
    },
    {
      "epoch": 1.9118637220576353,
      "grad_norm": 0.7887619433528998,
      "learning_rate": 4.7913470886624125e-08,
      "loss": 0.29566154479980467,
      "num_input_tokens_seen": 905143216,
      "step": 28395,
      "train_runtime": 74346.6274,
      "train_tokens_per_second": 12174.637
    },
    {
      "epoch": 1.9122003770535954,
      "grad_norm": 0.8722006140212124,
      "learning_rate": 4.754899690546544e-08,
      "loss": 0.2814087152481079,
      "num_input_tokens_seen": 905303424,
      "step": 28400,
      "train_runtime": 74359.7355,
      "train_tokens_per_second": 12174.646
    },
    {
      "epoch": 1.9125370320495556,
      "grad_norm": 0.8663852395026291,
      "learning_rate": 4.718590786143729e-08,
      "loss": 0.29212179183959963,
      "num_input_tokens_seen": 905463264,
      "step": 28405,
      "train_runtime": 74372.8358,
      "train_tokens_per_second": 12174.65
    },
    {
      "epoch": 1.9128736870455159,
      "grad_norm": 0.824399639283976,
      "learning_rate": 4.6824203856076776e-08,
      "loss": 0.2857960224151611,
      "num_input_tokens_seen": 905627104,
      "step": 28410,
      "train_runtime": 74385.2661,
      "train_tokens_per_second": 12174.818
    },
    {
      "epoch": 1.9132103420414759,
      "grad_norm": 0.7620293495013702,
      "learning_rate": 4.646388499053356e-08,
      "loss": 0.2752853870391846,
      "num_input_tokens_seen": 905786488,
      "step": 28415,
      "train_runtime": 74398.3072,
      "train_tokens_per_second": 12174.827
    },
    {
      "epoch": 1.913546997037436,
      "grad_norm": 0.8514217195501339,
      "learning_rate": 4.6104951365568696e-08,
      "loss": 0.28073368072509763,
      "num_input_tokens_seen": 905947448,
      "step": 28420,
      "train_runtime": 74411.4115,
      "train_tokens_per_second": 12174.846
    },
    {
      "epoch": 1.9138836520333962,
      "grad_norm": 0.8541526604217446,
      "learning_rate": 4.574740308155801e-08,
      "loss": 0.2872061252593994,
      "num_input_tokens_seen": 906106080,
      "step": 28425,
      "train_runtime": 74425.2764,
      "train_tokens_per_second": 12174.709
    },
    {
      "epoch": 1.9142203070293564,
      "grad_norm": 0.8723180176558147,
      "learning_rate": 4.539124023848762e-08,
      "loss": 0.29244508743286135,
      "num_input_tokens_seen": 906269080,
      "step": 28430,
      "train_runtime": 74438.1052,
      "train_tokens_per_second": 12174.8
    },
    {
      "epoch": 1.9145569620253164,
      "grad_norm": 0.8356327897044589,
      "learning_rate": 4.503646293595787e-08,
      "loss": 0.32239794731140137,
      "num_input_tokens_seen": 906425528,
      "step": 28435,
      "train_runtime": 74451.3665,
      "train_tokens_per_second": 12174.733
    },
    {
      "epoch": 1.9148936170212765,
      "grad_norm": 0.8731899153844053,
      "learning_rate": 4.468307127318105e-08,
      "loss": 0.29468064308166503,
      "num_input_tokens_seen": 906586472,
      "step": 28440,
      "train_runtime": 74463.8279,
      "train_tokens_per_second": 12174.857
    },
    {
      "epoch": 1.9152302720172367,
      "grad_norm": 0.942609216376018,
      "learning_rate": 4.4331065348982014e-08,
      "loss": 0.2866788864135742,
      "num_input_tokens_seen": 906746096,
      "step": 28445,
      "train_runtime": 74476.7126,
      "train_tokens_per_second": 12174.894
    },
    {
      "epoch": 1.915566927013197,
      "grad_norm": 0.8400296372211845,
      "learning_rate": 4.3980445261798125e-08,
      "loss": 0.29766771793365476,
      "num_input_tokens_seen": 906907192,
      "step": 28450,
      "train_runtime": 74490.2117,
      "train_tokens_per_second": 12174.851
    },
    {
      "epoch": 1.915903582009157,
      "grad_norm": 0.8320138520570873,
      "learning_rate": 4.363121110967927e-08,
      "loss": 0.270781946182251,
      "num_input_tokens_seen": 907063096,
      "step": 28455,
      "train_runtime": 74503.629,
      "train_tokens_per_second": 12174.751
    },
    {
      "epoch": 1.916240237005117,
      "grad_norm": 0.7568959791799943,
      "learning_rate": 4.328336299028735e-08,
      "loss": 0.27180256843566897,
      "num_input_tokens_seen": 907220912,
      "step": 28460,
      "train_runtime": 74517.1927,
      "train_tokens_per_second": 12174.652
    },
    {
      "epoch": 1.9165768920010773,
      "grad_norm": 0.8190191887227736,
      "learning_rate": 4.293690100089731e-08,
      "loss": 0.2916971921920776,
      "num_input_tokens_seen": 907379240,
      "step": 28465,
      "train_runtime": 74530.5673,
      "train_tokens_per_second": 12174.592
    },
    {
      "epoch": 1.9169135469970375,
      "grad_norm": 0.8893354425757927,
      "learning_rate": 4.259182523839556e-08,
      "loss": 0.2838243246078491,
      "num_input_tokens_seen": 907537048,
      "step": 28470,
      "train_runtime": 74543.2797,
      "train_tokens_per_second": 12174.633
    },
    {
      "epoch": 1.9172502019929976,
      "grad_norm": 0.8882472197317557,
      "learning_rate": 4.224813579928211e-08,
      "loss": 0.31717352867126464,
      "num_input_tokens_seen": 907697680,
      "step": 28475,
      "train_runtime": 74556.1353,
      "train_tokens_per_second": 12174.688
    },
    {
      "epoch": 1.9175868569889576,
      "grad_norm": 0.7955271893747686,
      "learning_rate": 4.190583277966842e-08,
      "loss": 0.300145697593689,
      "num_input_tokens_seen": 907859480,
      "step": 28480,
      "train_runtime": 74570.0307,
      "train_tokens_per_second": 12174.589
    },
    {
      "epoch": 1.9179235119849178,
      "grad_norm": 0.7797486900403856,
      "learning_rate": 4.1564916275278477e-08,
      "loss": 0.2774558782577515,
      "num_input_tokens_seen": 908015248,
      "step": 28485,
      "train_runtime": 74583.1816,
      "train_tokens_per_second": 12174.531
    },
    {
      "epoch": 1.918260166980878,
      "grad_norm": 0.8104200457045871,
      "learning_rate": 4.122538638144824e-08,
      "loss": 0.2908132553100586,
      "num_input_tokens_seen": 908176856,
      "step": 28490,
      "train_runtime": 74596.7921,
      "train_tokens_per_second": 12174.476
    },
    {
      "epoch": 1.9185968219768381,
      "grad_norm": 0.7546098434315309,
      "learning_rate": 4.08872431931262e-08,
      "loss": 0.2742162227630615,
      "num_input_tokens_seen": 908337488,
      "step": 28495,
      "train_runtime": 74610.2244,
      "train_tokens_per_second": 12174.437
    },
    {
      "epoch": 1.9189334769727981,
      "grad_norm": 0.8060257843943104,
      "learning_rate": 4.055048680487339e-08,
      "loss": 0.2883006572723389,
      "num_input_tokens_seen": 908498280,
      "step": 28500,
      "train_runtime": 74623.3296,
      "train_tokens_per_second": 12174.454
    },
    {
      "epoch": 1.9192701319687584,
      "grad_norm": 0.8154788993660225,
      "learning_rate": 4.0215117310862806e-08,
      "loss": 0.31098692417144774,
      "num_input_tokens_seen": 908647968,
      "step": 28505,
      "train_runtime": 74635.8841,
      "train_tokens_per_second": 12174.412
    },
    {
      "epoch": 1.9196067869647186,
      "grad_norm": 0.8901780919215156,
      "learning_rate": 3.9881134804878316e-08,
      "loss": 0.3031690359115601,
      "num_input_tokens_seen": 908802192,
      "step": 28510,
      "train_runtime": 74648.6039,
      "train_tokens_per_second": 12174.403
    },
    {
      "epoch": 1.9199434419606787,
      "grad_norm": 0.8754799599616853,
      "learning_rate": 3.9548539380318e-08,
      "loss": 0.3124077320098877,
      "num_input_tokens_seen": 908961976,
      "step": 28515,
      "train_runtime": 74661.0579,
      "train_tokens_per_second": 12174.512
    },
    {
      "epoch": 1.9202800969566387,
      "grad_norm": 0.8154962644748002,
      "learning_rate": 3.921733113019077e-08,
      "loss": 0.2959012985229492,
      "num_input_tokens_seen": 909115728,
      "step": 28520,
      "train_runtime": 74674.4707,
      "train_tokens_per_second": 12174.385
    },
    {
      "epoch": 1.920616751952599,
      "grad_norm": 0.7963800256502893,
      "learning_rate": 3.888751014711867e-08,
      "loss": 0.3138494253158569,
      "num_input_tokens_seen": 909268344,
      "step": 28525,
      "train_runtime": 74687.8668,
      "train_tokens_per_second": 12174.244
    },
    {
      "epoch": 1.9209534069485592,
      "grad_norm": 1.0836750738140328,
      "learning_rate": 3.855907652333402e-08,
      "loss": 0.3021607160568237,
      "num_input_tokens_seen": 909430832,
      "step": 28530,
      "train_runtime": 74700.5402,
      "train_tokens_per_second": 12174.354
    },
    {
      "epoch": 1.9212900619445192,
      "grad_norm": 0.8343585400415516,
      "learning_rate": 3.8232030350683344e-08,
      "loss": 0.2855525016784668,
      "num_input_tokens_seen": 909594672,
      "step": 28535,
      "train_runtime": 74713.5717,
      "train_tokens_per_second": 12174.424
    },
    {
      "epoch": 1.9216267169404793,
      "grad_norm": 0.897868415984019,
      "learning_rate": 3.7906371720622946e-08,
      "loss": 0.3006442070007324,
      "num_input_tokens_seen": 909755856,
      "step": 28540,
      "train_runtime": 74726.0465,
      "train_tokens_per_second": 12174.548
    },
    {
      "epoch": 1.9219633719364395,
      "grad_norm": 0.77685765574357,
      "learning_rate": 3.758210072422275e-08,
      "loss": 0.27744030952453613,
      "num_input_tokens_seen": 909913496,
      "step": 28545,
      "train_runtime": 74738.5227,
      "train_tokens_per_second": 12174.625
    },
    {
      "epoch": 1.9223000269323998,
      "grad_norm": 0.9330747557127657,
      "learning_rate": 3.72592174521641e-08,
      "loss": 0.32152185440063474,
      "num_input_tokens_seen": 910076432,
      "step": 28550,
      "train_runtime": 74751.2797,
      "train_tokens_per_second": 12174.727
    },
    {
      "epoch": 1.9226366819283598,
      "grad_norm": 0.7538603417361607,
      "learning_rate": 3.693772199474033e-08,
      "loss": 0.2951069355010986,
      "num_input_tokens_seen": 910237384,
      "step": 28555,
      "train_runtime": 74764.4317,
      "train_tokens_per_second": 12174.738
