{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 198,
  "global_step": 1980,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0010108668182966893,
      "grad_norm": 8.506893157958984,
      "learning_rate": 0.0,
      "loss": 0.2525153160095215,
      "num_input_tokens_seen": 2014,
      "step": 1,
      "train_runtime": 8.1467,
      "train_tokens_per_second": 247.217
    },
    {
      "epoch": 0.0020217336365933787,
      "grad_norm": 7.676785469055176,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.2333833873271942,
      "num_input_tokens_seen": 4302,
      "step": 2,
      "train_runtime": 10.5575,
      "train_tokens_per_second": 407.482
    },
    {
      "epoch": 0.003032600454890068,
      "grad_norm": 5.56013298034668,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.20100091397762299,
      "num_input_tokens_seen": 6256,
      "step": 3,
      "train_runtime": 12.8851,
      "train_tokens_per_second": 485.524
    },
    {
      "epoch": 0.004043467273186757,
      "grad_norm": 3.537539005279541,
      "learning_rate": 1.2e-05,
      "loss": 0.12631550431251526,
      "num_input_tokens_seen": 8140,
      "step": 4,
      "train_runtime": 15.1509,
      "train_tokens_per_second": 537.26
    },
    {
      "epoch": 0.005054334091483447,
      "grad_norm": 3.2350640296936035,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.1263377070426941,
      "num_input_tokens_seen": 9800,
      "step": 5,
      "train_runtime": 17.4855,
      "train_tokens_per_second": 560.463
    },
    {
      "epoch": 0.006065200909780136,
      "grad_norm": 3.040302038192749,
      "learning_rate": 2e-05,
      "loss": 0.15985995531082153,
      "num_input_tokens_seen": 11810,
      "step": 6,
      "train_runtime": 19.8808,
      "train_tokens_per_second": 594.039
    },
    {
      "epoch": 0.0070760677280768255,
      "grad_norm": 2.353942632675171,
      "learning_rate": 1.998987341772152e-05,
      "loss": 0.12493391335010529,
      "num_input_tokens_seen": 13386,
      "step": 7,
      "train_runtime": 22.1341,
      "train_tokens_per_second": 604.769
    },
    {
      "epoch": 0.008086934546373515,
      "grad_norm": 2.236783027648926,
      "learning_rate": 1.997974683544304e-05,
      "loss": 0.10446575284004211,
      "num_input_tokens_seen": 15448,
      "step": 8,
      "train_runtime": 24.4794,
      "train_tokens_per_second": 631.061
    },
    {
      "epoch": 0.009097801364670205,
      "grad_norm": 2.195513963699341,
      "learning_rate": 1.996962025316456e-05,
      "loss": 0.11554163694381714,
      "num_input_tokens_seen": 17280,
      "step": 9,
      "train_runtime": 26.7545,
      "train_tokens_per_second": 645.872
    },
    {
      "epoch": 0.010108668182966895,
      "grad_norm": 2.050267457962036,
      "learning_rate": 1.9959493670886078e-05,
      "loss": 0.1334109902381897,
      "num_input_tokens_seen": 19228,
      "step": 10,
      "train_runtime": 29.0256,
      "train_tokens_per_second": 662.45
    },
    {
      "epoch": 0.011119535001263583,
      "grad_norm": 2.197054862976074,
      "learning_rate": 1.9949367088607597e-05,
      "loss": 0.13452044129371643,
      "num_input_tokens_seen": 21282,
      "step": 11,
      "train_runtime": 31.4148,
      "train_tokens_per_second": 677.451
    },
    {
      "epoch": 0.012130401819560273,
      "grad_norm": 1.5404812097549438,
      "learning_rate": 1.9939240506329116e-05,
      "loss": 0.11211110651493073,
      "num_input_tokens_seen": 23866,
      "step": 12,
      "train_runtime": 33.8865,
      "train_tokens_per_second": 704.293
    },
    {
      "epoch": 0.013141268637856963,
      "grad_norm": 2.436166763305664,
      "learning_rate": 1.9929113924050635e-05,
      "loss": 0.15277458727359772,
      "num_input_tokens_seen": 25162,
      "step": 13,
      "train_runtime": 36.1,
      "train_tokens_per_second": 697.008
    },
    {
      "epoch": 0.014152135456153651,
      "grad_norm": 2.166334390640259,
      "learning_rate": 1.9918987341772154e-05,
      "loss": 0.1117962896823883,
      "num_input_tokens_seen": 27000,
      "step": 14,
      "train_runtime": 38.4314,
      "train_tokens_per_second": 702.551
    },
    {
      "epoch": 0.015163002274450341,
      "grad_norm": 1.7016150951385498,
      "learning_rate": 1.9908860759493673e-05,
      "loss": 0.14756187796592712,
      "num_input_tokens_seen": 29284,
      "step": 15,
      "train_runtime": 40.8298,
      "train_tokens_per_second": 717.221
    },
    {
      "epoch": 0.01617386909274703,
      "grad_norm": 1.590873122215271,
      "learning_rate": 1.9898734177215192e-05,
      "loss": 0.10208898782730103,
      "num_input_tokens_seen": 30880,
      "step": 16,
      "train_runtime": 43.0632,
      "train_tokens_per_second": 717.086
    },
    {
      "epoch": 0.01718473591104372,
      "grad_norm": 1.8935258388519287,
      "learning_rate": 1.988860759493671e-05,
      "loss": 0.12088144570589066,
      "num_input_tokens_seen": 32636,
      "step": 17,
      "train_runtime": 45.3162,
      "train_tokens_per_second": 720.183
    },
    {
      "epoch": 0.01819560272934041,
      "grad_norm": 1.7732003927230835,
      "learning_rate": 1.9878481012658227e-05,
      "loss": 0.11159995198249817,
      "num_input_tokens_seen": 34086,
      "step": 18,
      "train_runtime": 47.5312,
      "train_tokens_per_second": 717.129
    },
    {
      "epoch": 0.0192064695476371,
      "grad_norm": 1.8494051694869995,
      "learning_rate": 1.986835443037975e-05,
      "loss": 0.13126812875270844,
      "num_input_tokens_seen": 35890,
      "step": 19,
      "train_runtime": 49.7621,
      "train_tokens_per_second": 721.232
    },
    {
      "epoch": 0.02021733636593379,
      "grad_norm": 1.9393033981323242,
      "learning_rate": 1.985822784810127e-05,
      "loss": 0.12357176840305328,
      "num_input_tokens_seen": 37808,
      "step": 20,
      "train_runtime": 52.0773,
      "train_tokens_per_second": 725.997
    },
    {
      "epoch": 0.02122820318423048,
      "grad_norm": 2.33729887008667,
      "learning_rate": 1.9848101265822785e-05,
      "loss": 0.12302634119987488,
      "num_input_tokens_seen": 39772,
      "step": 21,
      "train_runtime": 54.3308,
      "train_tokens_per_second": 732.034
    },
    {
      "epoch": 0.022239070002527166,
      "grad_norm": 2.332458019256592,
      "learning_rate": 1.9837974683544307e-05,
      "loss": 0.135231152176857,
      "num_input_tokens_seen": 41968,
      "step": 22,
      "train_runtime": 56.6929,
      "train_tokens_per_second": 740.268
    },
    {
      "epoch": 0.023249936820823856,
      "grad_norm": 1.934679388999939,
      "learning_rate": 1.9827848101265826e-05,
      "loss": 0.13030485808849335,
      "num_input_tokens_seen": 43884,
      "step": 23,
      "train_runtime": 58.9618,
      "train_tokens_per_second": 744.278
    },
    {
      "epoch": 0.024260803639120546,
      "grad_norm": 2.044053554534912,
      "learning_rate": 1.9817721518987342e-05,
      "loss": 0.09511726349592209,
      "num_input_tokens_seen": 45820,
      "step": 24,
      "train_runtime": 61.2022,
      "train_tokens_per_second": 748.666
    },
    {
      "epoch": 0.025271670457417236,
      "grad_norm": 1.9673645496368408,
      "learning_rate": 1.9807594936708864e-05,
      "loss": 0.12695713341236115,
      "num_input_tokens_seen": 47302,
      "step": 25,
      "train_runtime": 63.3997,
      "train_tokens_per_second": 746.092
    },
    {
      "epoch": 0.026282537275713926,
      "grad_norm": 1.7110137939453125,
      "learning_rate": 1.979746835443038e-05,
      "loss": 0.10450948029756546,
      "num_input_tokens_seen": 48978,
      "step": 26,
      "train_runtime": 65.5816,
      "train_tokens_per_second": 746.825
    },
    {
      "epoch": 0.027293404094010616,
      "grad_norm": 2.1584558486938477,
      "learning_rate": 1.97873417721519e-05,
      "loss": 0.15018154680728912,
      "num_input_tokens_seen": 50812,
      "step": 27,
      "train_runtime": 67.8371,
      "train_tokens_per_second": 749.03
    },
    {
      "epoch": 0.028304270912307302,
      "grad_norm": 2.2640867233276367,
      "learning_rate": 1.977721518987342e-05,
      "loss": 0.11018655449151993,
      "num_input_tokens_seen": 52666,
      "step": 28,
      "train_runtime": 70.1303,
      "train_tokens_per_second": 750.973
    },
    {
      "epoch": 0.029315137730603992,
      "grad_norm": 2.3095314502716064,
      "learning_rate": 1.9767088607594937e-05,
      "loss": 0.14648404717445374,
      "num_input_tokens_seen": 54330,
      "step": 29,
      "train_runtime": 72.3367,
      "train_tokens_per_second": 751.071
    },
    {
      "epoch": 0.030326004548900682,
      "grad_norm": 2.0651280879974365,
      "learning_rate": 1.9756962025316456e-05,
      "loss": 0.12345363199710846,
      "num_input_tokens_seen": 55990,
      "step": 30,
      "train_runtime": 74.6124,
      "train_tokens_per_second": 750.411
    },
    {
      "epoch": 0.03133687136719737,
      "grad_norm": 1.8660341501235962,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.09264306724071503,
      "num_input_tokens_seen": 57646,
      "step": 31,
      "train_runtime": 92.9802,
      "train_tokens_per_second": 619.981
    },
    {
      "epoch": 0.03234773818549406,
      "grad_norm": 1.4165245294570923,
      "learning_rate": 1.9736708860759494e-05,
      "loss": 0.09966084361076355,
      "num_input_tokens_seen": 59978,
      "step": 32,
      "train_runtime": 95.3586,
      "train_tokens_per_second": 628.973
    },
    {
      "epoch": 0.03335860500379075,
      "grad_norm": 2.1475155353546143,
      "learning_rate": 1.9726582278481014e-05,
      "loss": 0.12514939904212952,
      "num_input_tokens_seen": 61296,
      "step": 33,
      "train_runtime": 97.5187,
      "train_tokens_per_second": 628.556
    },
    {
      "epoch": 0.03436947182208744,
      "grad_norm": 1.5792697668075562,
      "learning_rate": 1.9716455696202533e-05,
      "loss": 0.12737156450748444,
      "num_input_tokens_seen": 63650,
      "step": 34,
      "train_runtime": 99.8854,
      "train_tokens_per_second": 637.23
    },
    {
      "epoch": 0.03538033864038413,
      "grad_norm": 1.7555632591247559,
      "learning_rate": 1.970632911392405e-05,
      "loss": 0.12677490711212158,
      "num_input_tokens_seen": 65694,
      "step": 35,
      "train_runtime": 102.2476,
      "train_tokens_per_second": 642.499
    },
    {
      "epoch": 0.03639120545868082,
      "grad_norm": 1.6866122484207153,
      "learning_rate": 1.969620253164557e-05,
      "loss": 0.09945788979530334,
      "num_input_tokens_seen": 67306,
      "step": 36,
      "train_runtime": 104.4451,
      "train_tokens_per_second": 644.415
    },
    {
      "epoch": 0.037402072276977505,
      "grad_norm": 2.158679962158203,
      "learning_rate": 1.968607594936709e-05,
      "loss": 0.14948192238807678,
      "num_input_tokens_seen": 69350,
      "step": 37,
      "train_runtime": 106.808,
      "train_tokens_per_second": 649.296
    },
    {
      "epoch": 0.0384129390952742,
      "grad_norm": 2.2676992416381836,
      "learning_rate": 1.967594936708861e-05,
      "loss": 0.14130638539791107,
      "num_input_tokens_seen": 71108,
      "step": 38,
      "train_runtime": 109.0227,
      "train_tokens_per_second": 652.231
    },
    {
      "epoch": 0.039423805913570885,
      "grad_norm": 1.7943005561828613,
      "learning_rate": 1.9665822784810128e-05,
      "loss": 0.13869130611419678,
      "num_input_tokens_seen": 72894,
      "step": 39,
      "train_runtime": 111.2519,
      "train_tokens_per_second": 655.216
    },
    {
      "epoch": 0.04043467273186758,
      "grad_norm": 1.3528027534484863,
      "learning_rate": 1.9655696202531647e-05,
      "loss": 0.10072000324726105,
      "num_input_tokens_seen": 74832,
      "step": 40,
      "train_runtime": 113.5818,
      "train_tokens_per_second": 658.838
    },
    {
      "epoch": 0.041445539550164265,
      "grad_norm": 1.7628815174102783,
      "learning_rate": 1.9645569620253166e-05,
      "loss": 0.13465619087219238,
      "num_input_tokens_seen": 76986,
      "step": 41,
      "train_runtime": 115.9664,
      "train_tokens_per_second": 663.865
    },
    {
      "epoch": 0.04245640636846096,
      "grad_norm": 2.3988149166107178,
      "learning_rate": 1.9635443037974685e-05,
      "loss": 0.10453267395496368,
      "num_input_tokens_seen": 78728,
      "step": 42,
      "train_runtime": 118.3009,
      "train_tokens_per_second": 665.49
    },
    {
      "epoch": 0.043467273186757645,
      "grad_norm": 1.4457941055297852,
      "learning_rate": 1.9625316455696204e-05,
      "loss": 0.08477690815925598,
      "num_input_tokens_seen": 80352,
      "step": 43,
      "train_runtime": 120.5416,
      "train_tokens_per_second": 666.591
    },
    {
      "epoch": 0.04447814000505433,
      "grad_norm": 1.8616442680358887,
      "learning_rate": 1.9615189873417723e-05,
      "loss": 0.10402233153581619,
      "num_input_tokens_seen": 82186,
      "step": 44,
      "train_runtime": 122.844,
      "train_tokens_per_second": 669.028
    },
    {
      "epoch": 0.045489006823351025,
      "grad_norm": 1.9784412384033203,
      "learning_rate": 1.9605063291139243e-05,
      "loss": 0.12321233749389648,
      "num_input_tokens_seen": 84134,
      "step": 45,
      "train_runtime": 125.1089,
      "train_tokens_per_second": 672.486
    },
    {
      "epoch": 0.04649987364164771,
      "grad_norm": 1.38789701461792,
      "learning_rate": 1.959493670886076e-05,
      "loss": 0.11749634146690369,
      "num_input_tokens_seen": 86412,
      "step": 46,
      "train_runtime": 127.5308,
      "train_tokens_per_second": 677.577
    },
    {
      "epoch": 0.047510740459944405,
      "grad_norm": 1.7648074626922607,
      "learning_rate": 1.958481012658228e-05,
      "loss": 0.11442861706018448,
      "num_input_tokens_seen": 88426,
      "step": 47,
      "train_runtime": 129.9524,
      "train_tokens_per_second": 680.449
    },
    {
      "epoch": 0.04852160727824109,
      "grad_norm": 1.6100456714630127,
      "learning_rate": 1.95746835443038e-05,
      "loss": 0.11065111309289932,
      "num_input_tokens_seen": 90310,
      "step": 48,
      "train_runtime": 132.2487,
      "train_tokens_per_second": 682.88
    },
    {
      "epoch": 0.04953247409653778,
      "grad_norm": 1.5381323099136353,
      "learning_rate": 1.956455696202532e-05,
      "loss": 0.11945416778326035,
      "num_input_tokens_seen": 92530,
      "step": 49,
      "train_runtime": 134.598,
      "train_tokens_per_second": 687.455
    },
    {
      "epoch": 0.05054334091483447,
      "grad_norm": 1.450587511062622,
      "learning_rate": 1.9554430379746835e-05,
      "loss": 0.10527409613132477,
      "num_input_tokens_seen": 94718,
      "step": 50,
      "train_runtime": 136.9365,
      "train_tokens_per_second": 691.693
    },
    {
      "epoch": 0.05155420773313116,
      "grad_norm": 1.7181240320205688,
      "learning_rate": 1.9544303797468357e-05,
      "loss": 0.10942867398262024,
      "num_input_tokens_seen": 96530,
      "step": 51,
      "train_runtime": 139.1971,
      "train_tokens_per_second": 693.477
    },
    {
      "epoch": 0.05256507455142785,
      "grad_norm": 1.6302286386489868,
      "learning_rate": 1.9534177215189876e-05,
      "loss": 0.1107134222984314,
      "num_input_tokens_seen": 98434,
      "step": 52,
      "train_runtime": 141.4523,
      "train_tokens_per_second": 695.881
    },
    {