    },
    {
      "epoch": 1.9229733369243198,
      "grad_norm": 0.8115385877941742,
      "learning_rate": 3.661761444185674e-08,
      "loss": 0.29569182395935056,
      "num_input_tokens_seen": 910396240,
      "step": 28560,
      "train_runtime": 74777.2775,
      "train_tokens_per_second": 12174.771
    },
    {
      "epoch": 1.92330999192028,
      "grad_norm": 0.8793323894633963,
      "learning_rate": 3.6298894883030066e-08,
      "loss": 0.2983766317367554,
      "num_input_tokens_seen": 910556992,
      "step": 28565,
      "train_runtime": 74790.1602,
      "train_tokens_per_second": 12174.823
    },
    {
      "epoch": 1.9236466469162403,
      "grad_norm": 0.8652189388099313,
      "learning_rate": 3.5981563407389007e-08,
      "loss": 0.3157456159591675,
      "num_input_tokens_seen": 910716352,
      "step": 28570,
      "train_runtime": 74803.3056,
      "train_tokens_per_second": 12174.814
    },
    {
      "epoch": 1.9239833019122003,
      "grad_norm": 0.8477611396008407,
      "learning_rate": 3.566562010367536e-08,
      "loss": 0.2746917724609375,
      "num_input_tokens_seen": 910871640,
      "step": 28575,
      "train_runtime": 74816.2575,
      "train_tokens_per_second": 12174.782
    },
    {
      "epoch": 1.9243199569081604,
      "grad_norm": 0.7501412966120312,
      "learning_rate": 3.535106506024011e-08,
      "loss": 0.2954802274703979,
      "num_input_tokens_seen": 911033064,
      "step": 28580,
      "train_runtime": 74829.3391,
      "train_tokens_per_second": 12174.811
    },
    {
      "epoch": 1.9246566119041206,
      "grad_norm": 0.8401732696889453,
      "learning_rate": 3.5037898365049005e-08,
      "loss": 0.2661609411239624,
      "num_input_tokens_seen": 911193480,
      "step": 28585,
      "train_runtime": 74842.513,
      "train_tokens_per_second": 12174.811
    },
    {
      "epoch": 1.9249932669000809,
      "grad_norm": 0.8349397355536027,
      "learning_rate": 3.472612010567755e-08,
      "loss": 0.28289783000946045,
      "num_input_tokens_seen": 911346480,
      "step": 28590,
      "train_runtime": 74855.9391,
      "train_tokens_per_second": 12174.672
    },
    {
      "epoch": 1.925329921896041,
      "grad_norm": 0.824188973587861,
      "learning_rate": 3.441573036931323e-08,
      "loss": 0.311726975440979,
      "num_input_tokens_seen": 911509520,
      "step": 28595,
      "train_runtime": 74868.9895,
      "train_tokens_per_second": 12174.727
    },
    {
      "epoch": 1.925666576892001,
      "grad_norm": 0.8295410620750222,
      "learning_rate": 3.410672924275604e-08,
      "loss": 0.2714081287384033,
      "num_input_tokens_seen": 911672512,
      "step": 28600,
      "train_runtime": 74881.7961,
      "train_tokens_per_second": 12174.822
    },
    {
      "epoch": 1.9260032318879612,
      "grad_norm": 0.8383477799389843,
      "learning_rate": 3.3799116812416876e-08,
      "loss": 0.31270813941955566,
      "num_input_tokens_seen": 911832952,
      "step": 28605,
      "train_runtime": 74894.8388,
      "train_tokens_per_second": 12174.844
    },
    {
      "epoch": 1.9263398868839214,
      "grad_norm": 0.8982068931458073,
      "learning_rate": 3.349289316431803e-08,
      "loss": 0.3105414152145386,
      "num_input_tokens_seen": 911994056,
      "step": 28610,
      "train_runtime": 74907.7439,
      "train_tokens_per_second": 12174.897
    },
    {
      "epoch": 1.9266765418798815,
      "grad_norm": 0.7884981532710531,
      "learning_rate": 3.3188058384095446e-08,
      "loss": 0.2929044723510742,
      "num_input_tokens_seen": 912149824,
      "step": 28615,
      "train_runtime": 74920.5381,
      "train_tokens_per_second": 12174.897
    },
    {
      "epoch": 1.9270131968758415,
      "grad_norm": 0.8526771036725423,
      "learning_rate": 3.2884612556993714e-08,
      "loss": 0.3126600980758667,
      "num_input_tokens_seen": 912310496,
      "step": 28620,
      "train_runtime": 74932.9584,
      "train_tokens_per_second": 12175.023
    },
    {
      "epoch": 1.9273498518718017,
      "grad_norm": 0.8921230463010991,
      "learning_rate": 3.258255576787161e-08,
      "loss": 0.3061052322387695,
      "num_input_tokens_seen": 912468968,
      "step": 28625,
      "train_runtime": 74945.7275,
      "train_tokens_per_second": 12175.063
    },
    {
      "epoch": 1.927686506867762,
      "grad_norm": 0.93218844870514,
      "learning_rate": 3.22818881011977e-08,
      "loss": 0.31447610855102537,
      "num_input_tokens_seen": 912632688,
      "step": 28630,
      "train_runtime": 74959.7849,
      "train_tokens_per_second": 12174.964
    },
    {
      "epoch": 1.928023161863722,
      "grad_norm": 0.9752478569951576,
      "learning_rate": 3.198260964105249e-08,
      "loss": 0.27275815010070803,
      "num_input_tokens_seen": 912788024,
      "step": 28635,
      "train_runtime": 74972.4016,
      "train_tokens_per_second": 12174.987
    },
    {
      "epoch": 1.928359816859682,
      "grad_norm": 0.80392987898737,
      "learning_rate": 3.168472047112903e-08,
      "loss": 0.2988280773162842,
      "num_input_tokens_seen": 912945632,
      "step": 28640,
      "train_runtime": 74984.9197,
      "train_tokens_per_second": 12175.056
    },
    {
      "epoch": 1.9286964718556423,
      "grad_norm": 0.8516783107242409,
      "learning_rate": 3.138822067473068e-08,
      "loss": 0.282541823387146,
      "num_input_tokens_seen": 913104528,
      "step": 28645,
      "train_runtime": 74997.9625,
      "train_tokens_per_second": 12175.058
    },
    {
      "epoch": 1.9290331268516026,
      "grad_norm": 0.7826579926986188,
      "learning_rate": 3.10931103347728e-08,
      "loss": 0.30090980529785155,
      "num_input_tokens_seen": 913266256,
      "step": 28650,
      "train_runtime": 75010.4306,
      "train_tokens_per_second": 12175.19
    },
    {
      "epoch": 1.9293697818475626,
      "grad_norm": 0.8238918570773871,
      "learning_rate": 3.0799389533781587e-08,
      "loss": 0.3158176183700562,
      "num_input_tokens_seen": 913425064,
      "step": 28655,
      "train_runtime": 75022.9116,
      "train_tokens_per_second": 12175.281
    },
    {
      "epoch": 1.9297064368435226,
      "grad_norm": 0.8405112609956897,
      "learning_rate": 3.050705835389578e-08,
      "loss": 0.2814477443695068,
      "num_input_tokens_seen": 913583312,
      "step": 28660,
      "train_runtime": 75035.7511,
      "train_tokens_per_second": 12175.307
    },
    {
      "epoch": 1.9300430918394829,
      "grad_norm": 0.8347298052448817,
      "learning_rate": 3.021611687686443e-08,
      "loss": 0.28585822582244874,
      "num_input_tokens_seen": 913736344,
      "step": 28665,
      "train_runtime": 75048.7454,
      "train_tokens_per_second": 12175.238
    },
    {
      "epoch": 1.9303797468354431,
      "grad_norm": 0.8201032089668052,
      "learning_rate": 2.992656518404857e-08,
      "loss": 0.3008700370788574,
      "num_input_tokens_seen": 913893944,
      "step": 28670,
      "train_runtime": 75062.351,
      "train_tokens_per_second": 12175.131
    },
    {
      "epoch": 1.9307164018314031,
      "grad_norm": 0.7370696629095344,
      "learning_rate": 2.96384033564201e-08,
      "loss": 0.28259131908416746,
      "num_input_tokens_seen": 914053704,
      "step": 28675,
      "train_runtime": 75076.2412,
      "train_tokens_per_second": 12175.006
    },
    {
      "epoch": 1.9310530568273632,
      "grad_norm": 0.8593889260996549,
      "learning_rate": 2.935163147456288e-08,
      "loss": 0.25777606964111327,
      "num_input_tokens_seen": 914213928,
      "step": 28680,
      "train_runtime": 75089.4992,
      "train_tokens_per_second": 12174.99
    },
    {
      "epoch": 1.9313897118233234,
      "grad_norm": 0.8198702800037067,
      "learning_rate": 2.906624961867166e-08,
      "loss": 0.2948530912399292,
      "num_input_tokens_seen": 914373584,
      "step": 28685,
      "train_runtime": 75102.7862,
      "train_tokens_per_second": 12174.962
    },
    {
      "epoch": 1.9317263668192837,
      "grad_norm": 0.8289033133930079,
      "learning_rate": 2.8782257868553154e-08,
      "loss": 0.264910364151001,
      "num_input_tokens_seen": 914531520,
      "step": 28690,
      "train_runtime": 75115.9747,
      "train_tokens_per_second": 12174.927
    },
    {
      "epoch": 1.9320630218152437,
      "grad_norm": 0.8320342971157785,
      "learning_rate": 2.849965630362328e-08,
      "loss": 0.3063076019287109,
      "num_input_tokens_seen": 914683632,
      "step": 28695,
      "train_runtime": 75128.8173,
      "train_tokens_per_second": 12174.871
    },
    {
      "epoch": 1.9323996768112037,
      "grad_norm": 0.8522074930577093,
      "learning_rate": 2.8218445002912153e-08,
      "loss": 0.29024419784545896,
      "num_input_tokens_seen": 914846616,
      "step": 28700,
      "train_runtime": 75141.6334,
      "train_tokens_per_second": 12174.963
    },
    {
      "epoch": 1.932736331807164,
      "grad_norm": 0.8623257175216014,
      "learning_rate": 2.7938624045059094e-08,
      "loss": 0.27026126384735105,
      "num_input_tokens_seen": 915005264,
      "step": 28705,
      "train_runtime": 75155.0092,
      "train_tokens_per_second": 12174.907
    },
    {
      "epoch": 1.9330729868031242,
      "grad_norm": 0.9294517921718258,
      "learning_rate": 2.766019350831428e-08,
      "loss": 0.29651756286621095,
      "num_input_tokens_seen": 915169104,
      "step": 28710,
      "train_runtime": 75167.4435,
      "train_tokens_per_second": 12175.073
    },
    {
      "epoch": 1.9334096417990843,
      "grad_norm": 0.841329360502712,
      "learning_rate": 2.7383153470541546e-08,
      "loss": 0.28957486152648926,
      "num_input_tokens_seen": 915329968,
      "step": 28715,
      "train_runtime": 75179.8978,
      "train_tokens_per_second": 12175.196
    },
    {
      "epoch": 1.9337462967950443,
      "grad_norm": 0.7893135610406274,
      "learning_rate": 2.710750400921336e-08,
      "loss": 0.2849695682525635,
      "num_input_tokens_seen": 915482888,
      "step": 28720,
      "train_runtime": 75193.5317,
      "train_tokens_per_second": 12175.022