      "epoch": 0.05357594136972454,
      "grad_norm": 1.4967299699783325,
      "learning_rate": 1.9524050632911392e-05,
      "loss": 0.10694774240255356,
      "num_input_tokens_seen": 100400,
      "step": 53,
      "train_runtime": 143.7107,
      "train_tokens_per_second": 698.626
    },
    {
      "epoch": 0.05458680818802123,
      "grad_norm": 1.5010350942611694,
      "learning_rate": 1.9513924050632914e-05,
      "loss": 0.08706866204738617,
      "num_input_tokens_seen": 101800,
      "step": 54,
      "train_runtime": 145.902,
      "train_tokens_per_second": 697.729
    },
    {
      "epoch": 0.05559767500631792,
      "grad_norm": 1.9456924200057983,
      "learning_rate": 1.9503797468354433e-05,
      "loss": 0.1087537631392479,
      "num_input_tokens_seen": 103632,
      "step": 55,
      "train_runtime": 148.1733,
      "train_tokens_per_second": 699.397
    },
    {
      "epoch": 0.056608541824614604,
      "grad_norm": 1.7735391855239868,
      "learning_rate": 1.949367088607595e-05,
      "loss": 0.14347060024738312,
      "num_input_tokens_seen": 105654,
      "step": 56,
      "train_runtime": 150.5644,
      "train_tokens_per_second": 701.719
    },
    {
      "epoch": 0.0576194086429113,
      "grad_norm": 1.873093843460083,
      "learning_rate": 1.948354430379747e-05,
      "loss": 0.13866883516311646,
      "num_input_tokens_seen": 107452,
      "step": 57,
      "train_runtime": 152.8554,
      "train_tokens_per_second": 702.965
    },
    {
      "epoch": 0.058630275461207984,
      "grad_norm": 1.2524882555007935,
      "learning_rate": 1.9473417721518987e-05,
      "loss": 0.08895567059516907,
      "num_input_tokens_seen": 109702,
      "step": 58,
      "train_runtime": 155.2944,
      "train_tokens_per_second": 706.413
    },
    {
      "epoch": 0.05964114227950468,
      "grad_norm": 1.770918369293213,
      "learning_rate": 1.9463291139240506e-05,
      "loss": 0.11555160582065582,
      "num_input_tokens_seen": 111458,
      "step": 59,
      "train_runtime": 157.6024,
      "train_tokens_per_second": 707.21
    },
    {
      "epoch": 0.060652009097801364,
      "grad_norm": 2.2099945545196533,
      "learning_rate": 1.945316455696203e-05,
      "loss": 0.1112520769238472,
      "num_input_tokens_seen": 112866,
      "step": 60,
      "train_runtime": 159.7889,
      "train_tokens_per_second": 706.344
    },
    {
      "epoch": 0.06166287591609806,
      "grad_norm": 1.9527833461761475,
      "learning_rate": 1.9443037974683544e-05,
      "loss": 0.14135530591011047,
      "num_input_tokens_seen": 115128,
      "step": 61,
      "train_runtime": 175.6086,
      "train_tokens_per_second": 655.594
    },
    {
      "epoch": 0.06267374273439474,
      "grad_norm": 2.4236061573028564,
      "learning_rate": 1.9432911392405064e-05,
      "loss": 0.11961868405342102,
      "num_input_tokens_seen": 117058,
      "step": 62,
      "train_runtime": 177.917,
      "train_tokens_per_second": 657.936
    },
    {
      "epoch": 0.06368460955269144,
      "grad_norm": 1.5899832248687744,
      "learning_rate": 1.9422784810126586e-05,
      "loss": 0.1059185191988945,
      "num_input_tokens_seen": 119258,
      "step": 63,
      "train_runtime": 180.4015,
      "train_tokens_per_second": 661.07
    },
    {
      "epoch": 0.06469547637098812,
      "grad_norm": 1.4054566621780396,
      "learning_rate": 1.9412658227848102e-05,
      "loss": 0.0871809720993042,
      "num_input_tokens_seen": 121276,
      "step": 64,
      "train_runtime": 182.7583,
      "train_tokens_per_second": 663.587
    },
    {
      "epoch": 0.06570634318928481,
      "grad_norm": 1.5707159042358398,
      "learning_rate": 1.940253164556962e-05,
      "loss": 0.10112117230892181,
      "num_input_tokens_seen": 122996,
      "step": 65,
      "train_runtime": 185.035,
      "train_tokens_per_second": 664.718
    },
    {
      "epoch": 0.0667172100075815,
      "grad_norm": 1.552430510520935,
      "learning_rate": 1.939240506329114e-05,
      "loss": 0.107082799077034,
      "num_input_tokens_seen": 125242,
      "step": 66,
      "train_runtime": 187.3972,
      "train_tokens_per_second": 668.324
    },
    {
      "epoch": 0.0677280768258782,
      "grad_norm": 1.6381937265396118,
      "learning_rate": 1.938227848101266e-05,
      "loss": 0.1008966863155365,
      "num_input_tokens_seen": 127580,
      "step": 67,
      "train_runtime": 189.7902,
      "train_tokens_per_second": 672.216
    },
    {
      "epoch": 0.06873894364417488,
      "grad_norm": 1.7757294178009033,
      "learning_rate": 1.9372151898734178e-05,
      "loss": 0.09650348126888275,
      "num_input_tokens_seen": 129804,
      "step": 68,
      "train_runtime": 192.1347,
      "train_tokens_per_second": 675.589
    },
    {
      "epoch": 0.06974981046247157,
      "grad_norm": 1.7566887140274048,
      "learning_rate": 1.9362025316455697e-05,
      "loss": 0.1532202512025833,
      "num_input_tokens_seen": 131960,
      "step": 69,
      "train_runtime": 194.4645,
      "train_tokens_per_second": 678.581
    },
    {
      "epoch": 0.07076067728076826,
      "grad_norm": 1.4537709951400757,
      "learning_rate": 1.9351898734177216e-05,
      "loss": 0.12261821329593658,
      "num_input_tokens_seen": 133790,
      "step": 70,
      "train_runtime": 196.7306,
      "train_tokens_per_second": 680.067
    },
    {
      "epoch": 0.07177154409906494,
      "grad_norm": 1.4357869625091553,
      "learning_rate": 1.9341772151898735e-05,
      "loss": 0.09564615786075592,
      "num_input_tokens_seen": 135344,
      "step": 71,
      "train_runtime": 198.9071,
      "train_tokens_per_second": 680.438
    },
    {
      "epoch": 0.07278241091736164,
      "grad_norm": 1.570696234703064,
      "learning_rate": 1.9331645569620254e-05,
      "loss": 0.08187444508075714,
      "num_input_tokens_seen": 136974,
      "step": 72,
      "train_runtime": 201.0947,
      "train_tokens_per_second": 681.142
    },
    {
      "epoch": 0.07379327773565833,
      "grad_norm": 1.7900714874267578,
      "learning_rate": 1.9321518987341774e-05,
      "loss": 0.10589981079101562,
      "num_input_tokens_seen": 138742,
      "step": 73,
      "train_runtime": 203.3062,
      "train_tokens_per_second": 682.429
    },
    {
      "epoch": 0.07480414455395501,
      "grad_norm": 1.569770097732544,
      "learning_rate": 1.9311392405063293e-05,
      "loss": 0.10022330284118652,
      "num_input_tokens_seen": 140352,
      "step": 74,
      "train_runtime": 205.5147,
      "train_tokens_per_second": 682.929
    },
    {
      "epoch": 0.0758150113722517,
      "grad_norm": 1.275302767753601,
      "learning_rate": 1.930126582278481e-05,
      "loss": 0.08037956804037094,
      "num_input_tokens_seen": 142278,
      "step": 75,
      "train_runtime": 207.7919,
      "train_tokens_per_second": 684.714
    },
    {
      "epoch": 0.0768258781905484,
      "grad_norm": 1.855034351348877,
      "learning_rate": 1.929113924050633e-05,
      "loss": 0.10013698041439056,
      "num_input_tokens_seen": 144160,
      "step": 76,
      "train_runtime": 210.0556,
      "train_tokens_per_second": 686.294
    },
    {
      "epoch": 0.07783674500884509,
      "grad_norm": 5.506159782409668,
      "learning_rate": 1.928101265822785e-05,
      "loss": 0.10819685459136963,
      "num_input_tokens_seen": 146444,
      "step": 77,
      "train_runtime": 212.4903,
      "train_tokens_per_second": 689.18
    },
    {
      "epoch": 0.07884761182714177,
      "grad_norm": 1.6996967792510986,
      "learning_rate": 1.927088607594937e-05,
      "loss": 0.09614938497543335,
      "num_input_tokens_seen": 148724,
      "step": 78,
      "train_runtime": 214.8652,
      "train_tokens_per_second": 692.173
    },
    {
      "epoch": 0.07985847864543846,
      "grad_norm": 1.5823657512664795,
      "learning_rate": 1.9260759493670888e-05,
      "loss": 0.10873556137084961,
      "num_input_tokens_seen": 150364,
      "step": 79,
      "train_runtime": 217.0375,
      "train_tokens_per_second": 692.802
    },
    {
      "epoch": 0.08086934546373516,
      "grad_norm": 2.107003688812256,
      "learning_rate": 1.9250632911392407e-05,
      "loss": 0.14108863472938538,
      "num_input_tokens_seen": 152144,
      "step": 80,
      "train_runtime": 219.3032,
      "train_tokens_per_second": 693.761
    },
    {
      "epoch": 0.08188021228203184,
      "grad_norm": 1.9523966312408447,
      "learning_rate": 1.9240506329113926e-05,
      "loss": 0.11961855739355087,
      "num_input_tokens_seen": 153538,
      "step": 81,
      "train_runtime": 221.4523,
      "train_tokens_per_second": 693.323
    },
    {
      "epoch": 0.08289107910032853,
      "grad_norm": 1.4230780601501465,
      "learning_rate": 1.9230379746835445e-05,
      "loss": 0.10475742816925049,
      "num_input_tokens_seen": 155578,
      "step": 82,
      "train_runtime": 223.8001,
      "train_tokens_per_second": 695.165
    },
    {
      "epoch": 0.08390194591862522,
      "grad_norm": 1.6393123865127563,
      "learning_rate": 1.9220253164556964e-05,
      "loss": 0.10261191427707672,
      "num_input_tokens_seen": 157254,
      "step": 83,
      "train_runtime": 226.0656,
      "train_tokens_per_second": 695.612
    },
    {
      "epoch": 0.08491281273692192,
      "grad_norm": 1.2770360708236694,
      "learning_rate": 1.9210126582278483e-05,
      "loss": 0.07230052351951599,
      "num_input_tokens_seen": 158702,
      "step": 84,
      "train_runtime": 228.1978,
      "train_tokens_per_second": 695.458
    },
    {
      "epoch": 0.0859236795552186,
      "grad_norm": 1.7239381074905396,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.11455848067998886,
      "num_input_tokens_seen": 160346,
      "step": 85,
      "train_runtime": 230.4051,
      "train_tokens_per_second": 695.931
    },
    {
      "epoch": 0.08693454637351529,
      "grad_norm": 1.8309370279312134,
      "learning_rate": 1.918987341772152e-05,
      "loss": 0.13381218910217285,
      "num_input_tokens_seen": 162460,
      "step": 86,
      "train_runtime": 232.7105,
      "train_tokens_per_second": 698.121
    },
    {
      "epoch": 0.08794541319181198,
      "grad_norm": 1.411454200744629,
      "learning_rate": 1.917974683544304e-05,
      "loss": 0.10632367432117462,
      "num_input_tokens_seen": 164650,
      "step": 87,
      "train_runtime": 235.0917,
      "train_tokens_per_second": 700.365
    },
    {
      "epoch": 0.08895628001010866,
      "grad_norm": 1.9270180463790894,
      "learning_rate": 1.916962025316456e-05,
      "loss": 0.10820554196834564,
      "num_input_tokens_seen": 166154,
      "step": 88,
      "train_runtime": 237.2707,
      "train_tokens_per_second": 700.272
    },
    {
      "epoch": 0.08996714682840536,
      "grad_norm": 1.873599648475647,
      "learning_rate": 1.915949367088608e-05,
      "loss": 0.1287396252155304,
      "num_input_tokens_seen": 167978,
      "step": 89,
      "train_runtime": 239.5113,
      "train_tokens_per_second": 701.336
    },
    {
      "epoch": 0.09097801364670205,
      "grad_norm": 1.8736536502838135,
      "learning_rate": 1.9149367088607595e-05,
      "loss": 0.12423866987228394,
      "num_input_tokens_seen": 169662,
      "step": 90,
      "train_runtime": 241.7224,
      "train_tokens_per_second": 701.888
    },
    {
      "epoch": 0.09198888046499874,
      "grad_norm": 1.5817285776138306,
      "learning_rate": 1.9139240506329117e-05,
      "loss": 0.11752364039421082,
      "num_input_tokens_seen": 171702,
      "step": 91,
      "train_runtime": 257.4921,
      "train_tokens_per_second": 666.824
    },
    {
      "epoch": 0.09299974728329542,
      "grad_norm": 1.419991374015808,
      "learning_rate": 1.9129113924050636e-05,
      "loss": 0.0935627743601799,
      "num_input_tokens_seen": 173154,
      "step": 92,
      "train_runtime": 259.6397,
      "train_tokens_per_second": 666.901
    },
    {
      "epoch": 0.09401061410159212,
      "grad_norm": 1.9239012002944946,
      "learning_rate": 1.9118987341772152e-05,
      "loss": 0.10259293019771576,
      "num_input_tokens_seen": 175074,
      "step": 93,
      "train_runtime": 261.8772,
      "train_tokens_per_second": 668.535
    },
    {
      "epoch": 0.09502148091988881,
      "grad_norm": 1.7783845663070679,
      "learning_rate": 1.9108860759493674e-05,
      "loss": 0.08148714900016785,
      "num_input_tokens_seen": 177174,
      "step": 94,
      "train_runtime": 264.238,
      "train_tokens_per_second": 670.509
    },
    {
      "epoch": 0.09603234773818549,
      "grad_norm": 1.4966665506362915,
      "learning_rate": 1.9098734177215193e-05,
      "loss": 0.08918128907680511,
      "num_input_tokens_seen": 178894,
      "step": 95,
      "train_runtime": 266.4846,
      "train_tokens_per_second": 671.311
    },
    {
      "epoch": 0.09704321455648218,
      "grad_norm": 1.6010295152664185,
      "learning_rate": 1.908860759493671e-05,
      "loss": 0.10201150178909302,
      "num_input_tokens_seen": 180318,
      "step": 96,
      "train_runtime": 268.6632,
      "train_tokens_per_second": 671.167
    },
    {
      "epoch": 0.09805408137477888,
      "grad_norm": 1.4962912797927856,
      "learning_rate": 1.9078481012658228e-05,
      "loss": 0.11616621911525726,
      "num_input_tokens_seen": 182496,
      "step": 97,
      "train_runtime": 271.0206,
      "train_tokens_per_second": 673.366
    },
    {
      "epoch": 0.09906494819307556,
      "grad_norm": 1.4796838760375977,
      "learning_rate": 1.9068354430379747e-05,
      "loss": 0.12867720425128937,
      "num_input_tokens_seen": 184870,
      "step": 98,
      "train_runtime": 273.4063,
      "train_tokens_per_second": 676.173
    },
    {
      "epoch": 0.10007581501137225,
      "grad_norm": 1.7509914636611938,
      "learning_rate": 1.9058227848101266e-05,
      "loss": 0.12179767340421677,
      "num_input_tokens_seen": 187088,
      "step": 99,
      "train_runtime": 275.8123,
      "train_tokens_per_second": 678.316
    },
    {
      "epoch": 0.10108668182966894,
      "grad_norm": 1.5425728559494019,
      "learning_rate": 1.9048101265822785e-05,
      "loss": 0.11425593495368958,
      "num_input_tokens_seen": 188710,
      "step": 100,
      "train_runtime": 278.1176,
      "train_tokens_per_second": 678.526
    },
    {
      "epoch": 0.10209754864796564,
      "grad_norm": 1.8624966144561768,
      "learning_rate": 1.9037974683544304e-05,
      "loss": 0.14122828841209412,
      "num_input_tokens_seen": 190694,
      "step": 101,
      "train_runtime": 280.5004,
      "train_tokens_per_second": 679.835
    },
    {
      "epoch": 0.10310841546626232,
      "grad_norm": 1.6076759099960327,
      "learning_rate": 1.9027848101265824e-05,
      "loss": 0.1249576210975647,
      "num_input_tokens_seen": 192500,
      "step": 102,
      "train_runtime": 282.9772,
      "train_tokens_per_second": 680.267
    },
    {
      "epoch": 0.10411928228455901,
      "grad_norm": 1.8289830684661865,
      "learning_rate": 1.9017721518987343e-05,
      "loss": 0.12983912229537964,
      "num_input_tokens_seen": 194224,
      "step": 103,
      "train_runtime": 285.3966,
      "train_tokens_per_second": 680.541
    },
    {
      "epoch": 0.1051301491028557,
      "grad_norm": 1.4450889825820923,
      "learning_rate": 1.9007594936708862e-05,
      "loss": 0.1161910742521286,
      "num_input_tokens_seen": 196598,
      "step": 104,
      "train_runtime": 287.817,
      "train_tokens_per_second": 683.066
    },
    {
      "epoch": 0.10614101592115238,
      "grad_norm": 1.6481293439865112,
      "learning_rate": 1.899746835443038e-05,
      "loss": 0.12809798121452332,
      "num_input_tokens_seen": 199018,
      "step": 105,
      "train_runtime": 290.2358,
      "train_tokens_per_second": 685.711
    },
    {
      "epoch": 0.10715188273944908,
      "grad_norm": 1.6318658590316772,
      "learning_rate": 1.89873417721519e-05,
      "loss": 0.11569087207317352,
      "num_input_tokens_seen": 200604,
      "step": 106,
      "train_runtime": 292.5091,
      "train_tokens_per_second": 685.804
    },
    {
      "epoch": 0.10816274955774577,
      "grad_norm": 1.725794792175293,
      "learning_rate": 1.897721518987342e-05,
      "loss": 0.1173192709684372,
      "num_input_tokens_seen": 202286,
      "step": 107,
      "train_runtime": 294.7331,
      "train_tokens_per_second": 686.336
    },
    {
      "epoch": 0.10917361637604246,
      "grad_norm": 1.5027580261230469,
      "learning_rate": 1.8967088607594938e-05,
      "loss": 0.12167303264141083,
      "num_input_tokens_seen": 204376,
      "step": 108,
      "train_runtime": 297.0733,
      "train_tokens_per_second": 687.965
    },
    {
      "epoch": 0.11018448319433914,
      "grad_norm": 1.6913418769836426,
      "learning_rate": 1.8956962025316457e-05,
      "loss": 0.11281217634677887,
      "num_input_tokens_seen": 206010,
      "step": 109,
      "train_runtime": 299.2876,
      "train_tokens_per_second": 688.334
    },
    {
      "epoch": 0.11119535001263584,
      "grad_norm": 1.5609663724899292,
      "learning_rate": 1.8946835443037976e-05,
      "loss": 0.14416992664337158,
      "num_input_tokens_seen": 207814,
      "step": 110,
      "train_runtime": 301.5041,
      "train_tokens_per_second": 689.258
    },
    {
      "epoch": 0.11220621683093253,
      "grad_norm": 1.447195291519165,
      "learning_rate": 1.8936708860759495e-05,
      "loss": 0.08768196403980255,
      "num_input_tokens_seen": 209512,
      "step": 111,
      "train_runtime": 303.7733,
      "train_tokens_per_second": 689.699
    },
    {
      "epoch": 0.11321708364922921,
      "grad_norm": 1.609835147857666,
      "learning_rate": 1.8926582278481014e-05,
      "loss": 0.11692729592323303,
      "num_input_tokens_seen": 211092,
      "step": 112,
      "train_runtime": 305.9966,
      "train_tokens_per_second": 689.851
    },
    {
      "epoch": 0.1142279504675259,
      "grad_norm": 2.001396894454956,
      "learning_rate": 1.8916455696202533e-05,
      "loss": 0.12284542620182037,
      "num_input_tokens_seen": 212812,
      "step": 113,
      "train_runtime": 308.2443,
      "train_tokens_per_second": 690.4
    },
    {
      "epoch": 0.1152388172858226,
      "grad_norm": 1.9883241653442383,
      "learning_rate": 1.8906329113924053e-05,
      "loss": 0.12710854411125183,
      "num_input_tokens_seen": 214682,
      "step": 114,
      "train_runtime": 310.611,
      "train_tokens_per_second": 691.16
    },
    {
      "epoch": 0.11624968410411929,
      "grad_norm": 2.105234146118164,
      "learning_rate": 1.889620253164557e-05,
      "loss": 0.14509180188179016,
      "num_input_tokens_seen": 216826,
      "step": 115,
      "train_runtime": 313.0412,
      "train_tokens_per_second": 692.644
    },
    {
      "epoch": 0.11726055092241597,
      "grad_norm": 1.8560389280319214,
      "learning_rate": 1.888607594936709e-05,
      "loss": 0.13155341148376465,
      "num_input_tokens_seen": 218824,
      "step": 116,
      "train_runtime": 315.3893,
      "train_tokens_per_second": 693.822
    },
    {
      "epoch": 0.11827141774071266,
      "grad_norm": 1.8292335271835327,
      "learning_rate": 1.887594936708861e-05,
      "loss": 0.12597715854644775,
      "num_input_tokens_seen": 220542,
      "step": 117,
      "train_runtime": 317.6546,
      "train_tokens_per_second": 694.282
    },
    {
      "epoch": 0.11928228455900936,
      "grad_norm": 1.6812541484832764,
      "learning_rate": 1.886582278481013e-05,
      "loss": 0.10085409134626389,
      "num_input_tokens_seen": 222116,
      "step": 118,
      "train_runtime": 319.8691,
      "train_tokens_per_second": 694.396
    },
    {
      "epoch": 0.12029315137730603,
      "grad_norm": 1.4323360919952393,
      "learning_rate": 1.8855696202531648e-05,
      "loss": 0.08915829658508301,
      "num_input_tokens_seen": 224040,
      "step": 119,
      "train_runtime": 322.2311,
      "train_tokens_per_second": 695.277
    },
    {
      "epoch": 0.12130401819560273,
      "grad_norm": 1.7646902799606323,
      "learning_rate": 1.8845569620253167e-05,
      "loss": 0.11199093610048294,
      "num_input_tokens_seen": 225810,
      "step": 120,
      "train_runtime": 324.5507,
      "train_tokens_per_second": 695.762
    },
    {
      "epoch": 0.12231488501389942,
      "grad_norm": 1.4256528615951538,
      "learning_rate": 1.8835443037974686e-05,
      "loss": 0.0777023658156395,
      "num_input_tokens_seen": 227214,
      "step": 121,
      "train_runtime": 341.3149,
      "train_tokens_per_second": 665.702
    },
    {
      "epoch": 0.12332575183219611,
      "grad_norm": 1.4666156768798828,
      "learning_rate": 1.8825316455696202e-05,
      "loss": 0.10978391766548157,
      "num_input_tokens_seen": 229020,
      "step": 122,
      "train_runtime": 343.5689,
      "train_tokens_per_second": 666.591
    },
    {
      "epoch": 0.1243366186504928,
      "grad_norm": 1.4810764789581299,
      "learning_rate": 1.8815189873417724e-05,
      "loss": 0.09080948680639267,
      "num_input_tokens_seen": 230962,
      "step": 123,
      "train_runtime": 345.9213,
      "train_tokens_per_second": 667.672
    },
    {
      "epoch": 0.1253474854687895,
      "grad_norm": 1.444600224494934,
      "learning_rate": 1.8805063291139243e-05,
      "loss": 0.10851176083087921,
      "num_input_tokens_seen": 232580,
      "step": 124,
      "train_runtime": 348.1398,
      "train_tokens_per_second": 668.065
    },
    {
      "epoch": 0.12635835228708617,
      "grad_norm": 1.5433681011199951,
      "learning_rate": 1.879493670886076e-05,
      "loss": 0.11208166182041168,
      "num_input_tokens_seen": 234272,
      "step": 125,
      "train_runtime": 350.3289,
      "train_tokens_per_second": 668.72
    },
    {
      "epoch": 0.12736921910538287,
      "grad_norm": 1.2799385786056519,
      "learning_rate": 1.878481012658228e-05,
      "loss": 0.09377605468034744,
      "num_input_tokens_seen": 236512,
      "step": 126,
      "train_runtime": 352.664,
      "train_tokens_per_second": 670.644
    },
    {
      "epoch": 0.12838008592367955,
      "grad_norm": 1.6316152811050415,
      "learning_rate": 1.87746835443038e-05,
      "loss": 0.1482527256011963,
      "num_input_tokens_seen": 238912,
      "step": 127,
      "train_runtime": 355.0241,
      "train_tokens_per_second": 672.946
    },
    {
      "epoch": 0.12939095274197623,
      "grad_norm": 1.8516558408737183,
      "learning_rate": 1.8764556962025316e-05,
      "loss": 0.12970715761184692,
      "num_input_tokens_seen": 241040,
      "step": 128,
      "train_runtime": 357.3299,
      "train_tokens_per_second": 674.559
    },
    {
      "epoch": 0.13040181956027294,
      "grad_norm": 1.7439422607421875,
      "learning_rate": 1.875443037974684e-05,
      "loss": 0.10727793723344803,
      "num_input_tokens_seen": 242810,
      "step": 129,
      "train_runtime": 359.5624,
      "train_tokens_per_second": 675.293
    },
    {
      "epoch": 0.13141268637856962,
      "grad_norm": 1.7847952842712402,
      "learning_rate": 1.8744303797468355e-05,
      "loss": 0.12795516848564148,
      "num_input_tokens_seen": 244636,
      "step": 130,
      "train_runtime": 361.8498,
      "train_tokens_per_second": 676.071
    },
    {
      "epoch": 0.1324235531968663,
      "grad_norm": 1.4231994152069092,
      "learning_rate": 1.8734177215189874e-05,
      "loss": 0.09431955963373184,
      "num_input_tokens_seen": 246324,
      "step": 131,
      "train_runtime": 364.1371,
      "train_tokens_per_second": 676.46
    },
    {
      "epoch": 0.133434420015163,
      "grad_norm": 1.6861406564712524,
      "learning_rate": 1.8724050632911396e-05,
      "loss": 0.1439673751592636,
      "num_input_tokens_seen": 248122,
      "step": 132,
      "train_runtime": 366.5857,
      "train_tokens_per_second": 676.846
    },
    {
      "epoch": 0.1344452868334597,
      "grad_norm": 1.4552510976791382,
      "learning_rate": 1.8713924050632912e-05,
      "loss": 0.09559144824743271,
      "num_input_tokens_seen": 250064,
      "step": 133,
      "train_runtime": 368.913,
      "train_tokens_per_second": 677.84
    },
    {
      "epoch": 0.1354561536517564,
      "grad_norm": 1.485036015510559,
      "learning_rate": 1.870379746835443e-05,
      "loss": 0.0976657047867775,
      "num_input_tokens_seen": 251986,
      "step": 134,
      "train_runtime": 371.2416,
      "train_tokens_per_second": 678.766
    },
    {
      "epoch": 0.13646702047005307,
      "grad_norm": 2.0229406356811523,
      "learning_rate": 1.8693670886075953e-05,
      "loss": 0.10776853561401367,
      "num_input_tokens_seen": 253914,
      "step": 135,
      "train_runtime": 373.5523,
      "train_tokens_per_second": 679.728
    },
    {
      "epoch": 0.13747788728834975,
      "grad_norm": 1.7993659973144531,
      "learning_rate": 1.868354430379747e-05,
      "loss": 0.0977972149848938,
      "num_input_tokens_seen": 255648,
      "step": 136,
      "train_runtime": 375.7938,
      "train_tokens_per_second": 680.288
    },
    {
      "epoch": 0.13848875410664646,
      "grad_norm": 1.521706461906433,
      "learning_rate": 1.8673417721518988e-05,
      "loss": 0.08950497210025787,
      "num_input_tokens_seen": 257298,
      "step": 137,
      "train_runtime": 377.9829,
      "train_tokens_per_second": 680.713
    },
    {
      "epoch": 0.13949962092494314,
      "grad_norm": 1.6079732179641724,
      "learning_rate": 1.8663291139240507e-05,
      "loss": 0.1278381645679474,
      "num_input_tokens_seen": 258954,
      "step": 138,
      "train_runtime": 380.2355,
      "train_tokens_per_second": 681.036
    },
    {
      "epoch": 0.14051048774323982,
      "grad_norm": 1.5357948541641235,
      "learning_rate": 1.8653164556962026e-05,
      "loss": 0.09309880435466766,
      "num_input_tokens_seen": 260652,
      "step": 139,
      "train_runtime": 382.4921,
      "train_tokens_per_second": 681.457
    },
    {
      "epoch": 0.14152135456153653,
      "grad_norm": 1.5426337718963623,
      "learning_rate": 1.8643037974683545e-05,
      "loss": 0.09987395256757736,
      "num_input_tokens_seen": 262534,
      "step": 140,
      "train_runtime": 384.7794,
      "train_tokens_per_second": 682.297
    },
    {
      "epoch": 0.1425322213798332,
      "grad_norm": 1.3595482110977173,
      "learning_rate": 1.8632911392405064e-05,
      "loss": 0.08942777663469315,
      "num_input_tokens_seen": 264060,
      "step": 141,
      "train_runtime": 387.0331,
      "train_tokens_per_second": 682.267
    },
    {
      "epoch": 0.1435430881981299,
      "grad_norm": 1.641709804534912,
      "learning_rate": 1.8622784810126584e-05,
      "loss": 0.08591048419475555,
      "num_input_tokens_seen": 265636,
      "step": 142,
      "train_runtime": 389.2082,
      "train_tokens_per_second": 682.504
    },
    {
      "epoch": 0.1445539550164266,
      "grad_norm": 1.7246475219726562,
      "learning_rate": 1.8612658227848103e-05,
      "loss": 0.12977781891822815,
      "num_input_tokens_seen": 267074,
      "step": 143,
      "train_runtime": 391.3529,
      "train_tokens_per_second": 682.438
    },
    {
      "epoch": 0.14556482183472327,
      "grad_norm": 1.306073546409607,
      "learning_rate": 1.8602531645569622e-05,
      "loss": 0.09224815666675568,
      "num_input_tokens_seen": 269338,
      "step": 144,
      "train_runtime": 393.6818,
      "train_tokens_per_second": 684.151
    },
    {
      "epoch": 0.14657568865301995,
      "grad_norm": 1.6135250329971313,
      "learning_rate": 1.859240506329114e-05,
      "loss": 0.1096469908952713,
      "num_input_tokens_seen": 271342,
      "step": 145,
      "train_runtime": 395.971,
      "train_tokens_per_second": 685.257
    },
    {
      "epoch": 0.14758655547131666,
      "grad_norm": 1.6248900890350342,
      "learning_rate": 1.858227848101266e-05,
      "loss": 0.08891341090202332,
      "num_input_tokens_seen": 273232,
      "step": 146,
      "train_runtime": 398.2645,
      "train_tokens_per_second": 686.057
    },
    {
      "epoch": 0.14859742228961334,
      "grad_norm": 1.447414755821228,
      "learning_rate": 1.857215189873418e-05,
      "loss": 0.09662678837776184,
      "num_input_tokens_seen": 275106,
      "step": 147,
      "train_runtime": 400.5277,
      "train_tokens_per_second": 686.859
    },
    {
      "epoch": 0.14960828910791002,
      "grad_norm": 1.7878117561340332,
      "learning_rate": 1.8562025316455698e-05,
      "loss": 0.1260918527841568,
      "num_input_tokens_seen": 276940,
      "step": 148,
      "train_runtime": 402.7234,
      "train_tokens_per_second": 687.668
    },
    {
      "epoch": 0.15061915592620673,
      "grad_norm": 1.4992377758026123,
      "learning_rate": 1.8551898734177217e-05,
      "loss": 0.11068637669086456,
      "num_input_tokens_seen": 278862,
      "step": 149,
      "train_runtime": 405.0307,
      "train_tokens_per_second": 688.496
    },
    {
      "epoch": 0.1516300227445034,
      "grad_norm": 1.598223090171814,
      "learning_rate": 1.8541772151898736e-05,
      "loss": 0.1253667175769806,
      "num_input_tokens_seen": 280454,
      "step": 150,
      "train_runtime": 407.2075,
      "train_tokens_per_second": 688.725
    },
    {
      "epoch": 0.15264088956280011,
      "grad_norm": 1.4451793432235718,
      "learning_rate": 1.8531645569620255e-05,
      "loss": 0.09338060021400452,
      "num_input_tokens_seen": 282348,
      "step": 151,
      "train_runtime": 424.3627,
      "train_tokens_per_second": 665.346
    },
    {
      "epoch": 0.1536517563810968,
      "grad_norm": 1.3872953653335571,
      "learning_rate": 1.8521518987341774e-05,
      "loss": 0.0805031806230545,
      "num_input_tokens_seen": 284070,
      "step": 152,
      "train_runtime": 426.71,
      "train_tokens_per_second": 665.721
    },
    {
      "epoch": 0.15466262319939347,
      "grad_norm": 1.7340925931930542,
      "learning_rate": 1.8511392405063293e-05,
      "loss": 0.1279347985982895,
      "num_input_tokens_seen": 285946,
      "step": 153,
      "train_runtime": 429.024,
      "train_tokens_per_second": 666.504
    },
    {
      "epoch": 0.15567349001769018,
      "grad_norm": 1.3884550333023071,
      "learning_rate": 1.850126582278481e-05,
      "loss": 0.11289025843143463,
      "num_input_tokens_seen": 287824,
      "step": 154,
      "train_runtime": 431.2747,
      "train_tokens_per_second": 667.38
    },
    {
      "epoch": 0.15668435683598686,
      "grad_norm": 1.3569774627685547,
      "learning_rate": 1.849113924050633e-05,
      "loss": 0.10962873697280884,
      "num_input_tokens_seen": 290194,
      "step": 155,
      "train_runtime": 433.69,
      "train_tokens_per_second": 669.128
    },
    {
      "epoch": 0.15769522365428354,
      "grad_norm": 1.3600742816925049,
      "learning_rate": 1.848101265822785e-05,
      "loss": 0.10517730563879013,
      "num_input_tokens_seen": 292346,