    },
    {
      "epoch": 1.9340829517910045,
      "grad_norm": 0.8180457029686823,
      "learning_rate": 2.683324520141417e-08,
      "loss": 0.2861651420593262,
      "num_input_tokens_seen": 915643272,
      "step": 28725,
      "train_runtime": 75206.5938,
      "train_tokens_per_second": 12175.04
    },
    {
      "epoch": 1.9344196067869648,
      "grad_norm": 0.827301847670361,
      "learning_rate": 2.65603771238393e-08,
      "loss": 0.29910333156585694,
      "num_input_tokens_seen": 915803704,
      "step": 28730,
      "train_runtime": 75220.196,
      "train_tokens_per_second": 12174.971
    },
    {
      "epoch": 1.9347562617829248,
      "grad_norm": 0.836620763360012,
      "learning_rate": 2.6288899852796036e-08,
      "loss": 0.28818683624267577,
      "num_input_tokens_seen": 915967544,
      "step": 28735,
      "train_runtime": 75232.6234,
      "train_tokens_per_second": 12175.138
    },
    {
      "epoch": 1.9350929167788848,
      "grad_norm": 0.9175817750709617,
      "learning_rate": 2.6018813464202543e-08,
      "loss": 0.29910736083984374,
      "num_input_tokens_seen": 916124504,
      "step": 28740,
      "train_runtime": 75246.3502,
      "train_tokens_per_second": 12175.003
    },
    {
      "epoch": 1.935429571774845,
      "grad_norm": 0.783333397709063,
      "learning_rate": 2.575011803358618e-08,
      "loss": 0.29322926998138427,
      "num_input_tokens_seen": 916284824,
      "step": 28745,
      "train_runtime": 75259.7328,
      "train_tokens_per_second": 12174.968
    },
    {
      "epoch": 1.9357662267708053,
      "grad_norm": 0.701525719352334,
      "learning_rate": 2.5482813636087955e-08,
      "loss": 0.27543609142303466,
      "num_input_tokens_seen": 916439744,
      "step": 28750,
      "train_runtime": 75272.6038,
      "train_tokens_per_second": 12174.944
    },
    {
      "epoch": 1.9361028817667654,
      "grad_norm": 0.8298937022459818,
      "learning_rate": 2.5216900346458073e-08,
      "loss": 0.2751685380935669,
      "num_input_tokens_seen": 916599520,
      "step": 28755,
      "train_runtime": 75285.1376,
      "train_tokens_per_second": 12175.039
    },
    {
      "epoch": 1.9364395367627254,
      "grad_norm": 0.7697529406346881,
      "learning_rate": 2.4952378239058716e-08,
      "loss": 0.2747399091720581,
      "num_input_tokens_seen": 916759552,
      "step": 28760,
      "train_runtime": 75298.0327,
      "train_tokens_per_second": 12175.08
    },
    {
      "epoch": 1.9367761917586857,
      "grad_norm": 0.8079101443439097,
      "learning_rate": 2.4689247387862934e-08,
      "loss": 0.28501172065734864,
      "num_input_tokens_seen": 916917360,
      "step": 28765,
      "train_runtime": 75310.9807,
      "train_tokens_per_second": 12175.082
    },
    {
      "epoch": 1.937112846754646,
      "grad_norm": 0.7851748828534685,
      "learning_rate": 2.4427507866453537e-08,
      "loss": 0.2746150493621826,
      "num_input_tokens_seen": 917072808,
      "step": 28770,
      "train_runtime": 75325.7647,
      "train_tokens_per_second": 12174.756
    },
    {
      "epoch": 1.937449501750606,
      "grad_norm": 0.8896379996416431,
      "learning_rate": 2.416715974802586e-08,
      "loss": 0.2909178972244263,
      "num_input_tokens_seen": 917230184,
      "step": 28775,
      "train_runtime": 75338.8546,
      "train_tokens_per_second": 12174.73
    },
    {
      "epoch": 1.937786156746566,
      "grad_norm": 0.7736748263871922,
      "learning_rate": 2.3908203105384997e-08,
      "loss": 0.26843369007110596,
      "num_input_tokens_seen": 917390192,
      "step": 28780,
      "train_runtime": 75352.4592,
      "train_tokens_per_second": 12174.655
    },
    {
      "epoch": 1.9381228117425262,
      "grad_norm": 0.8213438795286101,
      "learning_rate": 2.365063801094747e-08,
      "loss": 0.27482404708862307,
      "num_input_tokens_seen": 917550872,
      "step": 28785,
      "train_runtime": 75365.8562,
      "train_tokens_per_second": 12174.623
    },
    {
      "epoch": 1.9384594667384865,
      "grad_norm": 0.7332889358688368,
      "learning_rate": 2.3394464536740658e-08,
      "loss": 0.2989178657531738,
      "num_input_tokens_seen": 917714712,
      "step": 28790,
      "train_runtime": 75378.2852,
      "train_tokens_per_second": 12174.789
    },
    {
      "epoch": 1.9387961217344465,
      "grad_norm": 0.9068751842099144,
      "learning_rate": 2.3139682754402816e-08,
      "loss": 0.3104221343994141,
      "num_input_tokens_seen": 917877496,
      "step": 28795,
      "train_runtime": 75391.0189,
      "train_tokens_per_second": 12174.892
    },
    {
      "epoch": 1.9391327767304065,
      "grad_norm": 0.724166890937452,
      "learning_rate": 2.2886292735182502e-08,
      "loss": 0.27684361934661866,
      "num_input_tokens_seen": 918041168,
      "step": 28800,
      "train_runtime": 75404.7281,
      "train_tokens_per_second": 12174.849
    },
    {
      "epoch": 1.9394694317263668,
      "grad_norm": 0.7982531372666478,
      "learning_rate": 2.2634294549939705e-08,
      "loss": 0.27920310497283934,
      "num_input_tokens_seen": 918204136,
      "step": 28805,
      "train_runtime": 75417.5531,
      "train_tokens_per_second": 12174.939
    },
    {
      "epoch": 1.939806086722327,
      "grad_norm": 0.8641827141609502,
      "learning_rate": 2.2383688269144723e-08,
      "loss": 0.35346529483795164,
      "num_input_tokens_seen": 918363104,
      "step": 28810,
      "train_runtime": 75430.1608,
      "train_tokens_per_second": 12175.012
    },
    {
      "epoch": 1.940142741718287,
      "grad_norm": 0.8240191785371843,
      "learning_rate": 2.2134473962878712e-08,
      "loss": 0.2800391674041748,
      "num_input_tokens_seen": 918522744,
      "step": 28815,
      "train_runtime": 75442.9041,
      "train_tokens_per_second": 12175.071
    },
    {
      "epoch": 1.940479396714247,
      "grad_norm": 0.98048516221259,
      "learning_rate": 2.1886651700833705e-08,
      "loss": 0.2802724838256836,
      "num_input_tokens_seen": 918672136,
      "step": 28820,
      "train_runtime": 75456.2678,
      "train_tokens_per_second": 12174.895
    },
    {
      "epoch": 1.9408160517102075,
      "grad_norm": 0.7949373780734291,
      "learning_rate": 2.1640221552312603e-08,
      "loss": 0.3030146598815918,
      "num_input_tokens_seen": 918834352,
      "step": 28825,
      "train_runtime": 75469.5797,
      "train_tokens_per_second": 12174.897
    },
    {
      "epoch": 1.9411527067061676,
      "grad_norm": 0.8264078243959045,
      "learning_rate": 2.1395183586229163e-08,
      "loss": 0.29787449836730956,
      "num_input_tokens_seen": 918997744,
      "step": 28830,
      "train_runtime": 75482.532,
      "train_tokens_per_second": 12174.972
    },
    {
      "epoch": 1.9414893617021276,
      "grad_norm": 0.8223092520169791,
      "learning_rate": 2.1151537871106353e-08,
      "loss": 0.29082038402557375,
      "num_input_tokens_seen": 919152352,
      "step": 28835,
      "train_runtime": 75495.6136,
      "train_tokens_per_second": 12174.911
    },
    {
      "epoch": 1.9418260166980879,
      "grad_norm": 0.8794137404330502,
      "learning_rate": 2.0909284475080226e-08,
      "loss": 0.2737832546234131,
      "num_input_tokens_seen": 919310648,
      "step": 28840,
      "train_runtime": 75508.7469,
      "train_tokens_per_second": 12174.889
    },
    {
      "epoch": 1.942162671694048,
      "grad_norm": 0.8439272284428391,
      "learning_rate": 2.0668423465894928e-08,
      "loss": 0.281263542175293,
      "num_input_tokens_seen": 919469136,
      "step": 28845,
      "train_runtime": 75521.5422,
      "train_tokens_per_second": 12174.925
    },
    {
      "epoch": 1.9424993266900081,
      "grad_norm": 0.912412397345424,
      "learning_rate": 2.0428954910907684e-08,
      "loss": 0.28661403656005857,
      "num_input_tokens_seen": 919619192,
      "step": 28850,
      "train_runtime": 75535.0224,
      "train_tokens_per_second": 12174.739
    },
    {
      "epoch": 1.9428359816859682,
      "grad_norm": 0.8729345054680056,
      "learning_rate": 2.019087887708382e-08,
      "loss": 0.30406627655029295,
      "num_input_tokens_seen": 919777024,
      "step": 28855,
      "train_runtime": 75548.5454,
      "train_tokens_per_second": 12174.649
    },
    {
      "epoch": 1.9431726366819284,
      "grad_norm": 0.8653081325127362,
      "learning_rate": 1.9954195431001744e-08,
      "loss": 0.2937690258026123,
      "num_input_tokens_seen": 919938880,
      "step": 28860,
      "train_runtime": 75561.044,
      "train_tokens_per_second": 12174.777
    },
    {
      "epoch": 1.9435092916778887,
      "grad_norm": 0.8578164733456476,
      "learning_rate": 1.9718904638848513e-08,
      "loss": 0.2698089122772217,
      "num_input_tokens_seen": 920096352,
      "step": 28865,
      "train_runtime": 75574.4778,
      "train_tokens_per_second": 12174.697
    },
    {
      "epoch": 1.9438459466738487,
      "grad_norm": 0.8582003830973761,
      "learning_rate": 1.9485006566422602e-08,
      "loss": 0.29454870223999025,
      "num_input_tokens_seen": 920258952,
      "step": 28870,
      "train_runtime": 75587.1592,
      "train_tokens_per_second": 12174.805
    },
    {
      "epoch": 1.9441826016698087,
      "grad_norm": 0.8871505496739368,
      "learning_rate": 1.9252501279132806e-08,
      "loss": 0.2968337059020996,
      "num_input_tokens_seen": 920416536,
      "step": 28875,
      "train_runtime": 75599.8966,
      "train_tokens_per_second": 12174.839
    },
    {
      "epoch": 1.944519256665769,
      "grad_norm": 0.7748574040561794,
      "learning_rate": 1.902138884199878e-08,
      "loss": 0.30783467292785643,
      "num_input_tokens_seen": 920576264,
      "step": 28880,
      "train_runtime": 75612.6835,
      "train_tokens_per_second": 12174.892
    },
    {
      "epoch": 1.9448559116617292,
      "grad_norm": 0.7882081854794212,
      "learning_rate": 1.879166931964993e-08,
      "loss": 0.28879528045654296,
      "num_input_tokens_seen": 920736728,
      "step": 28885,
      "train_runtime": 75626.6122,
      "train_tokens_per_second": 12174.772
    },
    {
      "epoch": 1.9451925666576892,
      "grad_norm": 0.7940039919708077,
      "learning_rate": 1.856334277632765e-08,
      "loss": 0.2984178066253662,
      "num_input_tokens_seen": 920900568,
      "step": 28890,
      "train_runtime": 75639.0331,
      "train_tokens_per_second": 12174.938
    },
    {
      "epoch": 1.9455292216536493,
      "grad_norm": 0.7490523577140491,
      "learning_rate": 1.8336409275880872e-08,
      "loss": 0.286403226852417,
      "num_input_tokens_seen": 921064184,
      "step": 28895,
      "train_runtime": 75652.0653,
      "train_tokens_per_second": 12175.004
    },
    {
      "epoch": 1.9458658766496095,
      "grad_norm": 0.8279865355364764,
      "learning_rate": 1.8110868881772713e-08,
      "loss": 0.28240299224853516,
      "num_input_tokens_seen": 921221224,
      "step": 28900,
      "train_runtime": 75665.8744,
      "train_tokens_per_second": 12174.857
    },
    {
      "epoch": 1.9462025316455698,
      "grad_norm": 0.8157353944704581,
      "learning_rate": 1.788672165707328e-08,
      "loss": 0.27784175872802735,
      "num_input_tokens_seen": 921378912,
      "step": 28905,
      "train_runtime": 75678.8574,
      "train_tokens_per_second": 12174.852
    },
    {
      "epoch": 1.9465391866415298,
      "grad_norm": 0.8776841962951271,
      "learning_rate": 1.7663967664465763e-08,
      "loss": 0.2834606647491455,
      "num_input_tokens_seen": 921531144,
      "step": 28910,
      "train_runtime": 75691.7084,
      "train_tokens_per_second": 12174.796
    },
    {
      "epoch": 1.9468758416374898,
      "grad_norm": 0.816195483283043,
      "learning_rate": 1.7442606966242005e-08,
      "loss": 0.30164783000946044,
      "num_input_tokens_seen": 921694984,
      "step": 28915,
      "train_runtime": 75704.1448,
      "train_tokens_per_second": 12174.961
    },
    {
      "epoch": 1.94721249663345,
      "grad_norm": 0.8283107796460333,
      "learning_rate": 1.722263962430526e-08,
      "loss": 0.2983541011810303,
      "num_input_tokens_seen": 921854680,
      "step": 28920,
      "train_runtime": 75716.6746,
      "train_tokens_per_second": 12175.055
    },
    {
      "epoch": 1.9475491516294103,
      "grad_norm": 0.8376735143756399,
      "learning_rate": 1.7004065700168547e-08,
      "loss": 0.270952033996582,
      "num_input_tokens_seen": 922015936,
      "step": 28925,
      "train_runtime": 75729.4109,
      "train_tokens_per_second": 12175.137
    },
    {
      "epoch": 1.9478858066253704,
      "grad_norm": 0.778301474218795,
      "learning_rate": 1.6786885254954644e-08,
      "loss": 0.2830474615097046,
      "num_input_tokens_seen": 922179744,
      "step": 28930,
      "train_runtime": 75742.5317,
      "train_tokens_per_second": 12175.19
    },
    {
      "epoch": 1.9482224616213304,
      "grad_norm": 0.906311661340437,
      "learning_rate": 1.6571098349398296e-08,
      "loss": 0.2901591777801514,
      "num_input_tokens_seen": 922332128,
      "step": 28935,
      "train_runtime": 75756.2107,
      "train_tokens_per_second": 12175.003
    },
    {
      "epoch": 1.9485591166172906,
      "grad_norm": 0.8518320646022326,
      "learning_rate": 1.635670504384346e-08,
      "loss": 0.30651230812072755,
      "num_input_tokens_seen": 922489824,
      "step": 28940,
      "train_runtime": 75769.9359,
      "train_tokens_per_second": 12174.879
    },
    {
      "epoch": 1.948895771613251,
      "grad_norm": 0.9377215700407561,
      "learning_rate": 1.6143705398243837e-08,
      "loss": 0.276810884475708,
      "num_input_tokens_seen": 922646472,
      "step": 28945,
      "train_runtime": 75783.8555,
      "train_tokens_per_second": 12174.71
    },
    {
      "epoch": 1.949232426609211,
      "grad_norm": 0.871398917601628,
      "learning_rate": 1.5932099472165674e-08,
      "loss": 0.2886288404464722,
      "num_input_tokens_seen": 922802576,
      "step": 28950,
      "train_runtime": 75797.3249,
      "train_tokens_per_second": 12174.606
    },
    {
      "epoch": 1.949569081605171,
      "grad_norm": 0.7707519205131593,
      "learning_rate": 1.572188732478164e-08,
      "loss": 0.2909992218017578,
      "num_input_tokens_seen": 922966216,
      "step": 28955,
      "train_runtime": 75810.3354,
      "train_tokens_per_second": 12174.675
    },
    {
      "epoch": 1.9499057366011312,
      "grad_norm": 0.7765522020228335,
      "learning_rate": 1.551306901487859e-08,
      "loss": 0.2856043815612793,
      "num_input_tokens_seen": 923124824,
      "step": 28960,
      "train_runtime": 75824.096,
      "train_tokens_per_second": 12174.558
    },
    {
      "epoch": 1.9502423915970915,
      "grad_norm": 0.7372001456959646,
      "learning_rate": 1.5305644600852043e-08,
      "loss": 0.2813275814056396,
      "num_input_tokens_seen": 923288568,
      "step": 28965,
      "train_runtime": 75837.8895,
      "train_tokens_per_second": 12174.502
    },
    {
      "epoch": 1.9505790465930515,
      "grad_norm": 0.8056836583340231,
      "learning_rate": 1.5099614140706154e-08,
      "loss": 0.2971454620361328,
      "num_input_tokens_seen": 923450168,
      "step": 28970,
      "train_runtime": 75851.2904,
      "train_tokens_per_second": 12174.482
    },
    {
      "epoch": 1.9509157015890115,
      "grad_norm": 0.8240533851004691,
      "learning_rate": 1.489497769205761e-08,
      "loss": 0.30417983531951903,
      "num_input_tokens_seen": 923610520,
      "step": 28975,
      "train_runtime": 75864.9032,
      "train_tokens_per_second": 12174.411
    },
    {
      "epoch": 1.9512523565849718,
      "grad_norm": 0.7947775367457869,
      "learning_rate": 1.4691735312132304e-08,
      "loss": 0.28345279693603515,
      "num_input_tokens_seen": 923774360,
      "step": 28980,
      "train_runtime": 75877.3131,
      "train_tokens_per_second": 12174.579
    },
    {
      "epoch": 1.951589011580932,
      "grad_norm": 0.8301241233047023,
      "learning_rate": 1.4489887057766439e-08,
      "loss": 0.30179758071899415,
      "num_input_tokens_seen": 923933984,
      "step": 28985,
      "train_runtime": 75890.3135,
      "train_tokens_per_second": 12174.597
    },
    {
      "epoch": 1.951925666576892,
      "grad_norm": 0.8885167147741369,
      "learning_rate": 1.428943298540597e-08,
      "loss": 0.28542351722717285,
      "num_input_tokens_seen": 924096184,
      "step": 28990,
      "train_runtime": 75903.5612,
      "train_tokens_per_second": 12174.609
    },
    {
      "epoch": 1.952262321572852,
      "grad_norm": 0.7594617444287598,
      "learning_rate": 1.4090373151107173e-08,
      "loss": 0.3091750144958496,
      "num_input_tokens_seen": 924259184,
      "step": 28995,
      "train_runtime": 75916.9871,
      "train_tokens_per_second": 12174.603
    },
    {
      "epoch": 1.9525989765688123,
      "grad_norm": 0.7757424716130054,
      "learning_rate": 1.3892707610536627e-08,
      "loss": 0.2546836853027344,
      "num_input_tokens_seen": 924416544,
      "step": 29000,
      "train_runtime": 75930.3122,
      "train_tokens_per_second": 12174.539
    },
    {
      "epoch": 1.9529356315647726,
      "grad_norm": 0.8035441875851526,
      "learning_rate": 1.3696436418970672e-08,
      "loss": 0.27744646072387696,
      "num_input_tokens_seen": 924573312,
      "step": 29005,
      "train_runtime": 75942.855,
      "train_tokens_per_second": 12174.592
    },
    {
      "epoch": 1.9532722865607326,
      "grad_norm": 0.8038054377745673,
      "learning_rate": 1.3501559631296512e-08,
      "loss": 0.29370999336242676,
      "num_input_tokens_seen": 924731136,
      "step": 29010,
      "train_runtime": 75956.3032,
      "train_tokens_per_second": 12174.515
    },
    {
      "epoch": 1.9536089415566926,
      "grad_norm": 0.7778274979718193,
      "learning_rate": 1.3308077302010557e-08,
      "loss": 0.31083633899688723,
      "num_input_tokens_seen": 924894376,
      "step": 29015,
      "train_runtime": 75969.1889,
      "train_tokens_per_second": 12174.599
    },
    {
      "epoch": 1.9539455965526529,
      "grad_norm": 0.8018929033041386,
      "learning_rate": 1.3115989485218972e-08,
      "loss": 0.290159010887146,
      "num_input_tokens_seen": 925048200,
      "step": 29020,
      "train_runtime": 75982.681,
      "train_tokens_per_second": 12174.461
    },
    {
      "epoch": 1.9542822515486131,
      "grad_norm": 1.1657468820625887,
      "learning_rate": 1.2925296234638784e-08,
      "loss": 0.27074203491210935,
      "num_input_tokens_seen": 925206936,
      "step": 29025,
      "train_runtime": 75996.0865,
      "train_tokens_per_second": 12174.402
    },
    {
      "epoch": 1.9546189065445732,
      "grad_norm": 0.9295764768955839,
      "learning_rate": 1.2735997603597339e-08,
      "loss": 0.28510513305664065,
      "num_input_tokens_seen": 925369616,
      "step": 29030,
      "train_runtime": 76008.7671,
      "train_tokens_per_second": 12174.512
    },
    {
      "epoch": 1.9549555615405332,
      "grad_norm": 0.8114093466436221,
      "learning_rate": 1.2548093645030623e-08,
      "loss": 0.2804151773452759,
      "num_input_tokens_seen": 925531952,
      "step": 29035,
      "train_runtime": 76021.3688,
      "train_tokens_per_second": 12174.629
    },
    {
      "epoch": 1.9552922165364934,
      "grad_norm": 0.823131232637778,
      "learning_rate": 1.236158441148605e-08,
      "loss": 0.28353567123413087,
      "num_input_tokens_seen": 925692544,
      "step": 29040,
      "train_runtime": 76034.2653,
      "train_tokens_per_second": 12174.676
    },
    {
      "epoch": 1.9556288715324537,
      "grad_norm": 0.8952614438662455,
      "learning_rate": 1.2176469955119119e-08,
      "loss": 0.32651076316833494,
      "num_input_tokens_seen": 925855240,
      "step": 29045,
      "train_runtime": 76046.9737,
      "train_tokens_per_second": 12174.781
    },
    {
      "epoch": 1.9559655265284137,
      "grad_norm": 0.8828414499503993,
      "learning_rate": 1.199275032769842e-08,
      "loss": 0.31638288497924805,
      "num_input_tokens_seen": 926015424,
      "step": 29050,