      "step": 156,
      "train_runtime": 436.0582,
      "train_tokens_per_second": 670.429
    },
    {
      "epoch": 0.15870609047258025,
      "grad_norm": 1.3142476081848145,
      "learning_rate": 1.8470886075949366e-05,
      "loss": 0.11763598769903183,
      "num_input_tokens_seen": 294520,
      "step": 157,
      "train_runtime": 438.4213,
      "train_tokens_per_second": 671.774
    },
    {
      "epoch": 0.15971695729087693,
      "grad_norm": 1.3365898132324219,
      "learning_rate": 1.846075949367089e-05,
      "loss": 0.11077255010604858,
      "num_input_tokens_seen": 296632,
      "step": 158,
      "train_runtime": 440.8273,
      "train_tokens_per_second": 672.898
    },
    {
      "epoch": 0.1607278241091736,
      "grad_norm": 1.4594699144363403,
      "learning_rate": 1.8450632911392408e-05,
      "loss": 0.09953862428665161,
      "num_input_tokens_seen": 298670,
      "step": 159,
      "train_runtime": 443.2049,
      "train_tokens_per_second": 673.887
    },
    {
      "epoch": 0.1617386909274703,
      "grad_norm": 1.6890226602554321,
      "learning_rate": 1.8440506329113924e-05,
      "loss": 0.11373594403266907,
      "num_input_tokens_seen": 300156,
      "step": 160,
      "train_runtime": 445.3824,
      "train_tokens_per_second": 673.929
    },
    {
      "epoch": 0.162749557745767,
      "grad_norm": 1.4438869953155518,
      "learning_rate": 1.8430379746835446e-05,
      "loss": 0.12429876625537872,
      "num_input_tokens_seen": 301862,
      "step": 161,
      "train_runtime": 447.5962,
      "train_tokens_per_second": 674.407
    },
    {
      "epoch": 0.16376042456406367,
      "grad_norm": 1.6493380069732666,
      "learning_rate": 1.8420253164556962e-05,
      "loss": 0.09954585134983063,
      "num_input_tokens_seen": 303282,
      "step": 162,
      "train_runtime": 449.7545,
      "train_tokens_per_second": 674.328
    },
    {
      "epoch": 0.16477129138236038,
      "grad_norm": 1.360890507698059,
      "learning_rate": 1.841012658227848e-05,
      "loss": 0.10392744094133377,
      "num_input_tokens_seen": 304986,
      "step": 163,
      "train_runtime": 451.9741,
      "train_tokens_per_second": 674.786
    },
    {
      "epoch": 0.16578215820065706,
      "grad_norm": 1.5973113775253296,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.10154791176319122,
      "num_input_tokens_seen": 306760,
      "step": 164,
      "train_runtime": 454.1921,
      "train_tokens_per_second": 675.397
    },
    {
      "epoch": 0.16679302501895377,
      "grad_norm": 1.3810741901397705,
      "learning_rate": 1.838987341772152e-05,
      "loss": 0.10939119011163712,
      "num_input_tokens_seen": 308746,
      "step": 165,
      "train_runtime": 456.5098,
      "train_tokens_per_second": 676.318
    },
    {
      "epoch": 0.16780389183725045,
      "grad_norm": 1.8896007537841797,
      "learning_rate": 1.8379746835443038e-05,
      "loss": 0.11585641652345657,
      "num_input_tokens_seen": 310382,
      "step": 166,
      "train_runtime": 458.7116,
      "train_tokens_per_second": 676.639
    },
    {
      "epoch": 0.16881475865554713,
      "grad_norm": 1.8823837041854858,
      "learning_rate": 1.836962025316456e-05,
      "loss": 0.1313817948102951,
      "num_input_tokens_seen": 312192,
      "step": 167,
      "train_runtime": 460.9528,
      "train_tokens_per_second": 677.275
    },
    {
      "epoch": 0.16982562547384383,
      "grad_norm": 1.5411574840545654,
      "learning_rate": 1.8359493670886076e-05,
      "loss": 0.10811397433280945,
      "num_input_tokens_seen": 313830,
      "step": 168,
      "train_runtime": 463.162,
      "train_tokens_per_second": 677.582
    },
    {
      "epoch": 0.1708364922921405,
      "grad_norm": 1.1355396509170532,
      "learning_rate": 1.8349367088607595e-05,
      "loss": 0.08281151205301285,
      "num_input_tokens_seen": 316264,
      "step": 169,
      "train_runtime": 465.5817,
      "train_tokens_per_second": 679.288
    },
    {
      "epoch": 0.1718473591104372,
      "grad_norm": 1.4537347555160522,
      "learning_rate": 1.8339240506329115e-05,
      "loss": 0.10685627907514572,
      "num_input_tokens_seen": 318554,
      "step": 170,
      "train_runtime": 467.9284,
      "train_tokens_per_second": 680.775
    },
    {
      "epoch": 0.1728582259287339,
      "grad_norm": 1.9830877780914307,
      "learning_rate": 1.8329113924050634e-05,
      "loss": 0.12449070811271667,
      "num_input_tokens_seen": 320016,
      "step": 171,
      "train_runtime": 470.0651,
      "train_tokens_per_second": 680.791
    },
    {
      "epoch": 0.17386909274703058,
      "grad_norm": 1.317903995513916,
      "learning_rate": 1.8318987341772153e-05,
      "loss": 0.09867916256189346,
      "num_input_tokens_seen": 321488,
      "step": 172,
      "train_runtime": 472.198,
      "train_tokens_per_second": 680.833
    },
    {
      "epoch": 0.17487995956532726,
      "grad_norm": 1.537931203842163,
      "learning_rate": 1.8308860759493672e-05,
      "loss": 0.11955583095550537,
      "num_input_tokens_seen": 323296,
      "step": 173,
      "train_runtime": 474.4565,
      "train_tokens_per_second": 681.403
    },
    {
      "epoch": 0.17589082638362397,
      "grad_norm": 1.729315161705017,
      "learning_rate": 1.829873417721519e-05,
      "loss": 0.15529963374137878,
      "num_input_tokens_seen": 325238,
      "step": 174,
      "train_runtime": 476.7604,
      "train_tokens_per_second": 682.183
    },
    {
      "epoch": 0.17690169320192065,
      "grad_norm": 1.2993214130401611,
      "learning_rate": 1.828860759493671e-05,
      "loss": 0.10498125106096268,
      "num_input_tokens_seen": 327534,
      "step": 175,
      "train_runtime": 479.0961,
      "train_tokens_per_second": 683.65
    },
    {
      "epoch": 0.17791256002021733,
      "grad_norm": 1.38814115524292,
      "learning_rate": 1.827848101265823e-05,
      "loss": 0.11244213581085205,
      "num_input_tokens_seen": 329454,
      "step": 176,
      "train_runtime": 481.3543,
      "train_tokens_per_second": 684.431
    },
    {
      "epoch": 0.17892342683851403,
      "grad_norm": 1.5756162405014038,
      "learning_rate": 1.8268354430379748e-05,
      "loss": 0.11264954507350922,
      "num_input_tokens_seen": 331260,
      "step": 177,
      "train_runtime": 483.5772,
      "train_tokens_per_second": 685.02
    },
    {
      "epoch": 0.1799342936568107,
      "grad_norm": 1.7160416841506958,
      "learning_rate": 1.8258227848101267e-05,
      "loss": 0.11802274733781815,
      "num_input_tokens_seen": 333080,
      "step": 178,
      "train_runtime": 485.801,
      "train_tokens_per_second": 685.631
    },
    {
      "epoch": 0.1809451604751074,
      "grad_norm": 1.4553736448287964,
      "learning_rate": 1.8248101265822786e-05,
      "loss": 0.08927750587463379,
      "num_input_tokens_seen": 334598,
      "step": 179,
      "train_runtime": 487.9588,
      "train_tokens_per_second": 685.71
    },
    {
      "epoch": 0.1819560272934041,
      "grad_norm": 1.1470091342926025,
      "learning_rate": 1.8237974683544305e-05,
      "loss": 0.07438832521438599,
      "num_input_tokens_seen": 336020,
      "step": 180,
      "train_runtime": 490.0918,
      "train_tokens_per_second": 685.627
    },
    {
      "epoch": 0.18296689411170078,
      "grad_norm": 1.8155831098556519,
      "learning_rate": 1.8227848101265824e-05,
      "loss": 0.1461377888917923,
      "num_input_tokens_seen": 337744,
      "step": 181,
      "train_runtime": 505.6862,
      "train_tokens_per_second": 667.892
    },
    {
      "epoch": 0.18397776092999749,
      "grad_norm": 1.4382243156433105,
      "learning_rate": 1.8217721518987344e-05,
      "loss": 0.06910036504268646,
      "num_input_tokens_seen": 339166,
      "step": 182,
      "train_runtime": 507.8957,
      "train_tokens_per_second": 667.787
    },
    {
      "epoch": 0.18498862774829417,
      "grad_norm": 1.170188307762146,
      "learning_rate": 1.8207594936708863e-05,
      "loss": 0.08987250924110413,
      "num_input_tokens_seen": 341148,
      "step": 183,
      "train_runtime": 510.1967,
      "train_tokens_per_second": 668.66
    },
    {
      "epoch": 0.18599949456659085,
      "grad_norm": 1.4776676893234253,
      "learning_rate": 1.8197468354430382e-05,
      "loss": 0.10269522666931152,
      "num_input_tokens_seen": 342926,
      "step": 184,
      "train_runtime": 512.4435,
      "train_tokens_per_second": 669.198
    },
    {
      "epoch": 0.18701036138488755,
      "grad_norm": 1.3591065406799316,
      "learning_rate": 1.81873417721519e-05,
      "loss": 0.1041112095117569,
      "num_input_tokens_seen": 344884,
      "step": 185,
      "train_runtime": 514.6948,
      "train_tokens_per_second": 670.075
    },
    {
      "epoch": 0.18802122820318423,
      "grad_norm": 1.5035899877548218,
      "learning_rate": 1.817721518987342e-05,
      "loss": 0.13766233623027802,
      "num_input_tokens_seen": 346828,
      "step": 186,
      "train_runtime": 517.0104,
      "train_tokens_per_second": 670.834
    },
    {
      "epoch": 0.1890320950214809,
      "grad_norm": 1.3790515661239624,
      "learning_rate": 1.816708860759494e-05,
      "loss": 0.07921651005744934,
      "num_input_tokens_seen": 348858,
      "step": 187,
      "train_runtime": 519.4003,
      "train_tokens_per_second": 671.655
    },
    {
      "epoch": 0.19004296183977762,
      "grad_norm": 1.6024411916732788,
      "learning_rate": 1.8156962025316458e-05,
      "loss": 0.13930165767669678,
      "num_input_tokens_seen": 350544,
      "step": 188,
      "train_runtime": 521.6743,
      "train_tokens_per_second": 671.959
    },
    {
      "epoch": 0.1910538286580743,
      "grad_norm": 1.5712974071502686,
      "learning_rate": 1.8146835443037977e-05,
      "loss": 0.11272571980953217,
      "num_input_tokens_seen": 352694,
      "step": 189,
      "train_runtime": 524.0062,
      "train_tokens_per_second": 673.072
    },
    {
      "epoch": 0.19206469547637098,
      "grad_norm": 1.1387803554534912,
      "learning_rate": 1.8136708860759496e-05,
      "loss": 0.10390694439411163,
      "num_input_tokens_seen": 354786,
      "step": 190,
      "train_runtime": 526.415,
      "train_tokens_per_second": 673.966
    },
    {
      "epoch": 0.19307556229466769,
      "grad_norm": 1.6854265928268433,
      "learning_rate": 1.8126582278481012e-05,
      "loss": 0.12850573658943176,
      "num_input_tokens_seen": 356610,
      "step": 191,
      "train_runtime": 528.7417,
      "train_tokens_per_second": 674.45
    },
    {
      "epoch": 0.19408642911296436,
      "grad_norm": 1.5621191263198853,
      "learning_rate": 1.8116455696202534e-05,
      "loss": 0.14934185147285461,
      "num_input_tokens_seen": 358904,
      "step": 192,
      "train_runtime": 531.1095,
      "train_tokens_per_second": 675.763
    },
    {
      "epoch": 0.19509729593126104,
      "grad_norm": 1.9637142419815063,
      "learning_rate": 1.8106329113924053e-05,
      "loss": 0.129349023103714,
      "num_input_tokens_seen": 360522,
      "step": 193,
      "train_runtime": 533.3131,
      "train_tokens_per_second": 676.004
    },
    {
      "epoch": 0.19610816274955775,
      "grad_norm": 1.415863037109375,
      "learning_rate": 1.809620253164557e-05,
      "loss": 0.1059366837143898,
      "num_input_tokens_seen": 362594,
      "step": 194,
      "train_runtime": 535.6788,
      "train_tokens_per_second": 676.887
    },
    {
      "epoch": 0.19711902956785443,
      "grad_norm": 1.4663597345352173,
      "learning_rate": 1.808607594936709e-05,
      "loss": 0.08218872547149658,
      "num_input_tokens_seen": 364210,
      "step": 195,
      "train_runtime": 538.0065,
      "train_tokens_per_second": 676.962
    },
    {
      "epoch": 0.1981298963861511,
      "grad_norm": 1.2336485385894775,
      "learning_rate": 1.807594936708861e-05,
      "loss": 0.09393628686666489,
      "num_input_tokens_seen": 366304,
      "step": 196,
      "train_runtime": 540.3737,
      "train_tokens_per_second": 677.872
    },
    {
      "epoch": 0.19914076320444782,
      "grad_norm": 1.206937313079834,
      "learning_rate": 1.8065822784810126e-05,
      "loss": 0.0911276713013649,
      "num_input_tokens_seen": 368402,
      "step": 197,
      "train_runtime": 542.7979,
      "train_tokens_per_second": 678.709
    },
    {
      "epoch": 0.2001516300227445,
      "grad_norm": 1.638015866279602,
      "learning_rate": 1.805569620253165e-05,
      "loss": 0.116812564432621,
      "num_input_tokens_seen": 370366,
      "step": 198,
      "train_runtime": 545.1086,
      "train_tokens_per_second": 679.435
    },
    {
      "epoch": 0.2001516300227445,
      "eval_loss": 0.4045993685722351,
      "eval_runtime": 57.2181,
      "eval_samples_per_second": 15.362,
      "eval_steps_per_second": 7.69,
      "num_input_tokens_seen": 370366,
      "step": 198
    },
    {
      "epoch": 0.2011624968410412,
      "grad_norm": 1.7983155250549316,
      "learning_rate": 1.8045569620253165e-05,
      "loss": 0.10726121068000793,
      "num_input_tokens_seen": 371792,
      "step": 199,
      "train_runtime": 604.9973,
      "train_tokens_per_second": 614.535
    },
    {
      "epoch": 0.20217336365933788,
      "grad_norm": 1.305222988128662,
      "learning_rate": 1.8035443037974684e-05,
      "loss": 0.09841762483119965,
      "num_input_tokens_seen": 373630,
      "step": 200,
      "train_runtime": 607.3184,
      "train_tokens_per_second": 615.213
    },
    {
      "epoch": 0.20318423047763456,
      "grad_norm": 1.360268473625183,
      "learning_rate": 1.8025316455696206e-05,
      "loss": 0.08387979865074158,
      "num_input_tokens_seen": 375662,
      "step": 201,
      "train_runtime": 609.6146,
      "train_tokens_per_second": 616.229
    },
    {
      "epoch": 0.20419509729593127,
      "grad_norm": 1.5405970811843872,
      "learning_rate": 1.8015189873417722e-05,
      "loss": 0.08348280191421509,
      "num_input_tokens_seen": 377436,
      "step": 202,
      "train_runtime": 611.8975,
      "train_tokens_per_second": 616.829
    },
    {
      "epoch": 0.20520596411422795,
      "grad_norm": 1.6074672937393188,
      "learning_rate": 1.800506329113924e-05,
      "loss": 0.12273652851581573,
      "num_input_tokens_seen": 379004,
      "step": 203,
      "train_runtime": 614.0767,
      "train_tokens_per_second": 617.193
    },
    {
      "epoch": 0.20621683093252463,
      "grad_norm": 1.5545850992202759,
      "learning_rate": 1.7994936708860763e-05,
      "loss": 0.11406390368938446,
      "num_input_tokens_seen": 380828,
      "step": 204,
      "train_runtime": 616.3233,
      "train_tokens_per_second": 617.903
    },
    {
      "epoch": 0.20722769775082134,
      "grad_norm": 1.4091166257858276,
      "learning_rate": 1.798481012658228e-05,
      "loss": 0.08966867625713348,
      "num_input_tokens_seen": 382788,
      "step": 205,
      "train_runtime": 618.7111,
      "train_tokens_per_second": 618.686
    },
    {
      "epoch": 0.20823856456911802,
      "grad_norm": 1.5181801319122314,
      "learning_rate": 1.7974683544303798e-05,
      "loss": 0.11340916901826859,
      "num_input_tokens_seen": 384470,
      "step": 206,
      "train_runtime": 620.9427,
      "train_tokens_per_second": 619.171
    },
    {
      "epoch": 0.2092494313874147,
      "grad_norm": 1.1864792108535767,
      "learning_rate": 1.7964556962025317e-05,
      "loss": 0.08574163168668747,
      "num_input_tokens_seen": 386452,
      "step": 207,
      "train_runtime": 623.2617,
      "train_tokens_per_second": 620.048
    },
    {
      "epoch": 0.2102602982057114,
      "grad_norm": 1.295786738395691,
      "learning_rate": 1.7954430379746836e-05,
      "loss": 0.1107814610004425,
      "num_input_tokens_seen": 388504,
      "step": 208,
      "train_runtime": 625.6026,
      "train_tokens_per_second": 621.008
    },
    {
      "epoch": 0.21127116502400808,
      "grad_norm": 1.6799001693725586,
      "learning_rate": 1.7944303797468355e-05,
      "loss": 0.10416647046804428,
      "num_input_tokens_seen": 390166,
      "step": 209,
      "train_runtime": 627.8041,
      "train_tokens_per_second": 621.477
    },
    {
      "epoch": 0.21228203184230476,
      "grad_norm": 1.3828239440917969,
      "learning_rate": 1.7934177215189875e-05,
      "loss": 0.09480056911706924,
      "num_input_tokens_seen": 391552,
      "step": 210,
      "train_runtime": 630.0066,
      "train_tokens_per_second": 621.505
    },
    {
      "epoch": 0.21329289866060147,
      "grad_norm": 1.36429762840271,
      "learning_rate": 1.7924050632911394e-05,
      "loss": 0.1117706447839737,
      "num_input_tokens_seen": 393310,
      "step": 211,
      "train_runtime": 645.8096,
      "train_tokens_per_second": 609.019
    },
    {
      "epoch": 0.21430376547889815,
      "grad_norm": 1.2817749977111816,
      "learning_rate": 1.7913924050632913e-05,
      "loss": 0.09771767258644104,
      "num_input_tokens_seen": 395432,
      "step": 212,
      "train_runtime": 648.1951,
      "train_tokens_per_second": 610.051
    },
    {
      "epoch": 0.21531463229719486,
      "grad_norm": 1.4492032527923584,
      "learning_rate": 1.7903797468354432e-05,
      "loss": 0.08509109169244766,
      "num_input_tokens_seen": 397112,
      "step": 213,
      "train_runtime": 650.5053,
      "train_tokens_per_second": 610.467
    },
    {
      "epoch": 0.21632549911549154,
      "grad_norm": 1.0825918912887573,
      "learning_rate": 1.789367088607595e-05,
      "loss": 0.07626418769359589,
      "num_input_tokens_seen": 399078,
      "step": 214,
      "train_runtime": 652.8474,
      "train_tokens_per_second": 611.288
    },
    {
      "epoch": 0.21733636593378822,
      "grad_norm": 1.3212746381759644,
      "learning_rate": 1.788354430379747e-05,
      "loss": 0.09869857132434845,
      "num_input_tokens_seen": 401186,
      "step": 215,
      "train_runtime": 655.1711,
      "train_tokens_per_second": 612.338
    },
    {
      "epoch": 0.21834723275208492,
      "grad_norm": 1.7335662841796875,
      "learning_rate": 1.787341772151899e-05,
      "loss": 0.12159474939107895,
      "num_input_tokens_seen": 402874,
      "step": 216,
      "train_runtime": 657.4287,
      "train_tokens_per_second": 612.803
    },
    {
      "epoch": 0.2193580995703816,
      "grad_norm": 1.2020677328109741,
      "learning_rate": 1.7863291139240508e-05,
      "loss": 0.08844815939664841,
      "num_input_tokens_seen": 404962,
      "step": 217,
      "train_runtime": 659.7366,
      "train_tokens_per_second": 613.824
    },
    {
      "epoch": 0.22036896638867828,
      "grad_norm": 1.1301757097244263,
      "learning_rate": 1.7853164556962027e-05,
      "loss": 0.10469231754541397,
      "num_input_tokens_seen": 407460,
      "step": 218,
      "train_runtime": 662.1986,
      "train_tokens_per_second": 615.314
    },
    {
      "epoch": 0.221379833206975,
      "grad_norm": 1.5773680210113525,
      "learning_rate": 1.7843037974683546e-05,
      "loss": 0.12331892549991608,
      "num_input_tokens_seen": 409224,
      "step": 219,
      "train_runtime": 664.4252,
      "train_tokens_per_second": 615.907
    },
    {
      "epoch": 0.22239070002527167,
      "grad_norm": 1.22328782081604,
      "learning_rate": 1.7832911392405065e-05,
      "loss": 0.08160631358623505,
      "num_input_tokens_seen": 411000,
      "step": 220,
      "train_runtime": 666.6697,
      "train_tokens_per_second": 616.497
    },
    {
      "epoch": 0.22340156684356835,
      "grad_norm": 1.4057965278625488,
      "learning_rate": 1.7822784810126584e-05,
      "loss": 0.12025749683380127,
      "num_input_tokens_seen": 412552,
      "step": 221,
      "train_runtime": 668.8419,
      "train_tokens_per_second": 616.815
    },
    {
      "epoch": 0.22441243366186506,
      "grad_norm": 1.2786431312561035,
      "learning_rate": 1.7812658227848104e-05,
      "loss": 0.0865008682012558,
      "num_input_tokens_seen": 414648,
      "step": 222,
      "train_runtime": 671.1527,
      "train_tokens_per_second": 617.815
    },
    {
      "epoch": 0.22542330048016174,
      "grad_norm": 1.1525317430496216,
      "learning_rate": 1.780253164556962e-05,
      "loss": 0.08533370494842529,
      "num_input_tokens_seen": 416936,
      "step": 223,
      "train_runtime": 673.5284,
      "train_tokens_per_second": 619.033
    },
    {
      "epoch": 0.22643416729845842,
      "grad_norm": 1.500102162361145,
      "learning_rate": 1.779240506329114e-05,
      "loss": 0.09850093722343445,
      "num_input_tokens_seen": 418750,
      "step": 224,
      "train_runtime": 675.7377,
      "train_tokens_per_second": 619.693
    },
    {
      "epoch": 0.22744503411675512,
      "grad_norm": 1.3471362590789795,
      "learning_rate": 1.778227848101266e-05,
      "loss": 0.09441721439361572,
      "num_input_tokens_seen": 420940,
      "step": 225,
      "train_runtime": 678.0766,
      "train_tokens_per_second": 620.785
    },
    {
      "epoch": 0.2284559009350518,
      "grad_norm": 1.6434186697006226,
      "learning_rate": 1.7772151898734176e-05,
      "loss": 0.09287134557962418,
      "num_input_tokens_seen": 422712,
      "step": 226,
      "train_runtime": 680.3077,
      "train_tokens_per_second": 621.354
    },
    {
      "epoch": 0.22946676775334848,
      "grad_norm": 1.1667215824127197,
      "learning_rate": 1.77620253164557e-05,
      "loss": 0.10266640782356262,
      "num_input_tokens_seen": 424832,
      "step": 227,
      "train_runtime": 682.642,
      "train_tokens_per_second": 622.335
    },
    {
      "epoch": 0.2304776345716452,
      "grad_norm": 1.6905391216278076,
      "learning_rate": 1.7751898734177218e-05,
      "loss": 0.11705461144447327,
      "num_input_tokens_seen": 426646,
      "step": 228,
      "train_runtime": 684.9234,
      "train_tokens_per_second": 622.911
    },
    {
      "epoch": 0.23148850138994187,
      "grad_norm": 1.4415863752365112,
      "learning_rate": 1.7741772151898734e-05,
      "loss": 0.1196313351392746,
      "num_input_tokens_seen": 428700,
      "step": 229,
      "train_runtime": 687.2438,
      "train_tokens_per_second": 623.796
    },
    {
      "epoch": 0.23249936820823858,
      "grad_norm": 1.5781841278076172,
      "learning_rate": 1.7731645569620256e-05,
      "loss": 0.1281026005744934,
      "num_input_tokens_seen": 430232,
      "step": 230,
      "train_runtime": 689.4211,
      "train_tokens_per_second": 624.048
    },
    {
      "epoch": 0.23351023502653526,
      "grad_norm": 1.4502201080322266,
      "learning_rate": 1.7721518987341772e-05,
      "loss": 0.09619516134262085,
      "num_input_tokens_seen": 431962,
      "step": 231,
      "train_runtime": 691.6605,
      "train_tokens_per_second": 624.529
    },
    {
      "epoch": 0.23452110184483194,
      "grad_norm": 1.558029294013977,
      "learning_rate": 1.771139240506329e-05,
      "loss": 0.11640623211860657,
      "num_input_tokens_seen": 434320,
      "step": 232,
      "train_runtime": 694.0479,
      "train_tokens_per_second": 625.778
    },
    {
      "epoch": 0.23553196866312864,
      "grad_norm": 1.0241588354110718,
      "learning_rate": 1.7701265822784813e-05,
      "loss": 0.07907229661941528,
      "num_input_tokens_seen": 436184,
      "step": 233,
      "train_runtime": 696.3154,
      "train_tokens_per_second": 626.417
    },
    {
      "epoch": 0.23654283548142532,
      "grad_norm": 2.456648826599121,
      "learning_rate": 1.769113924050633e-05,
      "loss": 0.14079484343528748,
      "num_input_tokens_seen": 438052,
      "step": 234,
      "train_runtime": 698.6063,
      "train_tokens_per_second": 627.037
    },
    {
      "epoch": 0.237553702299722,
      "grad_norm": 1.2903836965560913,
      "learning_rate": 1.7681012658227848e-05,
      "loss": 0.11079635471105576,
      "num_input_tokens_seen": 440310,
      "step": 235,
      "train_runtime": 700.9765,
      "train_tokens_per_second": 628.138
    },
    {
      "epoch": 0.2385645691180187,
      "grad_norm": 1.4764586687088013,
      "learning_rate": 1.767088607594937e-05,
      "loss": 0.13691546022891998,
      "num_input_tokens_seen": 442282,
      "step": 236,
      "train_runtime": 703.2773,
      "train_tokens_per_second": 628.887
    },
    {
      "epoch": 0.2395754359363154,
      "grad_norm": 1.6779723167419434,
      "learning_rate": 1.7660759493670886e-05,
      "loss": 0.14767500758171082,
      "num_input_tokens_seen": 444134,
      "step": 237,
      "train_runtime": 705.5514,
      "train_tokens_per_second": 629.485
    },
    {
      "epoch": 0.24058630275461207,
      "grad_norm": 1.5181294679641724,
      "learning_rate": 1.7650632911392405e-05,
      "loss": 0.10068044811487198,
      "num_input_tokens_seen": 445702,
      "step": 238,
      "train_runtime": 707.7385,
      "train_tokens_per_second": 629.755
    },
    {
      "epoch": 0.24159716957290878,
      "grad_norm": 1.3846091032028198,
      "learning_rate": 1.7640506329113925e-05,
      "loss": 0.13093288242816925,
      "num_input_tokens_seen": 447646,
      "step": 239,
      "train_runtime": 710.0388,
      "train_tokens_per_second": 630.453
    },
    {
      "epoch": 0.24260803639120546,
      "grad_norm": 1.7610591650009155,
      "learning_rate": 1.7630379746835444e-05,
      "loss": 0.09435681253671646,
      "num_input_tokens_seen": 448988,
      "step": 240,
      "train_runtime": 712.1873,
      "train_tokens_per_second": 630.435
    },
    {
      "epoch": 0.24361890320950214,
      "grad_norm": 1.635029911994934,
      "learning_rate": 1.7620253164556963e-05,
      "loss": 0.09653934091329575,
      "num_input_tokens_seen": 450462,
      "step": 241,
      "train_runtime": 729.847,
      "train_tokens_per_second": 617.201
    },
    {
      "epoch": 0.24462977002779884,
      "grad_norm": 1.7021474838256836,
      "learning_rate": 1.7610126582278482e-05,
      "loss": 0.0920029804110527,
      "num_input_tokens_seen": 451978,
      "step": 242,
      "train_runtime": 732.0506,
      "train_tokens_per_second": 617.414
    },
    {
      "epoch": 0.24564063684609552,
      "grad_norm": 1.4489452838897705,
      "learning_rate": 1.76e-05,
      "loss": 0.08425816893577576,
      "num_input_tokens_seen": 453610,
      "step": 243,
      "train_runtime": 734.2579,
      "train_tokens_per_second": 617.78
    },
    {
      "epoch": 0.24665150366439223,
      "grad_norm": 1.2246116399765015,
      "learning_rate": 1.758987341772152e-05,
      "loss": 0.10387931764125824,
      "num_input_tokens_seen": 455582,
      "step": 244,
      "train_runtime": 736.5776,
      "train_tokens_per_second": 618.512
    },
    {
      "epoch": 0.2476623704826889,
      "grad_norm": 1.3314871788024902,
      "learning_rate": 1.757974683544304e-05,
      "loss": 0.07409219443798065,
      "num_input_tokens_seen": 457100,
      "step": 245,
      "train_runtime": 738.7345,
      "train_tokens_per_second": 618.761
    },
    {
      "epoch": 0.2486732373009856,
      "grad_norm": 1.466596007347107,
      "learning_rate": 1.7569620253164558e-05,
      "loss": 0.10883258283138275,
      "num_input_tokens_seen": 459110,
      "step": 246,
      "train_runtime": 741.0587,
      "train_tokens_per_second": 619.533
    },
    {
      "epoch": 0.2496841041192823,
      "grad_norm": 2.061326026916504,
      "learning_rate": 1.7559493670886077e-05,
      "loss": 0.1471419781446457,
      "num_input_tokens_seen": 460896,
      "step": 247,
      "train_runtime": 743.3057,
      "train_tokens_per_second": 620.063
    },
    {
      "epoch": 0.250694970937579,
      "grad_norm": 1.279106616973877,
      "learning_rate": 1.7549367088607596e-05,
      "loss": 0.08752479404211044,
      "num_input_tokens_seen": 462456,
      "step": 248,
      "train_runtime": 745.4907,
      "train_tokens_per_second": 620.338
    },
    {
      "epoch": 0.2517058377558757,
      "grad_norm": 1.5109342336654663,
      "learning_rate": 1.7539240506329115e-05,
      "loss": 0.09314848482608795,
      "num_input_tokens_seen": 464000,
      "step": 249,
      "train_runtime": 747.7051,
      "train_tokens_per_second": 620.566
    },
    {
      "epoch": 0.25271670457417234,
      "grad_norm": 1.7662253379821777,
      "learning_rate": 1.7529113924050634e-05,
      "loss": 0.08689254522323608,
      "num_input_tokens_seen": 465810,
      "step": 250,
      "train_runtime": 750.1167,
      "train_tokens_per_second": 620.983
    },
    {
      "epoch": 0.25372757139246904,
      "grad_norm": 1.5193723440170288,
      "learning_rate": 1.7518987341772154e-05,
      "loss": 0.11294049024581909,
      "num_input_tokens_seen": 468190,
      "step": 251,
      "train_runtime": 752.6207,
      "train_tokens_per_second": 622.08
    },
    {
      "epoch": 0.25473843821076575,
      "grad_norm": 1.3169605731964111,
      "learning_rate": 1.7508860759493673e-05,
      "loss": 0.08772353082895279,
      "num_input_tokens_seen": 470162,
      "step": 252,
      "train_runtime": 754.9573,
      "train_tokens_per_second": 622.766
    },
    {
      "epoch": 0.2557493050290624,
      "grad_norm": 1.653164029121399,
      "learning_rate": 1.7498734177215192e-05,
      "loss": 0.12575408816337585,
      "num_input_tokens_seen": 472040,
      "step": 253,
      "train_runtime": 757.2276,
      "train_tokens_per_second": 623.379
    },
    {
      "epoch": 0.2567601718473591,
      "grad_norm": 1.513001799583435,
      "learning_rate": 1.748860759493671e-05,
      "loss": 0.10402000695466995,
      "num_input_tokens_seen": 474150,
      "step": 254,
      "train_runtime": 759.5128,
      "train_tokens_per_second": 624.282
    },
    {
      "epoch": 0.2577710386656558,
      "grad_norm": 1.3591423034667969,
      "learning_rate": 1.747848101265823e-05,
      "loss": 0.08404304832220078,
      "num_input_tokens_seen": 476552,
      "step": 255,
      "train_runtime": 761.9035,
      "train_tokens_per_second": 625.476
    },
    {
      "epoch": 0.25878190548395247,
      "grad_norm": 1.7220196723937988,
      "learning_rate": 1.746835443037975e-05,
      "loss": 0.10737068206071854,
      "num_input_tokens_seen": 478316,
      "step": 256,
      "train_runtime": 764.1431,
      "train_tokens_per_second": 625.951
    },
    {
      "epoch": 0.2597927723022492,
      "grad_norm": 1.4888932704925537,
      "learning_rate": 1.7458227848101268e-05,
      "loss": 0.11215661466121674,
      "num_input_tokens_seen": 480664,
      "step": 257,
      "train_runtime": 766.5332,
      "train_tokens_per_second": 627.062
    },
    {
      "epoch": 0.2608036391205459,
      "grad_norm": 1.499002456665039,
      "learning_rate": 1.7448101265822787e-05,
      "loss": 0.09889926016330719,
      "num_input_tokens_seen": 482522,
      "step": 258,
      "train_runtime": 768.7862,
      "train_tokens_per_second": 627.641
    },
    {
      "epoch": 0.26181450593884253,
      "grad_norm": 1.3719797134399414,
      "learning_rate": 1.7437974683544306e-05,