      "train_runtime": 76060.675,
      "train_tokens_per_second": 12174.694
    },
    {
      "epoch": 1.9563021815243737,
      "grad_norm": 0.8346491115460912,
      "learning_rate": 1.1810425580598971e-08,
      "loss": 0.3131925821304321,
      "num_input_tokens_seen": 926178040,
      "step": 29055,
      "train_runtime": 76073.344,
      "train_tokens_per_second": 12174.804
    },
    {
      "epoch": 1.956638836520334,
      "grad_norm": 0.9249118296844675,
      "learning_rate": 1.1629495764807763e-08,
      "loss": 0.3112865686416626,
      "num_input_tokens_seen": 926341880,
      "step": 29060,
      "train_runtime": 76085.7551,
      "train_tokens_per_second": 12174.971
    },
    {
      "epoch": 1.9569754915162942,
      "grad_norm": 0.903865299643591,
      "learning_rate": 1.1449960930921544e-08,
      "loss": 0.28887715339660647,
      "num_input_tokens_seen": 926500160,
      "step": 29065,
      "train_runtime": 76098.7958,
      "train_tokens_per_second": 12174.965
    },
    {
      "epoch": 1.9573121465122543,
      "grad_norm": 0.8445204757506805,
      "learning_rate": 1.1271821129146266e-08,
      "loss": 0.27671151161193847,
      "num_input_tokens_seen": 926660312,
      "step": 29070,
      "train_runtime": 76111.3016,
      "train_tokens_per_second": 12175.069
    },
    {
      "epoch": 1.9576488015082143,
      "grad_norm": 0.8774050856238764,
      "learning_rate": 1.1095076409298189e-08,
      "loss": 0.2734399318695068,
      "num_input_tokens_seen": 926815904,
      "step": 29075,
      "train_runtime": 76124.115,
      "train_tokens_per_second": 12175.063
    },
    {
      "epoch": 1.9579854565041745,
      "grad_norm": 0.8052245695910365,
      "learning_rate": 1.0919726820803889e-08,
      "loss": 0.2903740882873535,
      "num_input_tokens_seen": 926976136,
      "step": 29080,
      "train_runtime": 76137.3688,
      "train_tokens_per_second": 12175.048
    },
    {
      "epoch": 1.9583221115001348,
      "grad_norm": 0.7552641170358747,
      "learning_rate": 1.0745772412698585e-08,
      "loss": 0.2916219711303711,
      "num_input_tokens_seen": 927130296,
      "step": 29085,
      "train_runtime": 76150.3985,
      "train_tokens_per_second": 12174.989
    },
    {
      "epoch": 1.9586587664960948,
      "grad_norm": 0.886690885988374,
      "learning_rate": 1.057321323362892e-08,
      "loss": 0.29023659229278564,
      "num_input_tokens_seen": 927294136,
      "step": 29090,
      "train_runtime": 76162.8206,
      "train_tokens_per_second": 12175.155
    },
    {
      "epoch": 1.9589954214920549,
      "grad_norm": 0.8182399718478792,
      "learning_rate": 1.0402049331849629e-08,
      "loss": 0.2984285831451416,
      "num_input_tokens_seen": 927455264,
      "step": 29095,
      "train_runtime": 76176.857,
      "train_tokens_per_second": 12175.027
    },
    {
      "epoch": 1.959332076488015,
      "grad_norm": 0.7857156556666306,
      "learning_rate": 1.0232280755226865e-08,
      "loss": 0.3030875205993652,
      "num_input_tokens_seen": 927614112,
      "step": 29100,
      "train_runtime": 76190.4698,
      "train_tokens_per_second": 12174.936
    },
    {
      "epoch": 1.9596687314839754,
      "grad_norm": 0.8303205475206933,
      "learning_rate": 1.006390755123543e-08,
      "loss": 0.28156147003173826,
      "num_input_tokens_seen": 927774344,
      "step": 29105,
      "train_runtime": 76203.8038,
      "train_tokens_per_second": 12174.909
    },
    {
      "epoch": 1.9600053864799354,
      "grad_norm": 0.85512421200839,
      "learning_rate": 9.89692976696044e-09,
      "loss": 0.31151576042175294,
      "num_input_tokens_seen": 927931320,
      "step": 29110,
      "train_runtime": 76216.3109,
      "train_tokens_per_second": 12174.97
    },
    {
      "epoch": 1.9603420414758954,
      "grad_norm": 0.7459946251599527,
      "learning_rate": 9.731347449097316e-09,
      "loss": 0.30746827125549314,
      "num_input_tokens_seen": 928093392,
      "step": 29115,
      "train_runtime": 76229.9159,
      "train_tokens_per_second": 12174.923
    },
    {
      "epoch": 1.9606786964718557,
      "grad_norm": 0.8376215975248772,
      "learning_rate": 9.56716064394958e-09,
      "loss": 0.2999640703201294,
      "num_input_tokens_seen": 928256312,
      "step": 29120,
      "train_runtime": 76242.3376,
      "train_tokens_per_second": 12175.077
    },
    {
      "epoch": 1.961015351467816,
      "grad_norm": 0.9217112386948961,
      "learning_rate": 9.404369397432166e-09,
      "loss": 0.27633986473083494,
      "num_input_tokens_seen": 928412616,
      "step": 29125,
      "train_runtime": 76254.974,
      "train_tokens_per_second": 12175.109
    },
    {
      "epoch": 1.961352006463776,
      "grad_norm": 0.7845970165115492,
      "learning_rate": 9.242973755068663e-09,
      "loss": 0.28524107933044435,
      "num_input_tokens_seen": 928574528,
      "step": 29130,
      "train_runtime": 76267.515,
      "train_tokens_per_second": 12175.23
    },
    {
      "epoch": 1.961688661459736,
      "grad_norm": 0.8245440931407594,
      "learning_rate": 9.082973761993518e-09,
      "loss": 0.28589253425598143,
      "num_input_tokens_seen": 928731376,
      "step": 29135,
      "train_runtime": 76280.2296,
      "train_tokens_per_second": 12175.257
    },
    {
      "epoch": 1.9620253164556962,
      "grad_norm": 0.7917797055104286,
      "learning_rate": 8.924369462949834e-09,
      "loss": 0.32050471305847167,
      "num_input_tokens_seen": 928886440,
      "step": 29140,
      "train_runtime": 76293.4841,
      "train_tokens_per_second": 12175.174
    },
    {
      "epoch": 1.9623619714516565,
      "grad_norm": 0.7795201316001087,
      "learning_rate": 8.767160902291016e-09,
      "loss": 0.29182920455932615,
      "num_input_tokens_seen": 929047344,
      "step": 29145,
      "train_runtime": 76306.6584,
      "train_tokens_per_second": 12175.181
    },
    {
      "epoch": 1.9626986264476165,
      "grad_norm": 0.7527944642683888,
      "learning_rate": 8.61134812397968e-09,
      "loss": 0.2893436193466187,
      "num_input_tokens_seen": 929207368,
      "step": 29150,
      "train_runtime": 76320.0126,
      "train_tokens_per_second": 12175.147
    },
    {
      "epoch": 1.9630352814435765,
      "grad_norm": 0.792256109923035,
      "learning_rate": 8.45693117158819e-09,
      "loss": 0.2741499900817871,
      "num_input_tokens_seen": 929362520,
      "step": 29155,
      "train_runtime": 76333.2266,
      "train_tokens_per_second": 12175.072
    },
    {
      "epoch": 1.9633719364395368,
      "grad_norm": 0.8823355011819821,
      "learning_rate": 8.303910088299228e-09,
      "loss": 0.2860186815261841,
      "num_input_tokens_seen": 929522096,
      "step": 29160,
      "train_runtime": 76345.878,
      "train_tokens_per_second": 12175.144
    },
    {
      "epoch": 1.963708591435497,
      "grad_norm": 0.7999233681376358,
      "learning_rate": 8.152284916904674e-09,
      "loss": 0.3026837587356567,
      "num_input_tokens_seen": 929682912,
      "step": 29165,
      "train_runtime": 76358.3206,
      "train_tokens_per_second": 12175.267
    },
    {
      "epoch": 1.964045246431457,
      "grad_norm": 0.808755355800198,
      "learning_rate": 8.002055699805612e-09,
      "loss": 0.27841997146606445,
      "num_input_tokens_seen": 929842064,
      "step": 29170,
      "train_runtime": 76371.1149,
      "train_tokens_per_second": 12175.311
    },
    {
      "epoch": 1.964381901427417,
      "grad_norm": 0.7353375648412412,
      "learning_rate": 7.853222479013434e-09,
      "loss": 0.2710365056991577,
      "num_input_tokens_seen": 930005904,
      "step": 29175,
      "train_runtime": 76383.5289,
      "train_tokens_per_second": 12175.477
    },
    {
      "epoch": 1.9647185564233773,
      "grad_norm": 0.730225348719828,
      "learning_rate": 7.705785296148737e-09,
      "loss": 0.28692214488983153,
      "num_input_tokens_seen": 930165272,
      "step": 29180,
      "train_runtime": 76396.9893,
      "train_tokens_per_second": 12175.418
    },
    {
      "epoch": 1.9650552114193376,
      "grad_norm": 0.9179627967196287,
      "learning_rate": 7.55974419244132e-09,
      "loss": 0.31088976860046386,
      "num_input_tokens_seen": 930325152,
      "step": 29185,
      "train_runtime": 76410.3498,
      "train_tokens_per_second": 12175.381
    },
    {
      "epoch": 1.9653918664152976,
      "grad_norm": 0.7973398060217323,
      "learning_rate": 7.415099208732402e-09,
      "loss": 0.2749485492706299,
      "num_input_tokens_seen": 930480584,
      "step": 29190,
      "train_runtime": 76422.7714,
      "train_tokens_per_second": 12175.436
    },
    {
      "epoch": 1.9657285214112576,
      "grad_norm": 0.777671908082559,
      "learning_rate": 7.2718503854707355e-09,
      "loss": 0.29342141151428225,
      "num_input_tokens_seen": 930639232,
      "step": 29195,
      "train_runtime": 76435.5933,
      "train_tokens_per_second": 12175.469
    },
    {
      "epoch": 1.966065176407218,
      "grad_norm": 0.7634360715449815,
      "learning_rate": 7.129997762715391e-09,
      "loss": 0.2619804859161377,
      "num_input_tokens_seen": 930797192,
      "step": 29200,
      "train_runtime": 76448.5183,
      "train_tokens_per_second": 12175.477
    },
    {
      "epoch": 1.9664018314031781,
      "grad_norm": 0.8170596591393084,
      "learning_rate": 6.98954138013519e-09,
      "loss": 0.2897006034851074,
      "num_input_tokens_seen": 930955752,
      "step": 29205,
      "train_runtime": 76461.4633,
      "train_tokens_per_second": 12175.49
    },
    {
      "epoch": 1.9667384863991382,
      "grad_norm": 0.8224837112087526,
      "learning_rate": 6.850481277008159e-09,
      "loss": 0.3011942863464355,
      "num_input_tokens_seen": 931118392,
      "step": 29210,
      "train_runtime": 76474.5278,
      "train_tokens_per_second": 12175.536
    },
    {
      "epoch": 1.9670751413950982,
      "grad_norm": 0.8590772388903853,
      "learning_rate": 6.712817492222079e-09,
      "loss": 0.3154435157775879,
      "num_input_tokens_seen": 931276200,
      "step": 29215,