      "loss": 0.12259243428707123,
      "num_input_tokens_seen": 484608,
      "step": 259,
      "train_runtime": 771.1057,
      "train_tokens_per_second": 628.459
    },
    {
      "epoch": 0.26282537275713924,
      "grad_norm": 1.3795220851898193,
      "learning_rate": 1.7427848101265825e-05,
      "loss": 0.10820159316062927,
      "num_input_tokens_seen": 486676,
      "step": 260,
      "train_runtime": 773.398,
      "train_tokens_per_second": 629.27
    },
    {
      "epoch": 0.26383623957543595,
      "grad_norm": 1.8952531814575195,
      "learning_rate": 1.7417721518987344e-05,
      "loss": 0.1369628757238388,
      "num_input_tokens_seen": 488252,
      "step": 261,
      "train_runtime": 775.574,
      "train_tokens_per_second": 629.536
    },
    {
      "epoch": 0.2648471063937326,
      "grad_norm": 1.487768292427063,
      "learning_rate": 1.7407594936708863e-05,
      "loss": 0.12266696989536285,
      "num_input_tokens_seen": 490056,
      "step": 262,
      "train_runtime": 777.8833,
      "train_tokens_per_second": 629.987
    },
    {
      "epoch": 0.2658579732120293,
      "grad_norm": 1.667852520942688,
      "learning_rate": 1.739746835443038e-05,
      "loss": 0.13039934635162354,
      "num_input_tokens_seen": 492050,
      "step": 263,
      "train_runtime": 780.2294,
      "train_tokens_per_second": 630.648
    },
    {
      "epoch": 0.266868840030326,
      "grad_norm": 1.686392068862915,
      "learning_rate": 1.73873417721519e-05,
      "loss": 0.09256687760353088,
      "num_input_tokens_seen": 494088,
      "step": 264,
      "train_runtime": 782.5517,
      "train_tokens_per_second": 631.381
    },
    {
      "epoch": 0.26787970684862267,
      "grad_norm": 1.3775935173034668,
      "learning_rate": 1.737721518987342e-05,
      "loss": 0.13270871341228485,
      "num_input_tokens_seen": 496250,
      "step": 265,
      "train_runtime": 784.8891,
      "train_tokens_per_second": 632.255
    },
    {
      "epoch": 0.2688905736669194,
      "grad_norm": 1.4882550239562988,
      "learning_rate": 1.7367088607594936e-05,
      "loss": 0.11100231111049652,
      "num_input_tokens_seen": 498554,
      "step": 266,
      "train_runtime": 787.3206,
      "train_tokens_per_second": 633.229
    },
    {
      "epoch": 0.2699014404852161,
      "grad_norm": 1.6049178838729858,
      "learning_rate": 1.735696202531646e-05,
      "loss": 0.10452968627214432,
      "num_input_tokens_seen": 500336,
      "step": 267,
      "train_runtime": 789.5692,
      "train_tokens_per_second": 633.682
    },
    {
      "epoch": 0.2709123073035128,
      "grad_norm": 1.414746880531311,
      "learning_rate": 1.7346835443037978e-05,
      "loss": 0.11845418810844421,
      "num_input_tokens_seen": 502388,
      "step": 268,
      "train_runtime": 791.8439,
      "train_tokens_per_second": 634.453
    },
    {
      "epoch": 0.27192317412180944,
      "grad_norm": 1.485299825668335,
      "learning_rate": 1.7336708860759494e-05,
      "loss": 0.08825024962425232,
      "num_input_tokens_seen": 504016,
      "step": 269,
      "train_runtime": 794.1155,
      "train_tokens_per_second": 634.689
    },
    {
      "epoch": 0.27293404094010615,
      "grad_norm": 1.4826604127883911,
      "learning_rate": 1.7326582278481016e-05,
      "loss": 0.13610872626304626,
      "num_input_tokens_seen": 506228,
      "step": 270,
      "train_runtime": 796.5551,
      "train_tokens_per_second": 635.522
    },
    {
      "epoch": 0.27394490775840286,
      "grad_norm": 1.5785086154937744,
      "learning_rate": 1.7316455696202532e-05,
      "loss": 0.09215106070041656,
      "num_input_tokens_seen": 507924,
      "step": 271,
      "train_runtime": 814.0631,
      "train_tokens_per_second": 623.937
    },
    {
      "epoch": 0.2749557745766995,
      "grad_norm": 1.530111312866211,
      "learning_rate": 1.730632911392405e-05,
      "loss": 0.10139001160860062,
      "num_input_tokens_seen": 509700,
      "step": 272,
      "train_runtime": 816.4309,
      "train_tokens_per_second": 624.303
    },
    {
      "epoch": 0.2759666413949962,
      "grad_norm": 1.2713271379470825,
      "learning_rate": 1.729620253164557e-05,
      "loss": 0.1057865172624588,
      "num_input_tokens_seen": 511482,
      "step": 273,
      "train_runtime": 818.8016,
      "train_tokens_per_second": 624.671
    },
    {
      "epoch": 0.2769775082132929,
      "grad_norm": 1.2762155532836914,
      "learning_rate": 1.728607594936709e-05,
      "loss": 0.10544420778751373,
      "num_input_tokens_seen": 513976,
      "step": 274,
      "train_runtime": 821.2782,
      "train_tokens_per_second": 625.824
    },
    {
      "epoch": 0.2779883750315896,
      "grad_norm": 1.5279234647750854,
      "learning_rate": 1.7275949367088608e-05,
      "loss": 0.1072789803147316,
      "num_input_tokens_seen": 515382,
      "step": 275,
      "train_runtime": 823.4179,
      "train_tokens_per_second": 625.906
    },
    {
      "epoch": 0.2789992418498863,
      "grad_norm": 1.2843568325042725,
      "learning_rate": 1.7265822784810127e-05,
      "loss": 0.08076569437980652,
      "num_input_tokens_seen": 517038,
      "step": 276,
      "train_runtime": 825.793,
      "train_tokens_per_second": 626.111
    },
    {
      "epoch": 0.280010108668183,
      "grad_norm": 1.492134928703308,
      "learning_rate": 1.7255696202531646e-05,
      "loss": 0.11192678660154343,
      "num_input_tokens_seen": 519080,
      "step": 277,
      "train_runtime": 828.1851,
      "train_tokens_per_second": 626.768
    },
    {
      "epoch": 0.28102097548647964,
      "grad_norm": 1.2823917865753174,
      "learning_rate": 1.7245569620253165e-05,
      "loss": 0.11176803708076477,
      "num_input_tokens_seen": 520804,
      "step": 278,
      "train_runtime": 830.4401,
      "train_tokens_per_second": 627.142
    },
    {
      "epoch": 0.28203184230477635,
      "grad_norm": 1.2782809734344482,
      "learning_rate": 1.7235443037974685e-05,
      "loss": 0.11687865853309631,
      "num_input_tokens_seen": 523082,
      "step": 279,
      "train_runtime": 832.7674,
      "train_tokens_per_second": 628.125
    },
    {
      "epoch": 0.28304270912307306,
      "grad_norm": 1.5529136657714844,
      "learning_rate": 1.7225316455696204e-05,
      "loss": 0.11955727636814117,
      "num_input_tokens_seen": 524988,
      "step": 280,
      "train_runtime": 835.0226,
      "train_tokens_per_second": 628.711
    },
    {
      "epoch": 0.2840535759413697,
      "grad_norm": 2.094829797744751,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.09432722628116608,
      "num_input_tokens_seen": 526770,
      "step": 281,
      "train_runtime": 837.2395,
      "train_tokens_per_second": 629.175
    },
    {
      "epoch": 0.2850644427596664,
      "grad_norm": 1.9160499572753906,
      "learning_rate": 1.7205063291139242e-05,
      "loss": 0.09831471741199493,
      "num_input_tokens_seen": 528400,
      "step": 282,
      "train_runtime": 839.4927,
      "train_tokens_per_second": 629.428
    },
    {
      "epoch": 0.2860753095779631,
      "grad_norm": 1.2781708240509033,
      "learning_rate": 1.719493670886076e-05,
      "loss": 0.09331564605236053,
      "num_input_tokens_seen": 530224,
      "step": 283,
      "train_runtime": 841.7166,
      "train_tokens_per_second": 629.932
    },
    {
      "epoch": 0.2870861763962598,
      "grad_norm": 1.4574735164642334,
      "learning_rate": 1.718481012658228e-05,
      "loss": 0.10699676722288132,
      "num_input_tokens_seen": 532286,
      "step": 284,
      "train_runtime": 844.0128,
      "train_tokens_per_second": 630.661
    },
    {
      "epoch": 0.2880970432145565,
      "grad_norm": 1.2516075372695923,
      "learning_rate": 1.71746835443038e-05,
      "loss": 0.0777752548456192,
      "num_input_tokens_seen": 534178,
      "step": 285,
      "train_runtime": 846.5658,
      "train_tokens_per_second": 630.994
    },
    {
      "epoch": 0.2891079100328532,
      "grad_norm": 1.3657443523406982,
      "learning_rate": 1.7164556962025318e-05,
      "loss": 0.10243025422096252,
      "num_input_tokens_seen": 535870,
      "step": 286,
      "train_runtime": 848.7713,
      "train_tokens_per_second": 631.348
    },
    {
      "epoch": 0.29011877685114984,
      "grad_norm": 1.4470164775848389,
      "learning_rate": 1.7154430379746837e-05,
      "loss": 0.11428594589233398,
      "num_input_tokens_seen": 537938,
      "step": 287,
      "train_runtime": 851.0535,
      "train_tokens_per_second": 632.085
    },
    {
      "epoch": 0.29112964366944655,
      "grad_norm": 1.3402222394943237,
      "learning_rate": 1.7144303797468356e-05,
      "loss": 0.09148528426885605,
      "num_input_tokens_seen": 540342,
      "step": 288,
      "train_runtime": 853.4611,
      "train_tokens_per_second": 633.118
    },
    {
      "epoch": 0.29214051048774325,
      "grad_norm": 1.4701645374298096,
      "learning_rate": 1.7134177215189875e-05,
      "loss": 0.10003398358821869,
      "num_input_tokens_seen": 542224,
      "step": 289,
      "train_runtime": 855.7411,
      "train_tokens_per_second": 633.631
    },
    {
      "epoch": 0.2931513773060399,
      "grad_norm": 1.7833036184310913,
      "learning_rate": 1.7124050632911394e-05,
      "loss": 0.10028425604104996,
      "num_input_tokens_seen": 544196,
      "step": 290,
      "train_runtime": 858.0045,
      "train_tokens_per_second": 634.258
    },
    {
      "epoch": 0.2941622441243366,
      "grad_norm": 1.3784419298171997,
      "learning_rate": 1.7113924050632914e-05,
      "loss": 0.09718849509954453,
      "num_input_tokens_seen": 546366,
      "step": 291,
      "train_runtime": 860.3662,
      "train_tokens_per_second": 635.039
    },
    {
      "epoch": 0.2951731109426333,
      "grad_norm": 1.7228463888168335,
      "learning_rate": 1.7103797468354433e-05,
      "loss": 0.09118913114070892,
      "num_input_tokens_seen": 548322,
      "step": 292,
      "train_runtime": 862.6776,
      "train_tokens_per_second": 635.605
    },
    {
      "epoch": 0.29618397776093,
      "grad_norm": 1.278310775756836,
      "learning_rate": 1.7093670886075952e-05,
      "loss": 0.08900696039199829,
      "num_input_tokens_seen": 550206,
      "step": 293,
      "train_runtime": 864.9696,
      "train_tokens_per_second": 636.099
    },
    {
      "epoch": 0.2971948445792267,
      "grad_norm": 1.6270031929016113,
      "learning_rate": 1.708354430379747e-05,
      "loss": 0.09637334942817688,
      "num_input_tokens_seen": 552534,
      "step": 294,
      "train_runtime": 867.3358,
      "train_tokens_per_second": 637.047
    },
    {
      "epoch": 0.2982057113975234,
      "grad_norm": 1.4598493576049805,
      "learning_rate": 1.7073417721518987e-05,
      "loss": 0.13069799542427063,
      "num_input_tokens_seen": 554498,
      "step": 295,
      "train_runtime": 869.6432,
      "train_tokens_per_second": 637.615
    },
    {
      "epoch": 0.29921657821582004,
      "grad_norm": 1.857382893562317,
      "learning_rate": 1.706329113924051e-05,
      "loss": 0.08895819634199142,
      "num_input_tokens_seen": 555778,
      "step": 296,
      "train_runtime": 871.7636,
      "train_tokens_per_second": 637.533
    },
    {
      "epoch": 0.30022744503411675,
      "grad_norm": 1.45383882522583,
      "learning_rate": 1.7053164556962028e-05,
      "loss": 0.08400879800319672,
      "num_input_tokens_seen": 557814,
      "step": 297,
      "train_runtime": 874.0642,
      "train_tokens_per_second": 638.184
    },
    {
      "epoch": 0.30123831185241345,
      "grad_norm": 1.2739516496658325,
      "learning_rate": 1.7043037974683544e-05,
      "loss": 0.09343541413545609,
      "num_input_tokens_seen": 560236,
      "step": 298,
      "train_runtime": 876.4843,
      "train_tokens_per_second": 639.185
    },
    {
      "epoch": 0.30224917867071016,
      "grad_norm": 1.4769498109817505,
      "learning_rate": 1.7032911392405066e-05,
      "loss": 0.0803271234035492,
      "num_input_tokens_seen": 561910,
      "step": 299,
      "train_runtime": 878.688,
      "train_tokens_per_second": 639.488
    },
    {
      "epoch": 0.3032600454890068,
      "grad_norm": 1.2992624044418335,
      "learning_rate": 1.7022784810126585e-05,
      "loss": 0.07486756145954132,
      "num_input_tokens_seen": 563520,
      "step": 300,
      "train_runtime": 880.8268,
      "train_tokens_per_second": 639.763
    },
    {
      "epoch": 0.3042709123073035,
      "grad_norm": 1.409480333328247,
      "learning_rate": 1.70126582278481e-05,
      "loss": 0.12226992845535278,
      "num_input_tokens_seen": 565432,
      "step": 301,
      "train_runtime": 897.9704,
      "train_tokens_per_second": 629.678
    },
    {
      "epoch": 0.30528177912560023,
      "grad_norm": 1.3607956171035767,
      "learning_rate": 1.7002531645569623e-05,
      "loss": 0.08789639174938202,
      "num_input_tokens_seen": 567426,
      "step": 302,
      "train_runtime": 900.2444,
      "train_tokens_per_second": 630.302
    },
    {
      "epoch": 0.3062926459438969,
      "grad_norm": 1.2299937009811401,
      "learning_rate": 1.699240506329114e-05,
      "loss": 0.08111349493265152,
      "num_input_tokens_seen": 569566,
      "step": 303,
      "train_runtime": 902.5401,
      "train_tokens_per_second": 631.07
    },
    {
      "epoch": 0.3073035127621936,
      "grad_norm": 1.6185789108276367,
      "learning_rate": 1.6982278481012658e-05,
      "loss": 0.09807392954826355,
      "num_input_tokens_seen": 570788,
      "step": 304,
      "train_runtime": 904.614,
      "train_tokens_per_second": 630.974
    },
    {
      "epoch": 0.3083143795804903,
      "grad_norm": 1.7131162881851196,
      "learning_rate": 1.697215189873418e-05,
      "loss": 0.12074774503707886,
      "num_input_tokens_seen": 572294,
      "step": 305,
      "train_runtime": 906.7949,
      "train_tokens_per_second": 631.117
    },
    {
      "epoch": 0.30932524639878695,
      "grad_norm": 1.5033224821090698,
      "learning_rate": 1.6962025316455696e-05,
      "loss": 0.13393236696720123,
      "num_input_tokens_seen": 574604,
      "step": 306,
      "train_runtime": 909.1603,
      "train_tokens_per_second": 632.016
    },
    {
      "epoch": 0.31033611321708365,
      "grad_norm": 1.6546261310577393,
      "learning_rate": 1.6951898734177216e-05,
      "loss": 0.10731075704097748,
      "num_input_tokens_seen": 576448,
      "step": 307,
      "train_runtime": 911.393,
      "train_tokens_per_second": 632.491
    },
    {
      "epoch": 0.31134698003538036,
      "grad_norm": 1.6043623685836792,
      "learning_rate": 1.6941772151898738e-05,
      "loss": 0.10167066007852554,
      "num_input_tokens_seen": 578256,
      "step": 308,
      "train_runtime": 913.6297,
      "train_tokens_per_second": 632.922
    },
    {
      "epoch": 0.312357846853677,
      "grad_norm": 1.7753088474273682,
      "learning_rate": 1.6931645569620254e-05,
      "loss": 0.13217665255069733,
      "num_input_tokens_seen": 580722,
      "step": 309,
      "train_runtime": 916.0141,
      "train_tokens_per_second": 633.966
    },
    {
      "epoch": 0.3133687136719737,
      "grad_norm": 1.5941412448883057,
      "learning_rate": 1.6921518987341773e-05,
      "loss": 0.10651037096977234,
      "num_input_tokens_seen": 582702,
      "step": 310,
      "train_runtime": 918.317,
      "train_tokens_per_second": 634.533
    },
    {
      "epoch": 0.3143795804902704,