      "train_runtime": 76486.9791,
      "train_tokens_per_second": 12175.617
    },
    {
      "epoch": 1.9674117963910585,
      "grad_norm": 0.7370067370233849,
      "learning_rate": 6.576550064273934e-09,
      "loss": 0.290617847442627,
      "num_input_tokens_seen": 931433344,
      "step": 29220,
      "train_runtime": 76500.118,
      "train_tokens_per_second": 12175.58
    },
    {
      "epoch": 1.9677484513870187,
      "grad_norm": 0.8349058841392281,
      "learning_rate": 6.44167903127102e-09,
      "loss": 0.3119671821594238,
      "num_input_tokens_seen": 931594304,
      "step": 29225,
      "train_runtime": 76513.0002,
      "train_tokens_per_second": 12175.634
    },
    {
      "epoch": 1.9680851063829787,
      "grad_norm": 0.8873494314559613,
      "learning_rate": 6.3082044309287215e-09,
      "loss": 0.28221855163574217,
      "num_input_tokens_seen": 931756176,
      "step": 29230,
      "train_runtime": 76526.0816,
      "train_tokens_per_second": 12175.668
    },
    {
      "epoch": 1.9684217613789388,
      "grad_norm": 0.7763560275554647,
      "learning_rate": 6.176126300573848e-09,
      "loss": 0.292108154296875,
      "num_input_tokens_seen": 931916984,
      "step": 29235,
      "train_runtime": 76538.835,
      "train_tokens_per_second": 12175.74
    },
    {
      "epoch": 1.968758416374899,
      "grad_norm": 0.858610586630292,
      "learning_rate": 6.045444677140744e-09,
      "loss": 0.27784552574157717,
      "num_input_tokens_seen": 932079816,
      "step": 29240,
      "train_runtime": 76551.5814,
      "train_tokens_per_second": 12175.84
    },
    {
      "epoch": 1.9690950713708593,
      "grad_norm": 0.8491615942184385,
      "learning_rate": 5.91615959717462e-09,
      "loss": 0.28981566429138184,
      "num_input_tokens_seen": 932237248,
      "step": 29245,
      "train_runtime": 76564.0612,
      "train_tokens_per_second": 12175.912
    },
    {
      "epoch": 1.9694317263668193,
      "grad_norm": 0.6818760385053637,
      "learning_rate": 5.78827109682989e-09,
      "loss": 0.24991152286529542,
      "num_input_tokens_seen": 932398016,
      "step": 29250,
      "train_runtime": 76577.1815,
      "train_tokens_per_second": 12175.925
    },
    {
      "epoch": 1.9697683813627793,
      "grad_norm": 0.8100713265658934,
      "learning_rate": 5.661779211869056e-09,
      "loss": 0.2760288238525391,
      "num_input_tokens_seen": 932559336,
      "step": 29255,
      "train_runtime": 76590.4102,
      "train_tokens_per_second": 12175.928
    },
    {
      "epoch": 1.9701050363587396,
      "grad_norm": 0.8854628387437574,
      "learning_rate": 5.536683977666601e-09,
      "loss": 0.2981565475463867,
      "num_input_tokens_seen": 932723096,
      "step": 29260,
      "train_runtime": 76603.5043,
      "train_tokens_per_second": 12175.985
    },
    {
      "epoch": 1.9704416913546998,
      "grad_norm": 0.7458527606708117,
      "learning_rate": 5.412985429203988e-09,
      "loss": 0.2862668514251709,
      "num_input_tokens_seen": 932879296,
      "step": 29265,
      "train_runtime": 76616.3147,
      "train_tokens_per_second": 12175.988
    },
    {
      "epoch": 1.9707783463506598,
      "grad_norm": 0.8295145138160456,
      "learning_rate": 5.290683601073543e-09,
      "loss": 0.28243634700775144,
      "num_input_tokens_seen": 933039944,
      "step": 29270,
      "train_runtime": 76629.7985,
      "train_tokens_per_second": 12175.942
    },
    {
      "epoch": 1.9711150013466199,
      "grad_norm": 0.8003242943795617,
      "learning_rate": 5.169778527476243e-09,
      "loss": 0.2951035022735596,
      "num_input_tokens_seen": 933201704,
      "step": 29275,
      "train_runtime": 76642.7107,
      "train_tokens_per_second": 12176.001
    },
    {
      "epoch": 1.9714516563425801,
      "grad_norm": 0.8411573280682593,
      "learning_rate": 5.050270242222821e-09,
      "loss": 0.30519585609436034,
      "num_input_tokens_seen": 933355320,
      "step": 29280,
      "train_runtime": 76656.117,
      "train_tokens_per_second": 12175.875
    },
    {
      "epoch": 1.9717883113385404,
      "grad_norm": 0.7972038728645241,
      "learning_rate": 4.932158778733765e-09,
      "loss": 0.2804659128189087,
      "num_input_tokens_seen": 933515528,
      "step": 29285,
      "train_runtime": 76668.8967,
      "train_tokens_per_second": 12175.935
    },
    {
      "epoch": 1.9721249663345004,
      "grad_norm": 0.8866508167217122,
      "learning_rate": 4.8154441700387636e-09,
      "loss": 0.3104846715927124,
      "num_input_tokens_seen": 933666640,
      "step": 29290,
      "train_runtime": 76682.5124,
      "train_tokens_per_second": 12175.744
    },
    {
      "epoch": 1.9724616213304604,
      "grad_norm": 0.8016374092314424,
      "learning_rate": 4.7001264487761545e-09,
      "loss": 0.3072397232055664,
      "num_input_tokens_seen": 933827072,
      "step": 29295,
      "train_runtime": 76694.9164,
      "train_tokens_per_second": 12175.867
    },
    {
      "epoch": 1.9727982763264207,
      "grad_norm": 0.7521807178697871,
      "learning_rate": 4.586205647194031e-09,
      "loss": 0.3076127529144287,
      "num_input_tokens_seen": 933988224,
      "step": 29300,
      "train_runtime": 76708.8358,
      "train_tokens_per_second": 12175.758
    },
    {
      "epoch": 1.973134931322381,
      "grad_norm": 0.9051066643736196,
      "learning_rate": 4.4736817971507974e-09,
      "loss": 0.2979921817779541,
      "num_input_tokens_seen": 934146248,
      "step": 29305,
      "train_runtime": 76721.9656,
      "train_tokens_per_second": 12175.734
    },
    {
      "epoch": 1.973471586318341,
      "grad_norm": 0.8034825681678837,
      "learning_rate": 4.362554930112395e-09,
      "loss": 0.2878642797470093,
      "num_input_tokens_seen": 934301736,
      "step": 29310,
      "train_runtime": 76735.3903,
      "train_tokens_per_second": 12175.63
    },
    {
      "epoch": 1.973808241314301,
      "grad_norm": 0.8420632395335224,
      "learning_rate": 4.252825077155631e-09,
      "loss": 0.2979853630065918,
      "num_input_tokens_seen": 934461944,
      "step": 29315,
      "train_runtime": 76748.8677,
      "train_tokens_per_second": 12175.58
    },
    {
      "epoch": 1.9741448963102612,
      "grad_norm": 0.8675204154127251,
      "learning_rate": 4.1444922689665156e-09,
      "loss": 0.2956474781036377,
      "num_input_tokens_seen": 934620336,
      "step": 29320,
      "train_runtime": 76762.5477,
      "train_tokens_per_second": 12175.473
    },
    {
      "epoch": 1.9744815513062215,
      "grad_norm": 0.86023449238149,
      "learning_rate": 4.037556535839149e-09,
      "loss": 0.26177656650543213,
      "num_input_tokens_seen": 934779232,
      "step": 29325,
      "train_runtime": 76775.4866,
      "train_tokens_per_second": 12175.491
    },
    {
      "epoch": 1.9748182063021815,
      "grad_norm": 0.73820683149687,
      "learning_rate": 3.932017907677943e-09,
      "loss": 0.2710430145263672,
      "num_input_tokens_seen": 934942280,
      "step": 29330,
      "train_runtime": 76788.3128,
      "train_tokens_per_second": 12175.58
    },
    {
      "epoch": 1.9751548612981416,
      "grad_norm": 0.7840797682346614,
      "learning_rate": 3.827876413997067e-09,
      "loss": 0.29810845851898193,
      "num_input_tokens_seen": 935094576,
      "step": 29335,
      "train_runtime": 76802.4877,
      "train_tokens_per_second": 12175.316
    },
    {
      "epoch": 1.9754915162941018,
      "grad_norm": 0.8375281112371067,
      "learning_rate": 3.725132083918781e-09,
      "loss": 0.2911087989807129,
      "num_input_tokens_seen": 935252952,
      "step": 29340,
      "train_runtime": 76816.3797,
      "train_tokens_per_second": 12175.176
    },
    {
      "epoch": 1.975828171290062,
      "grad_norm": 0.7657521510165652,
      "learning_rate": 3.623784946175102e-09,
      "loss": 0.29820995330810546,
      "num_input_tokens_seen": 935407608,
      "step": 29345,
      "train_runtime": 76829.5433,
      "train_tokens_per_second": 12175.103
    },
    {
      "epoch": 1.976164826286022,
      "grad_norm": 0.8104504671700982,
      "learning_rate": 3.5238350291083577e-09,
      "loss": 0.275114917755127,
      "num_input_tokens_seen": 935561536,
      "step": 29350,
      "train_runtime": 76843.2615,
      "train_tokens_per_second": 12174.933
    },
    {
      "epoch": 1.976501481281982,
      "grad_norm": 0.806214156225629,
      "learning_rate": 3.4252823606678588e-09,
      "loss": 0.3068329572677612,
      "num_input_tokens_seen": 935722776,
      "step": 29355,
      "train_runtime": 76856.2485,
      "train_tokens_per_second": 12174.973
    },
    {
      "epoch": 1.9768381362779424,
      "grad_norm": 0.8617761238487917,
      "learning_rate": 3.328126968414336e-09,
      "loss": 0.311933708190918,
      "num_input_tokens_seen": 935876760,
      "step": 29360,
      "train_runtime": 76869.7495,
      "train_tokens_per_second": 12174.838
    },
    {
      "epoch": 1.9771747912739026,
      "grad_norm": 0.7910248620353353,
      "learning_rate": 3.232368879517167e-09,
      "loss": 0.30498623847961426,
      "num_input_tokens_seen": 936039376,
      "step": 29365,
      "train_runtime": 76883.1192,
      "train_tokens_per_second": 12174.836
    },
    {
      "epoch": 1.9775114462698626,
      "grad_norm": 0.8446123317850052,
      "learning_rate": 3.1380081207543766e-09,
      "loss": 0.2721656560897827,
      "num_input_tokens_seen": 936198504,
      "step": 29370,
      "train_runtime": 76896.8241,
      "train_tokens_per_second": 12174.736
    },
    {
      "epoch": 1.9778481012658227,
      "grad_norm": 0.8337056911954362,
      "learning_rate": 3.0450447185137454e-09,
      "loss": 0.3107349634170532,
      "num_input_tokens_seen": 936358888,
      "step": 29375,
      "train_runtime": 76910.1774,
      "train_tokens_per_second": 12174.707
    },
    {
      "epoch": 1.978184756261783,
      "grad_norm": 0.8228387935666606,
      "learning_rate": 2.953478698792256e-09,
      "loss": 0.2757991552352905,
      "num_input_tokens_seen": 936521040,
      "step": 29380,
      "train_runtime": 76922.7544,
      "train_tokens_per_second": 12174.825
    },
    {
      "epoch": 1.9785214112577432,
      "grad_norm": 0.8213008480511698,
      "learning_rate": 2.8633100871960917e-09,
      "loss": 0.27904744148254396,
      "num_input_tokens_seen": 936674440,
      "step": 29385,
      "train_runtime": 76935.6325,
      "train_tokens_per_second": 12174.781
    },
    {
      "epoch": 1.9788580662537032,
      "grad_norm": 0.8761024589761595,
      "learning_rate": 2.774538908940638e-09,
      "loss": 0.2947150945663452,
      "num_input_tokens_seen": 936830064,
      "step": 29390,
      "train_runtime": 76948.5595,
      "train_tokens_per_second": 12174.758
    },
    {
      "epoch": 1.9791947212496632,
      "grad_norm": 0.7589192973955758,
      "learning_rate": 2.687165188849927e-09,
      "loss": 0.28190290927886963,
      "num_input_tokens_seen": 936992840,
      "step": 29395,
      "train_runtime": 76962.0049,
      "train_tokens_per_second": 12174.746
    },
    {
      "epoch": 1.9795313762456235,
      "grad_norm": 0.8593300926575295,
      "learning_rate": 2.6011889513588574e-09,
      "loss": 0.2864360332489014,
      "num_input_tokens_seen": 937151104,
      "step": 29400,
      "train_runtime": 76975.414,
      "train_tokens_per_second": 12174.681
    },
    {
      "epoch": 1.9798680312415837,
      "grad_norm": 0.7991317606658112,
      "learning_rate": 2.5166102205093077e-09,
      "loss": 0.3007565975189209,
      "num_input_tokens_seen": 937310744,
      "step": 29405,
      "train_runtime": 76988.4696,
      "train_tokens_per_second": 12174.69
    },
    {
      "epoch": 1.9802046862375438,
      "grad_norm": 0.8902602754945289,
      "learning_rate": 2.4334290199540254e-09,
      "loss": 0.28476266860961913,
      "num_input_tokens_seen": 937467656,
      "step": 29410,
      "train_runtime": 77000.8973,
      "train_tokens_per_second": 12174.763
    },
    {
      "epoch": 1.9805413412335038,
      "grad_norm": 0.8224627931710313,
      "learning_rate": 2.351645372953848e-09,
      "loss": 0.2673612594604492,
      "num_input_tokens_seen": 937628192,
      "step": 29415,
      "train_runtime": 77013.9117,
      "train_tokens_per_second": 12174.79
    },
    {
      "epoch": 1.980877996229464,
      "grad_norm": 0.901435471279115,
      "learning_rate": 2.2712593023804795e-09,
      "loss": 0.2813266277313232,
      "num_input_tokens_seen": 937781104,
      "step": 29420,
      "train_runtime": 77026.6802,
      "train_tokens_per_second": 12174.757
    },
    {
      "epoch": 1.9812146512254243,
      "grad_norm": 0.7871345186400073,
      "learning_rate": 2.1922708307120512e-09,
      "loss": 0.2950688362121582,
      "num_input_tokens_seen": 937942752,
      "step": 29425,
      "train_runtime": 77039.8264,
      "train_tokens_per_second": 12174.778
    },
    {
      "epoch": 1.9815513062213843,
      "grad_norm": 0.7976326292391558,
      "learning_rate": 2.1146799800386698e-09,
      "loss": 0.30443737506866453,
      "num_input_tokens_seen": 938100608,
      "step": 29430,
      "train_runtime": 77052.8194,
      "train_tokens_per_second": 12174.773
    },
    {
      "epoch": 1.9818879612173443,
      "grad_norm": 0.848508194537515,
      "learning_rate": 2.0384867720579794e-09,
      "loss": 0.27435779571533203,
      "num_input_tokens_seen": 938256584,
      "step": 29435,
      "train_runtime": 77066.0985,
      "train_tokens_per_second": 12174.699
    },
    {
      "epoch": 1.9822246162133046,
      "grad_norm": 0.9298469051325583,
      "learning_rate": 1.96369122807738e-09,
      "loss": 0.3082582473754883,
      "num_input_tokens_seen": 938417288,
      "step": 29440,
      "train_runtime": 77079.1281,
      "train_tokens_per_second": 12174.726
    },
    {
      "epoch": 1.9825612712092648,
      "grad_norm": 0.8481077027374472,
      "learning_rate": 1.8902933690123637e-09,
      "loss": 0.2730572700500488,
      "num_input_tokens_seen": 938575840,
      "step": 29445,
      "train_runtime": 77092.1315,
      "train_tokens_per_second": 12174.729
    },
    {
      "epoch": 1.9828979262052249,
      "grad_norm": 0.8010328499854671,
      "learning_rate": 1.8182932153892884e-09,
      "loss": 0.29535622596740724,
      "num_input_tokens_seen": 938736072,
      "step": 29450,
      "train_runtime": 77104.8565,
      "train_tokens_per_second": 12174.798
    },
    {
      "epoch": 1.983234581201185,
      "grad_norm": 0.8489537335074749,
      "learning_rate": 1.7476907873426041e-09,
      "loss": 0.27869572639465334,
      "num_input_tokens_seen": 938895864,
      "step": 29455,
      "train_runtime": 77117.3055,
      "train_tokens_per_second": 12174.905
    },
    {
      "epoch": 1.9835712361971451,
      "grad_norm": 0.7983175571632403,
      "learning_rate": 1.6784861046159617e-09,
      "loss": 0.2912956714630127,
      "num_input_tokens_seen": 939055744,
      "step": 29460,
      "train_runtime": 77130.6823,
      "train_tokens_per_second": 12174.866
    },
    {
      "epoch": 1.9839078911931054,
      "grad_norm": 0.7598396639412233,
      "learning_rate": 1.610679186561659e-09,
      "loss": 0.29097628593444824,
      "num_input_tokens_seen": 939219408,
      "step": 29465,
      "train_runtime": 77144.3712,
      "train_tokens_per_second": 12174.827
    },
    {
      "epoch": 1.9842445461890654,
      "grad_norm": 0.7172928944345176,
      "learning_rate": 1.5442700521428599e-09,
      "loss": 0.2674097061157227,
      "num_input_tokens_seen": 939377952,
      "step": 29470,
      "train_runtime": 77157.5723,
      "train_tokens_per_second": 12174.799
    },
    {
      "epoch": 1.9845812011850255,
      "grad_norm": 0.819529932769271,
      "learning_rate": 1.4792587199297103e-09,
      "loss": 0.28331961631774905,
      "num_input_tokens_seen": 939538568,
      "step": 29475,
      "train_runtime": 77170.1586,
      "train_tokens_per_second": 12174.895
    },
    {
      "epoch": 1.9849178561809857,
      "grad_norm": 0.8639167652586511,
      "learning_rate": 1.4156452081026672e-09,
      "loss": 0.29461016654968264,
      "num_input_tokens_seen": 939699048,
      "step": 29480,
      "train_runtime": 77182.5833,
      "train_tokens_per_second": 12175.014
    },
    {
      "epoch": 1.985254511176946,
      "grad_norm": 0.9108738154013134,
      "learning_rate": 1.3534295344513893e-09,
      "loss": 0.2977909088134766,
      "num_input_tokens_seen": 939861112,
      "step": 29485,
      "train_runtime": 77195.2025,
      "train_tokens_per_second": 12175.123
    },
    {
      "epoch": 1.985591166172906,
      "grad_norm": 0.9389041567530285,
      "learning_rate": 1.2926117163741813e-09,
      "loss": 0.27971372604370115,
      "num_input_tokens_seen": 940023104,
      "step": 29490,
      "train_runtime": 77207.7548,
      "train_tokens_per_second": 12175.242
    },
    {
      "epoch": 1.985927821168866,
      "grad_norm": 0.7918353716275476,
      "learning_rate": 1.233191770877995e-09,
      "loss": 0.30565314292907714,
      "num_input_tokens_seen": 940186016,
      "step": 29495,
      "train_runtime": 77221.2233,
      "train_tokens_per_second": 12175.228
    },
    {
      "epoch": 1.9862644761648263,
      "grad_norm": 0.8468452486759673,
      "learning_rate": 1.1751697145800934e-09,
      "loss": 0.2905928134918213,
      "num_input_tokens_seen": 940348768,
      "step": 29500,
      "train_runtime": 77233.8541,
      "train_tokens_per_second": 12175.344
    },
    {
      "epoch": 1.9866011311607865,
      "grad_norm": 0.8378997663580825,
      "learning_rate": 1.1185455637052756e-09,
      "loss": 0.33621797561645506,
      "num_input_tokens_seen": 940509424,
      "step": 29505,
      "train_runtime": 77247.1311,
      "train_tokens_per_second": 12175.331
    },
    {
      "epoch": 1.9869377861567465,
      "grad_norm": 0.8968025348299586,
      "learning_rate": 1.063319334089763e-09,
      "loss": 0.3021702289581299,
      "num_input_tokens_seen": 940665144,
      "step": 29510,
      "train_runtime": 77259.8774,
      "train_tokens_per_second": 12175.338
    },
    {
      "epoch": 1.9872744411527066,
      "grad_norm": 0.7727664332513424,
      "learning_rate": 1.0094910411762027e-09,
      "loss": 0.2760730266571045,
      "num_input_tokens_seen": 940823912,
      "step": 29515,
      "train_runtime": 77272.8938,
      "train_tokens_per_second": 12175.342
    },
    {
      "epoch": 1.9876110961486668,
      "grad_norm": 0.864133591980906,
      "learning_rate": 9.570607000175536e-10,
      "loss": 0.30960218906402587,
      "num_input_tokens_seen": 940985400,
      "step": 29520,
      "train_runtime": 77286.504,
      "train_tokens_per_second": 12175.287
    },
    {
      "epoch": 1.987947751144627,
      "grad_norm": 0.899554970278401,
      "learning_rate": 9.060283252765311e-10,
      "loss": 0.2835289716720581,
      "num_input_tokens_seen": 941140752,
      "step": 29525,
      "train_runtime": 77299.5955,
      "train_tokens_per_second": 12175.235
    },
    {
      "epoch": 1.988284406140587,
      "grad_norm": 0.8206640583717952,
      "learning_rate": 8.563939312233871e-10,
      "loss": 0.2835988521575928,
      "num_input_tokens_seen": 941295728,
      "step": 29530,
      "train_runtime": 77312.1621,
      "train_tokens_per_second": 12175.261
    },
    {
      "epoch": 1.9886210611365471,
      "grad_norm": 0.848395996343161,
      "learning_rate": 8.081575317386847e-10,
      "loss": 0.309305477142334,
      "num_input_tokens_seen": 941458128,
      "step": 29535,
      "train_runtime": 77325.7283,
      "train_tokens_per_second": 12175.225
    },
    {
      "epoch": 1.9889577161325074,
      "grad_norm": 0.8074536469579221,
      "learning_rate": 7.613191403116338e-10,
      "loss": 0.31104748249053954,
      "num_input_tokens_seen": 941619424,
      "step": 29540,
      "train_runtime": 77338.4672,
      "train_tokens_per_second": 12175.305
    },
    {
      "epoch": 1.9892943711284676,
      "grad_norm": 0.8528998101186868,
      "learning_rate": 7.158787700400905e-10,
      "loss": 0.3015526056289673,
      "num_input_tokens_seen": 941782352,