      "grad_norm": 1.6881039142608643,
      "learning_rate": 1.6911392405063292e-05,
      "loss": 0.13192051649093628,
      "num_input_tokens_seen": 584540,
      "step": 311,
      "train_runtime": 920.6956,
      "train_tokens_per_second": 634.89
    },
    {
      "epoch": 0.3153904473085671,
      "grad_norm": 1.2865118980407715,
      "learning_rate": 1.690126582278481e-05,
      "loss": 0.09769687056541443,
      "num_input_tokens_seen": 586196,
      "step": 312,
      "train_runtime": 922.9055,
      "train_tokens_per_second": 635.164
    },
    {
      "epoch": 0.3164013141268638,
      "grad_norm": 1.3329963684082031,
      "learning_rate": 1.689113924050633e-05,
      "loss": 0.08811003714799881,
      "num_input_tokens_seen": 588350,
      "step": 313,
      "train_runtime": 925.2492,
      "train_tokens_per_second": 635.883
    },
    {
      "epoch": 0.3174121809451605,
      "grad_norm": 1.3556103706359863,
      "learning_rate": 1.688101265822785e-05,
      "loss": 0.10852400213479996,
      "num_input_tokens_seen": 590494,
      "step": 314,
      "train_runtime": 927.5584,
      "train_tokens_per_second": 636.611
    },
    {
      "epoch": 0.31842304776345715,
      "grad_norm": 1.3499561548233032,
      "learning_rate": 1.6870886075949368e-05,
      "loss": 0.08952528238296509,
      "num_input_tokens_seen": 592296,
      "step": 315,
      "train_runtime": 929.8427,
      "train_tokens_per_second": 636.985
    },
    {
      "epoch": 0.31943391458175385,
      "grad_norm": 1.6870841979980469,
      "learning_rate": 1.6860759493670887e-05,
      "loss": 0.14597761631011963,
      "num_input_tokens_seen": 594442,
      "step": 316,
      "train_runtime": 932.2967,
      "train_tokens_per_second": 637.61
    },
    {
      "epoch": 0.32044478140005056,
      "grad_norm": 1.2874904870986938,
      "learning_rate": 1.6850632911392406e-05,
      "loss": 0.10051897168159485,
      "num_input_tokens_seen": 596700,
      "step": 317,
      "train_runtime": 934.7422,
      "train_tokens_per_second": 638.358
    },
    {
      "epoch": 0.3214556482183472,
      "grad_norm": 1.3767913579940796,
      "learning_rate": 1.6840506329113925e-05,
      "loss": 0.1061960756778717,
      "num_input_tokens_seen": 598890,
      "step": 318,
      "train_runtime": 937.2301,
      "train_tokens_per_second": 639.0
    },
    {
      "epoch": 0.3224665150366439,
      "grad_norm": 1.6041687726974487,
      "learning_rate": 1.6830379746835445e-05,
      "loss": 0.10163749754428864,
      "num_input_tokens_seen": 600748,
      "step": 319,
      "train_runtime": 939.6168,
      "train_tokens_per_second": 639.354
    },
    {
      "epoch": 0.3234773818549406,
      "grad_norm": 1.4598690271377563,
      "learning_rate": 1.6820253164556964e-05,
      "loss": 0.09881944954395294,
      "num_input_tokens_seen": 602598,
      "step": 320,
      "train_runtime": 941.9622,
      "train_tokens_per_second": 639.726
    },
    {
      "epoch": 0.3244882486732373,
      "grad_norm": 1.4148222208023071,
      "learning_rate": 1.6810126582278483e-05,
      "loss": 0.0919995978474617,
      "num_input_tokens_seen": 604068,
      "step": 321,
      "train_runtime": 944.1339,
      "train_tokens_per_second": 639.812
    },
    {
      "epoch": 0.325499115491534,
      "grad_norm": 1.1756342649459839,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 0.0852050930261612,
      "num_input_tokens_seen": 605770,
      "step": 322,
      "train_runtime": 946.385,
      "train_tokens_per_second": 640.088
    },
    {
      "epoch": 0.3265099823098307,
      "grad_norm": 1.5971572399139404,
      "learning_rate": 1.678987341772152e-05,
      "loss": 0.11823036521673203,
      "num_input_tokens_seen": 607382,
      "step": 323,
      "train_runtime": 948.592,
      "train_tokens_per_second": 640.298
    },
    {
      "epoch": 0.32752084912812734,
      "grad_norm": 1.3547089099884033,
      "learning_rate": 1.677974683544304e-05,
      "loss": 0.11181734502315521,
      "num_input_tokens_seen": 609506,
      "step": 324,
      "train_runtime": 950.9562,
      "train_tokens_per_second": 640.94
    },
    {
      "epoch": 0.32853171594642405,
      "grad_norm": 1.2220993041992188,
      "learning_rate": 1.676962025316456e-05,
      "loss": 0.0703224316239357,
      "num_input_tokens_seen": 611400,
      "step": 325,
      "train_runtime": 953.2391,
      "train_tokens_per_second": 641.392
    },
    {
      "epoch": 0.32954258276472076,
      "grad_norm": 1.8233991861343384,
      "learning_rate": 1.6759493670886078e-05,
      "loss": 0.11083690822124481,
      "num_input_tokens_seen": 612836,
      "step": 326,
      "train_runtime": 955.3964,
      "train_tokens_per_second": 641.447
    },
    {
      "epoch": 0.3305534495830174,
      "grad_norm": 1.5295997858047485,
      "learning_rate": 1.6749367088607594e-05,
      "loss": 0.1028873473405838,
      "num_input_tokens_seen": 614146,
      "step": 327,
      "train_runtime": 957.5511,
      "train_tokens_per_second": 641.371
    },
    {
      "epoch": 0.3315643164013141,
      "grad_norm": 2.0421392917633057,
      "learning_rate": 1.6739240506329116e-05,
      "loss": 0.08458258956670761,
      "num_input_tokens_seen": 615256,
      "step": 328,
      "train_runtime": 959.6481,
      "train_tokens_per_second": 641.127
    },
    {
      "epoch": 0.3325751832196108,
      "grad_norm": 1.3370336294174194,
      "learning_rate": 1.6729113924050635e-05,
      "loss": 0.08434239029884338,
      "num_input_tokens_seen": 617376,
      "step": 329,
      "train_runtime": 961.9827,
      "train_tokens_per_second": 641.775
    },
    {
      "epoch": 0.33358605003790753,
      "grad_norm": 1.5026538372039795,
      "learning_rate": 1.671898734177215e-05,
      "loss": 0.12171722948551178,
      "num_input_tokens_seen": 619276,
      "step": 330,
      "train_runtime": 964.2576,
      "train_tokens_per_second": 642.231
    },
    {
      "epoch": 0.3345969168562042,
      "grad_norm": 1.5519927740097046,
      "learning_rate": 1.6708860759493674e-05,
      "loss": 0.11145951598882675,
      "num_input_tokens_seen": 621170,
      "step": 331,
      "train_runtime": 981.4765,
      "train_tokens_per_second": 632.893
    },
    {
      "epoch": 0.3356077836745009,
      "grad_norm": 1.5354214906692505,
      "learning_rate": 1.6698734177215193e-05,
      "loss": 0.1101895347237587,
      "num_input_tokens_seen": 623702,
      "step": 332,
      "train_runtime": 983.9355,
      "train_tokens_per_second": 633.885
    },
    {
      "epoch": 0.3366186504927976,
      "grad_norm": 1.3286933898925781,
      "learning_rate": 1.668860759493671e-05,
      "loss": 0.07348203659057617,
      "num_input_tokens_seen": 625398,
      "step": 333,
      "train_runtime": 986.1737,
      "train_tokens_per_second": 634.166
    },
    {
      "epoch": 0.33762951731109425,
      "grad_norm": 1.9805306196212769,
      "learning_rate": 1.667848101265823e-05,
      "loss": 0.09451054036617279,
      "num_input_tokens_seen": 627014,
      "step": 334,
      "train_runtime": 988.3484,
      "train_tokens_per_second": 634.406
    },
    {
      "epoch": 0.33864038412939096,
      "grad_norm": 1.3444042205810547,
      "learning_rate": 1.6668354430379746e-05,
      "loss": 0.09824031591415405,
      "num_input_tokens_seen": 628980,
      "step": 335,
      "train_runtime": 990.6394,
      "train_tokens_per_second": 634.923
    },
    {
      "epoch": 0.33965125094768767,
      "grad_norm": 1.9391517639160156,
      "learning_rate": 1.6658227848101266e-05,
      "loss": 0.11450466513633728,
      "num_input_tokens_seen": 630750,
      "step": 336,
      "train_runtime": 992.9136,
      "train_tokens_per_second": 635.252
    },
    {
      "epoch": 0.3406621177659843,
      "grad_norm": 1.6187993288040161,
      "learning_rate": 1.6648101265822788e-05,
      "loss": 0.12155002355575562,
      "num_input_tokens_seen": 632848,
      "step": 337,
      "train_runtime": 995.2466,
      "train_tokens_per_second": 635.871
    },
    {
      "epoch": 0.341672984584281,
      "grad_norm": 1.7981492280960083,
      "learning_rate": 1.6637974683544304e-05,
      "loss": 0.13006402552127838,
      "num_input_tokens_seen": 634710,
      "step": 338,
      "train_runtime": 997.5703,
      "train_tokens_per_second": 636.256
    },
    {
      "epoch": 0.34268385140257773,
      "grad_norm": 1.4146748781204224,
      "learning_rate": 1.6627848101265823e-05,
      "loss": 0.07810388505458832,
      "num_input_tokens_seen": 636400,
      "step": 339,
      "train_runtime": 999.8783,
      "train_tokens_per_second": 636.477
    },
    {
      "epoch": 0.3436947182208744,
      "grad_norm": 1.3834339380264282,
      "learning_rate": 1.6617721518987345e-05,
      "loss": 0.09099924564361572,
      "num_input_tokens_seen": 638424,
      "step": 340,
      "train_runtime": 1002.2323,
      "train_tokens_per_second": 637.002
    },
    {
      "epoch": 0.3447055850391711,
      "grad_norm": 1.5695958137512207,
      "learning_rate": 1.660759493670886e-05,
      "loss": 0.10461483895778656,
      "num_input_tokens_seen": 639900,
      "step": 341,
      "train_runtime": 1004.4667,
      "train_tokens_per_second": 637.054
    },
    {
      "epoch": 0.3457164518574678,
      "grad_norm": 1.3106541633605957,
      "learning_rate": 1.659746835443038e-05,
      "loss": 0.10517305135726929,
      "num_input_tokens_seen": 641800,
      "step": 342,
      "train_runtime": 1006.7328,
      "train_tokens_per_second": 637.508
    },
    {
      "epoch": 0.34672731867576445,
      "grad_norm": 1.5017896890640259,
      "learning_rate": 1.65873417721519e-05,
      "loss": 0.11018554121255875,
      "num_input_tokens_seen": 644028,
      "step": 343,
      "train_runtime": 1009.0752,
      "train_tokens_per_second": 638.236
    },
    {
      "epoch": 0.34773818549406116,
      "grad_norm": 1.2032376527786255,
      "learning_rate": 1.6577215189873418e-05,
      "loss": 0.08843756467103958,
      "num_input_tokens_seen": 645904,
      "step": 344,
      "train_runtime": 1011.346,
      "train_tokens_per_second": 638.658
    },
    {
      "epoch": 0.34874905231235787,
      "grad_norm": 1.1695470809936523,
      "learning_rate": 1.6567088607594937e-05,
      "loss": 0.06325731426477432,
      "num_input_tokens_seen": 647174,
      "step": 345,
      "train_runtime": 1013.4667,
      "train_tokens_per_second": 638.574
    },
    {
      "epoch": 0.3497599191306545,
      "grad_norm": 1.8199679851531982,
      "learning_rate": 1.6556962025316456e-05,
      "loss": 0.12382033467292786,
      "num_input_tokens_seen": 648866,
      "step": 346,
      "train_runtime": 1015.6887,
      "train_tokens_per_second": 638.843
    },
    {
      "epoch": 0.3507707859489512,
      "grad_norm": 1.3210959434509277,
      "learning_rate": 1.6546835443037976e-05,
      "loss": 0.10739161819219589,
      "num_input_tokens_seen": 651554,
      "step": 347,
      "train_runtime": 1018.2033,
      "train_tokens_per_second": 639.906
    },
    {
      "epoch": 0.35178165276724793,
      "grad_norm": 1.9572030305862427,
      "learning_rate": 1.6536708860759495e-05,
      "loss": 0.08785548061132431,
      "num_input_tokens_seen": 652762,
      "step": 348,
      "train_runtime": 1020.3918,
      "train_tokens_per_second": 639.717
    },
    {
      "epoch": 0.3527925195855446,
      "grad_norm": 1.2821578979492188,
      "learning_rate": 1.6526582278481014e-05,
      "loss": 0.07090479135513306,
      "num_input_tokens_seen": 654322,
      "step": 349,
      "train_runtime": 1022.5927,
      "train_tokens_per_second": 639.866
    },
    {
      "epoch": 0.3538033864038413,
      "grad_norm": 1.4195822477340698,
      "learning_rate": 1.6516455696202533e-05,
      "loss": 0.08603788167238235,
      "num_input_tokens_seen": 656156,
      "step": 350,
      "train_runtime": 1024.8406,
      "train_tokens_per_second": 640.252
    },
    {
      "epoch": 0.354814253222138,
      "grad_norm": 1.4972152709960938,
      "learning_rate": 1.6506329113924052e-05,
      "loss": 0.09922720491886139,
      "num_input_tokens_seen": 657980,
      "step": 351,
      "train_runtime": 1027.1208,
      "train_tokens_per_second": 640.606
    },
    {
      "epoch": 0.35582512004043465,
      "grad_norm": 1.589414358139038,
      "learning_rate": 1.649620253164557e-05,
      "loss": 0.11485034972429276,
      "num_input_tokens_seen": 659612,
      "step": 352,
      "train_runtime": 1029.33,
      "train_tokens_per_second": 640.817
    },
    {
      "epoch": 0.35683598685873136,
      "grad_norm": 1.634354829788208,
      "learning_rate": 1.648607594936709e-05,
      "loss": 0.11900883913040161,
      "num_input_tokens_seen": 661560,
      "step": 353,
      "train_runtime": 1031.6102,
      "train_tokens_per_second": 641.289
    },
    {
      "epoch": 0.35784685367702807,
      "grad_norm": 1.4277557134628296,
      "learning_rate": 1.647594936708861e-05,
      "loss": 0.09803351014852524,
      "num_input_tokens_seen": 663524,
      "step": 354,
      "train_runtime": 1033.9392,
      "train_tokens_per_second": 641.744
    },
    {
      "epoch": 0.3588577204953247,
      "grad_norm": 1.006697654724121,
      "learning_rate": 1.6465822784810128e-05,
      "loss": 0.07800620794296265,
      "num_input_tokens_seen": 666486,
      "step": 355,
      "train_runtime": 1036.5554,
      "train_tokens_per_second": 642.982
    },
    {
      "epoch": 0.3598685873136214,
      "grad_norm": 1.6165040731430054,
      "learning_rate": 1.6455696202531647e-05,
      "loss": 0.07517444342374802,
      "num_input_tokens_seen": 667976,
      "step": 356,
      "train_runtime": 1038.7334,
      "train_tokens_per_second": 643.068
    },
    {
      "epoch": 0.36087945413191813,
      "grad_norm": 1.2826602458953857,
      "learning_rate": 1.6445569620253166e-05,
      "loss": 0.09816960990428925,
      "num_input_tokens_seen": 670104,
      "step": 357,
      "train_runtime": 1041.1268,
      "train_tokens_per_second": 643.633
    },
    {
      "epoch": 0.3618903209502148,
      "grad_norm": 1.544905424118042,
      "learning_rate": 1.6435443037974685e-05,
      "loss": 0.07951736450195312,
      "num_input_tokens_seen": 671812,
      "step": 358,
      "train_runtime": 1043.3338,
      "train_tokens_per_second": 643.909
    },
    {
      "epoch": 0.3629011877685115,
      "grad_norm": 1.5321018695831299,
      "learning_rate": 1.6425316455696205e-05,
      "loss": 0.1257629692554474,
      "num_input_tokens_seen": 673840,
      "step": 359,
      "train_runtime": 1045.6789,
      "train_tokens_per_second": 644.404
    },
    {
      "epoch": 0.3639120545868082,
      "grad_norm": 1.5514605045318604,
      "learning_rate": 1.6415189873417724e-05,
      "loss": 0.10741757601499557,
      "num_input_tokens_seen": 675820,
      "step": 360,
      "train_runtime": 1048.0645,
      "train_tokens_per_second": 644.827
    },
    {
      "epoch": 0.36492292140510485,
      "grad_norm": 1.1789236068725586,
      "learning_rate": 1.6405063291139243e-05,
      "loss": 0.07961886376142502,
      "num_input_tokens_seen": 677590,
      "step": 361,
      "train_runtime": 1065.5145,
      "train_tokens_per_second": 635.928
    },
    {
      "epoch": 0.36593378822340156,
      "grad_norm": 1.5250983238220215,
      "learning_rate": 1.6394936708860762e-05,
      "loss": 0.1249462366104126,
      "num_input_tokens_seen": 679656,
      "step": 362,
      "train_runtime": 1067.9822,