      "step": 29545,
      "train_runtime": 77351.927,
      "train_tokens_per_second": 12175.293
    },
    {
      "epoch": 1.9896310261244277,
      "grad_norm": 0.7491203828873338,
      "learning_rate": 6.718364336316674e-10,
      "loss": 0.28328001499176025,
      "num_input_tokens_seen": 941943128,
      "step": 29550,
      "train_runtime": 77364.9824,
      "train_tokens_per_second": 12175.316
    },
    {
      "epoch": 1.9899676811203877,
      "grad_norm": 0.7587661311854949,
      "learning_rate": 6.291921434026238e-10,
      "loss": 0.27706184387207033,
      "num_input_tokens_seen": 942105032,
      "step": 29555,
      "train_runtime": 77377.5266,
      "train_tokens_per_second": 12175.435
    },
    {
      "epoch": 1.990304336116348,
      "grad_norm": 0.7278038598180633,
      "learning_rate": 5.879459112778651e-10,
      "loss": 0.26283950805664064,
      "num_input_tokens_seen": 942266520,
      "step": 29560,
      "train_runtime": 77391.3468,
      "train_tokens_per_second": 12175.347
    },
    {
      "epoch": 1.9906409911123082,
      "grad_norm": 0.7474019015144875,
      "learning_rate": 5.480977487920536e-10,
      "loss": 0.29887566566467283,
      "num_input_tokens_seen": 942430360,
      "step": 29565,
      "train_runtime": 77403.7559,
      "train_tokens_per_second": 12175.512
    },
    {
      "epoch": 1.9909776461082682,
      "grad_norm": 0.7540203833863748,
      "learning_rate": 5.096476670890526e-10,
      "loss": 0.29635598659515383,
      "num_input_tokens_seen": 942594200,
      "step": 29570,
      "train_runtime": 77416.1724,
      "train_tokens_per_second": 12175.676
    },
    {
      "epoch": 1.9913143011042282,
      "grad_norm": 0.8376997866780798,
      "learning_rate": 4.725956769202622e-10,
      "loss": 0.3023070812225342,
      "num_input_tokens_seen": 942750888,
      "step": 29575,
      "train_runtime": 77429.0143,
      "train_tokens_per_second": 12175.68
    },
    {
      "epoch": 1.9916509561001885,
      "grad_norm": 0.7409331766308338,
      "learning_rate": 4.369417886485039e-10,
      "loss": 0.2884089946746826,
      "num_input_tokens_seen": 942909592,
      "step": 29580,
      "train_runtime": 77442.9272,
      "train_tokens_per_second": 12175.542
    },
    {
      "epoch": 1.9919876110961487,
      "grad_norm": 0.8010841721462845,
      "learning_rate": 4.026860122430254e-10,
      "loss": 0.2792597770690918,
      "num_input_tokens_seen": 943070080,
      "step": 29585,
      "train_runtime": 77456.2648,
      "train_tokens_per_second": 12175.517
    },
    {
      "epoch": 1.9923242660921088,
      "grad_norm": 0.8653515440495158,
      "learning_rate": 3.698283572839412e-10,
      "loss": 0.29066541194915774,
      "num_input_tokens_seen": 943230024,
      "step": 29590,
      "train_runtime": 77469.2872,
      "train_tokens_per_second": 12175.535
    },
    {
      "epoch": 1.9926609210880688,
      "grad_norm": 1.0419336259041754,
      "learning_rate": 3.38368832959457e-10,
      "loss": 0.30399882793426514,
      "num_input_tokens_seen": 943383728,
      "step": 29595,
      "train_runtime": 77481.8642,
      "train_tokens_per_second": 12175.542
    },
    {
      "epoch": 1.992997576084029,
      "grad_norm": 0.8656319607881853,
      "learning_rate": 3.0830744806753523e-10,
      "loss": 0.3008615016937256,
      "num_input_tokens_seen": 943541952,
      "step": 29600,
      "train_runtime": 77494.3808,
      "train_tokens_per_second": 12175.618
    },
    {
      "epoch": 1.9933342310799893,
      "grad_norm": 0.8429220436900494,
      "learning_rate": 2.7964421101478457e-10,
      "loss": 0.2879644393920898,
      "num_input_tokens_seen": 943697008,
      "step": 29605,
      "train_runtime": 77508.4928,
      "train_tokens_per_second": 12175.401
    },
    {
      "epoch": 1.9936708860759493,
      "grad_norm": 0.790454005331594,
      "learning_rate": 2.5237912981646016e-10,
      "loss": 0.31576874256134035,
      "num_input_tokens_seen": 943857608,
      "step": 29610,
      "train_runtime": 77521.4136,
      "train_tokens_per_second": 12175.444
    },
    {
      "epoch": 1.9940075410719094,
      "grad_norm": 0.8755205578226211,
      "learning_rate": 2.265122120975738e-10,
      "loss": 0.28240337371826174,
      "num_input_tokens_seen": 944015136,
      "step": 29615,
      "train_runtime": 77534.6057,
      "train_tokens_per_second": 12175.404
    },
    {
      "epoch": 1.9943441960678696,
      "grad_norm": 0.7413368540241816,
      "learning_rate": 2.020434650906733e-10,
      "loss": 0.28460862636566164,
      "num_input_tokens_seen": 944176456,
      "step": 29620,
      "train_runtime": 77547.1945,
      "train_tokens_per_second": 12175.508
    },
    {
      "epoch": 1.9946808510638299,
      "grad_norm": 0.8225014042874448,
      "learning_rate": 1.7897289563972854e-10,
      "loss": 0.2820748805999756,
      "num_input_tokens_seen": 944339848,
      "step": 29625,
      "train_runtime": 77560.1349,
      "train_tokens_per_second": 12175.583
    },
    {
      "epoch": 1.99501750605979,
      "grad_norm": 0.8050614746202819,
      "learning_rate": 1.5730051019569037e-10,
      "loss": 0.27139968872070314,
      "num_input_tokens_seen": 944499152,
      "step": 29630,
      "train_runtime": 77573.6502,
      "train_tokens_per_second": 12175.515
    },
    {
      "epoch": 1.99535416105575,
      "grad_norm": 0.7561090790771114,
      "learning_rate": 1.3702631481926633e-10,
      "loss": 0.28660154342651367,
      "num_input_tokens_seen": 944659792,
      "step": 29635,
      "train_runtime": 77586.6032,
      "train_tokens_per_second": 12175.553
    },
    {
      "epoch": 1.9956908160517102,
      "grad_norm": 0.7728508250728632,
      "learning_rate": 1.1815031517981023e-10,
      "loss": 0.2818227529525757,
      "num_input_tokens_seen": 944815192,
      "step": 29640,
      "train_runtime": 77599.9837,
      "train_tokens_per_second": 12175.456
    },
    {
      "epoch": 1.9960274710476704,
      "grad_norm": 0.8242452009121897,
      "learning_rate": 1.0067251655643262e-10,
      "loss": 0.28234565258026123,
      "num_input_tokens_seen": 944978864,
      "step": 29645,
      "train_runtime": 77612.4105,
      "train_tokens_per_second": 12175.615
    },
    {
      "epoch": 1.9963641260436304,
      "grad_norm": 0.8919854040581632,
      "learning_rate": 8.459292383633522e-11,
      "loss": 0.29870238304138186,
      "num_input_tokens_seen": 945135840,
      "step": 29650,
      "train_runtime": 77626.3508,
      "train_tokens_per_second": 12175.451
    },
    {
      "epoch": 1.9967007810395905,
      "grad_norm": 0.7828172510925471,
      "learning_rate": 6.991154151592128e-11,
      "loss": 0.27906227111816406,
      "num_input_tokens_seen": 945295128,
      "step": 29655,
      "train_runtime": 77638.9734,
      "train_tokens_per_second": 12175.523
    },
    {
      "epoch": 1.9970374360355507,
      "grad_norm": 0.7833960276821675,
      "learning_rate": 5.662837370135066e-11,
      "loss": 0.30341286659240724,
      "num_input_tokens_seen": 945455848,
      "step": 29660,
      "train_runtime": 77651.4297,
      "train_tokens_per_second": 12175.64
    },
    {
      "epoch": 1.997374091031511,
      "grad_norm": 0.818635008156073,
      "learning_rate": 4.47434241074296e-11,
      "loss": 0.29509916305541994,
      "num_input_tokens_seen": 945612864,
      "step": 29665,
      "train_runtime": 77664.7853,
      "train_tokens_per_second": 12175.568
    },
    {
      "epoch": 1.997710746027471,
      "grad_norm": 0.9822177117892062,
      "learning_rate": 3.4256696057055615e-11,
      "loss": 0.29454774856567384,
      "num_input_tokens_seen": 945769840,
      "step": 29670,
      "train_runtime": 77678.4914,
      "train_tokens_per_second": 12175.44
    },
    {
      "epoch": 1.998047401023431,
      "grad_norm": 0.8269150531407821,
      "learning_rate": 2.5168192482882824e-11,
      "loss": 0.30178217887878417,
      "num_input_tokens_seen": 945931296,
      "step": 29675,
      "train_runtime": 77691.6155,
      "train_tokens_per_second": 12175.462
    },
    {
      "epoch": 1.9983840560193913,
      "grad_norm": 0.847094917279186,
      "learning_rate": 1.747791592676684e-11,
      "loss": 0.26887195110321044,
      "num_input_tokens_seen": 946095136,
      "step": 29680,
      "train_runtime": 77704.05,
      "train_tokens_per_second": 12175.622
    },
    {
      "epoch": 1.9987207110153515,
      "grad_norm": 0.7990417245340667,
      "learning_rate": 1.1185868539209665e-11,
      "loss": 0.28517296314239504,
      "num_input_tokens_seen": 946255432,
      "step": 29685,
      "train_runtime": 77718.7118,
      "train_tokens_per_second": 12175.388
    },
    {
      "epoch": 1.9990573660113116,
      "grad_norm": 0.8777814215437981,
      "learning_rate": 6.29205207991479e-12,
      "loss": 0.2979507207870483,
      "num_input_tokens_seen": 946409480,
      "step": 29690,
      "train_runtime": 77732.7486,
      "train_tokens_per_second": 12175.171
    },
    {
      "epoch": 1.9993940210072716,
      "grad_norm": 0.790297442863995,
      "learning_rate": 2.7964679166769815e-12,
      "loss": 0.29178709983825685,
      "num_input_tokens_seen": 946571528,
      "step": 29695,
      "train_runtime": 77745.2748,
      "train_tokens_per_second": 12175.293
    },
    {
      "epoch": 1.9997306760032318,
      "grad_norm": 0.9216521251008282,
      "learning_rate": 6.991170281578364e-13,
      "loss": 0.2546474695205688,
      "num_input_tokens_seen": 946724720,
      "step": 29700,
      "train_runtime": 77758.7355,
      "train_tokens_per_second": 12175.156
    },
    {
      "epoch": 2.0,
      "num_input_tokens_seen": 946850504,
      "step": 29704,
      "total_flos": 1751052299468800.0,
      "train_loss": 0.34826657584565623,
      "train_runtime": 77931.6678,
      "train_samples_per_second": 1.525,
      "train_steps_per_second": 0.381
    }
  ],
  "logging_steps": 5,
  "max_steps": 29704,
  "num_input_tokens_seen": 946850504,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1751052299468800.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}