      "train_tokens_per_second": 636.393
    },
    {
      "epoch": 0.36694465504169826,
      "grad_norm": 1.3443119525909424,
      "learning_rate": 1.638481012658228e-05,
      "loss": 0.11207073926925659,
      "num_input_tokens_seen": 681542,
      "step": 363,
      "train_runtime": 1070.3137,
      "train_tokens_per_second": 636.768
    },
    {
      "epoch": 0.36795552185999497,
      "grad_norm": 1.2133140563964844,
      "learning_rate": 1.63746835443038e-05,
      "loss": 0.08309883624315262,
      "num_input_tokens_seen": 683072,
      "step": 364,
      "train_runtime": 1072.6198,
      "train_tokens_per_second": 636.826
    },
    {
      "epoch": 0.3689663886782916,
      "grad_norm": 1.3501054048538208,
      "learning_rate": 1.636455696202532e-05,
      "loss": 0.10249121487140656,
      "num_input_tokens_seen": 684780,
      "step": 365,
      "train_runtime": 1074.8369,
      "train_tokens_per_second": 637.101
    },
    {
      "epoch": 0.36997725549658833,
      "grad_norm": 1.3758456707000732,
      "learning_rate": 1.6354430379746838e-05,
      "loss": 0.13326986134052277,
      "num_input_tokens_seen": 687052,
      "step": 366,
      "train_runtime": 1077.2431,
      "train_tokens_per_second": 637.787
    },
    {
      "epoch": 0.37098812231488504,
      "grad_norm": 1.2025797367095947,
      "learning_rate": 1.6344303797468354e-05,
      "loss": 0.11053229868412018,
      "num_input_tokens_seen": 689058,
      "step": 367,
      "train_runtime": 1079.5103,
      "train_tokens_per_second": 638.306
    },
    {
      "epoch": 0.3719989891331817,
      "grad_norm": 1.3191109895706177,
      "learning_rate": 1.6334177215189876e-05,
      "loss": 0.1132110059261322,
      "num_input_tokens_seen": 690736,
      "step": 368,
      "train_runtime": 1081.7273,
      "train_tokens_per_second": 638.549
    },
    {
      "epoch": 0.3730098559514784,
      "grad_norm": 1.3614546060562134,
      "learning_rate": 1.6324050632911395e-05,
      "loss": 0.09920522570610046,
      "num_input_tokens_seen": 692596,
      "step": 369,
      "train_runtime": 1083.9848,
      "train_tokens_per_second": 638.935
    },
    {
      "epoch": 0.3740207227697751,
      "grad_norm": 1.4268485307693481,
      "learning_rate": 1.631392405063291e-05,
      "loss": 0.11578072607517242,
      "num_input_tokens_seen": 694712,
      "step": 370,
      "train_runtime": 1086.2939,
      "train_tokens_per_second": 639.525
    },
    {
      "epoch": 0.37503158958807176,
      "grad_norm": 1.1919304132461548,
      "learning_rate": 1.6303797468354434e-05,
      "loss": 0.07230079919099808,
      "num_input_tokens_seen": 696466,
      "step": 371,
      "train_runtime": 1088.5269,
      "train_tokens_per_second": 639.824
    },
    {
      "epoch": 0.37604245640636846,
      "grad_norm": 1.2799185514450073,
      "learning_rate": 1.629367088607595e-05,
      "loss": 0.06554621458053589,
      "num_input_tokens_seen": 697804,
      "step": 372,
      "train_runtime": 1090.6703,
      "train_tokens_per_second": 639.794
    },
    {
      "epoch": 0.37705332322466517,
      "grad_norm": 1.4671586751937866,
      "learning_rate": 1.6283544303797468e-05,
      "loss": 0.10723006725311279,
      "num_input_tokens_seen": 699614,
      "step": 373,
      "train_runtime": 1092.9454,
      "train_tokens_per_second": 640.118
    },
    {
      "epoch": 0.3780641900429618,
      "grad_norm": 1.632125973701477,
      "learning_rate": 1.627341772151899e-05,
      "loss": 0.11110654473304749,
      "num_input_tokens_seen": 700990,
      "step": 374,
      "train_runtime": 1095.0854,
      "train_tokens_per_second": 640.124
    },
    {
      "epoch": 0.37907505686125853,
      "grad_norm": 1.1445481777191162,
      "learning_rate": 1.6263291139240506e-05,
      "loss": 0.10859709978103638,
      "num_input_tokens_seen": 703404,
      "step": 375,
      "train_runtime": 1097.4698,
      "train_tokens_per_second": 640.932
    },
    {
      "epoch": 0.38008592367955524,
      "grad_norm": 1.453262209892273,
      "learning_rate": 1.6253164556962026e-05,
      "loss": 0.10258646309375763,
      "num_input_tokens_seen": 705094,
      "step": 376,
      "train_runtime": 1099.6843,
      "train_tokens_per_second": 641.179
    },
    {
      "epoch": 0.3810967904978519,
      "grad_norm": 1.593936800956726,
      "learning_rate": 1.6243037974683548e-05,
      "loss": 0.11771631240844727,
      "num_input_tokens_seen": 706698,
      "step": 377,
      "train_runtime": 1101.865,
      "train_tokens_per_second": 641.365
    },
    {
      "epoch": 0.3821076573161486,
      "grad_norm": 1.1572998762130737,
      "learning_rate": 1.6232911392405064e-05,
      "loss": 0.06967834383249283,
      "num_input_tokens_seen": 708394,
      "step": 378,
      "train_runtime": 1104.1007,
      "train_tokens_per_second": 641.603
    },
    {
      "epoch": 0.3831185241344453,
      "grad_norm": 1.2723830938339233,
      "learning_rate": 1.6222784810126583e-05,
      "loss": 0.08734267204999924,
      "num_input_tokens_seen": 710096,
      "step": 379,
      "train_runtime": 1106.2998,
      "train_tokens_per_second": 641.866
    },
    {
      "epoch": 0.38412939095274196,
      "grad_norm": 1.3895487785339355,
      "learning_rate": 1.6212658227848102e-05,
      "loss": 0.06834635883569717,
      "num_input_tokens_seen": 711458,
      "step": 380,
      "train_runtime": 1108.4048,
      "train_tokens_per_second": 641.876
    },
    {
      "epoch": 0.38514025777103866,
      "grad_norm": 1.4300158023834229,
      "learning_rate": 1.620253164556962e-05,
      "loss": 0.12024091929197311,
      "num_input_tokens_seen": 713746,
      "step": 381,
      "train_runtime": 1110.7597,
      "train_tokens_per_second": 642.575
    },
    {
      "epoch": 0.38615112458933537,
      "grad_norm": 1.3179868459701538,
      "learning_rate": 1.619240506329114e-05,
      "loss": 0.07144320756196976,
      "num_input_tokens_seen": 715434,
      "step": 382,
      "train_runtime": 1112.9623,
      "train_tokens_per_second": 642.82
    },
    {
      "epoch": 0.387161991407632,
      "grad_norm": 1.2376477718353271,
      "learning_rate": 1.618227848101266e-05,
      "loss": 0.08733730763196945,
      "num_input_tokens_seen": 717224,
      "step": 383,
      "train_runtime": 1115.1583,
      "train_tokens_per_second": 643.159
    },
    {
      "epoch": 0.38817285822592873,
      "grad_norm": 1.5439473390579224,
      "learning_rate": 1.6172151898734178e-05,
      "loss": 0.12377951294183731,
      "num_input_tokens_seen": 718734,
      "step": 384,
      "train_runtime": 1117.3266,
      "train_tokens_per_second": 643.262
    },
    {
      "epoch": 0.38918372504422544,
      "grad_norm": 1.0904730558395386,
      "learning_rate": 1.6162025316455697e-05,
      "loss": 0.07275289297103882,
      "num_input_tokens_seen": 720200,
      "step": 385,
      "train_runtime": 1119.516,
      "train_tokens_per_second": 643.314
    },
    {
      "epoch": 0.3901945918625221,
      "grad_norm": 1.4814702272415161,
      "learning_rate": 1.6151898734177216e-05,
      "loss": 0.11074559390544891,
      "num_input_tokens_seen": 722210,
      "step": 386,
      "train_runtime": 1121.8135,
      "train_tokens_per_second": 643.788
    },
    {
      "epoch": 0.3912054586808188,
      "grad_norm": 1.364432454109192,
      "learning_rate": 1.6141772151898735e-05,
      "loss": 0.09846647828817368,
      "num_input_tokens_seen": 724340,
      "step": 387,
      "train_runtime": 1124.1489,
      "train_tokens_per_second": 644.345
    },
    {
      "epoch": 0.3922163254991155,
      "grad_norm": 1.567051649093628,
      "learning_rate": 1.6131645569620255e-05,
      "loss": 0.10375119745731354,
      "num_input_tokens_seen": 725928,
      "step": 388,
      "train_runtime": 1126.3304,
      "train_tokens_per_second": 644.507
    },
    {
      "epoch": 0.39322719231741216,
      "grad_norm": 1.3150216341018677,
      "learning_rate": 1.6121518987341774e-05,
      "loss": 0.09941066801548004,
      "num_input_tokens_seen": 728306,
      "step": 389,
      "train_runtime": 1128.726,
      "train_tokens_per_second": 645.246
    },
    {
      "epoch": 0.39423805913570886,
      "grad_norm": 1.4323986768722534,
      "learning_rate": 1.6111392405063293e-05,
      "loss": 0.12157496809959412,
      "num_input_tokens_seen": 730248,
      "step": 390,
      "train_runtime": 1130.9986,
      "train_tokens_per_second": 645.667
    },
    {
      "epoch": 0.39524892595400557,
      "grad_norm": 1.3855671882629395,
      "learning_rate": 1.6101265822784812e-05,
      "loss": 0.08571504056453705,
      "num_input_tokens_seen": 732022,
      "step": 391,
      "train_runtime": 1148.4282,
      "train_tokens_per_second": 637.412
    },
    {
      "epoch": 0.3962597927723022,
      "grad_norm": 1.5683603286743164,
      "learning_rate": 1.609113924050633e-05,
      "loss": 0.13069680333137512,
      "num_input_tokens_seen": 734084,
      "step": 392,
      "train_runtime": 1150.74,
      "train_tokens_per_second": 637.923
    },
    {
      "epoch": 0.39727065959059893,
      "grad_norm": 1.5007070302963257,
      "learning_rate": 1.608101265822785e-05,
      "loss": 0.11729087680578232,
      "num_input_tokens_seen": 735922,
      "step": 393,
      "train_runtime": 1152.9967,
      "train_tokens_per_second": 638.269
    },
    {
      "epoch": 0.39828152640889564,
      "grad_norm": 1.3590445518493652,
      "learning_rate": 1.607088607594937e-05,
      "loss": 0.10647516697645187,
      "num_input_tokens_seen": 737962,
      "step": 394,
      "train_runtime": 1155.2977,
      "train_tokens_per_second": 638.763
    },
    {
      "epoch": 0.39929239322719234,
      "grad_norm": 1.802383542060852,
      "learning_rate": 1.6060759493670888e-05,
      "loss": 0.12328140437602997,
      "num_input_tokens_seen": 739550,
      "step": 395,
      "train_runtime": 1157.5286,
      "train_tokens_per_second": 638.904
    },
    {
      "epoch": 0.400303260045489,
      "grad_norm": 1.37345290184021,
      "learning_rate": 1.6050632911392404e-05,
      "loss": 0.0954669788479805,
      "num_input_tokens_seen": 741172,
      "step": 396,
      "train_runtime": 1159.7827,
      "train_tokens_per_second": 639.061
    },
    {
      "epoch": 0.400303260045489,
      "eval_loss": 0.3872518241405487,
      "eval_runtime": 55.0988,
      "eval_samples_per_second": 15.953,
      "eval_steps_per_second": 7.986,
      "num_input_tokens_seen": 741172,
      "step": 396
    },
    {
      "epoch": 0.4013141268637857,
      "grad_norm": 1.3639532327651978,
      "learning_rate": 1.6040506329113926e-05,
      "loss": 0.08709590882062912,
      "num_input_tokens_seen": 743004,
      "step": 397,
      "train_runtime": 1217.1537,
      "train_tokens_per_second": 610.444
    },
    {
      "epoch": 0.4023249936820824,
      "grad_norm": 1.1442697048187256,
      "learning_rate": 1.6030379746835445e-05,
      "loss": 0.09475848823785782,
      "num_input_tokens_seen": 744848,
      "step": 398,
      "train_runtime": 1219.4153,
      "train_tokens_per_second": 610.824
    },
    {
      "epoch": 0.40333586050037906,
      "grad_norm": 1.2756831645965576,
      "learning_rate": 1.602025316455696e-05,
      "loss": 0.0921563133597374,
      "num_input_tokens_seen": 746680,
      "step": 399,
      "train_runtime": 1221.7141,
      "train_tokens_per_second": 611.174
    },
    {
      "epoch": 0.40434672731867577,
      "grad_norm": 1.3857290744781494,
      "learning_rate": 1.6010126582278484e-05,
      "loss": 0.10650795698165894,
      "num_input_tokens_seen": 748696,
      "step": 400,
      "train_runtime": 1224.0205,
      "train_tokens_per_second": 611.669
    },
    {
      "epoch": 0.4053575941369725,
      "grad_norm": 1.2589681148529053,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.08366907387971878,
      "num_input_tokens_seen": 750616,
      "step": 401,
      "train_runtime": 1226.3543,
      "train_tokens_per_second": 612.071
    },
    {
      "epoch": 0.40636846095526913,
      "grad_norm": 1.5011835098266602,
      "learning_rate": 1.598987341772152e-05,
      "loss": 0.11901634931564331,
      "num_input_tokens_seen": 752420,
      "step": 402,
      "train_runtime": 1228.6163,
      "train_tokens_per_second": 612.413
    },
    {
      "epoch": 0.40737932777356584,
      "grad_norm": 1.5548064708709717,
      "learning_rate": 1.597974683544304e-05,
      "loss": 0.1128741055727005,
      "num_input_tokens_seen": 754336,
      "step": 403,
      "train_runtime": 1230.8945,
      "train_tokens_per_second": 612.836
    },
    {
      "epoch": 0.40839019459186254,
      "grad_norm": 1.4540143013000488,
      "learning_rate": 1.5969620253164557e-05,
      "loss": 0.09682419896125793,
      "num_input_tokens_seen": 756468,
      "step": 404,
      "train_runtime": 1233.2427,
      "train_tokens_per_second": 613.397
    },
    {
      "epoch": 0.4094010614101592,
      "grad_norm": 1.2885997295379639,
      "learning_rate": 1.5959493670886076e-05,
      "loss": 0.10913906246423721,
      "num_input_tokens_seen": 758516,
      "step": 405,
      "train_runtime": 1235.6416,
      "train_tokens_per_second": 613.864
    },
    {
      "epoch": 0.4104119282284559,
      "grad_norm": 1.3824455738067627,
      "learning_rate": 1.5949367088607598e-05,
      "loss": 0.10396812856197357,
      "num_input_tokens_seen": 760364,
      "step": 406,
      "train_runtime": 1237.8792,
      "train_tokens_per_second": 614.247
    },
    {
      "epoch": 0.4114227950467526,
      "grad_norm": 2.2940287590026855,
      "learning_rate": 1.5939240506329114e-05,
      "loss": 0.10810710489749908,
      "num_input_tokens_seen": 762296,
      "step": 407,
      "train_runtime": 1240.2508,
      "train_tokens_per_second": 614.631
    },
    {
      "epoch": 0.41243366186504926,
      "grad_norm": 1.3997433185577393,
      "learning_rate": 1.5929113924050633e-05,
      "loss": 0.08424016833305359,
      "num_input_tokens_seen": 764002,
      "step": 408,
      "train_runtime": 1242.5433,
      "train_tokens_per_second": 614.87
    },
    {
      "epoch": 0.41344452868334597,
      "grad_norm": 1.3747771978378296,
      "learning_rate": 1.5918987341772155e-05,
      "loss": 0.13588640093803406,
      "num_input_tokens_seen": 766318,
      "step": 409,
      "train_runtime": 1244.9952,
      "train_tokens_per_second": 615.519
    },
    {
      "epoch": 0.4144553955016427,
      "grad_norm": 1.525220274925232,
      "learning_rate": 1.590886075949367e-05,
      "loss": 0.09460221230983734,
      "num_input_tokens_seen": 768542,
      "step": 410,
      "train_runtime": 1247.3324,
      "train_tokens_per_second": 616.149
    },
    {
      "epoch": 0.41546626231993933,
      "grad_norm": 1.5507614612579346,
      "learning_rate": 1.589873417721519e-05,
      "loss": 0.08481799066066742,
      "num_input_tokens_seen": 770400,
      "step": 411,
      "train_runtime": 1249.59,
      "train_tokens_per_second": 616.522
    },
    {
      "epoch": 0.41647712913823604,
      "grad_norm": 1.088625192642212,
      "learning_rate": 1.588860759493671e-05,
      "loss": 0.08084049820899963,
      "num_input_tokens_seen": 773532,
      "step": 412,
      "train_runtime": 1252.1702,
      "train_tokens_per_second": 617.753
    },
    {
      "epoch": 0.41748799595653274,
      "grad_norm": 1.157347559928894,
      "learning_rate": 1.5878481012658228e-05,
      "loss": 0.07886244356632233,
      "num_input_tokens_seen": 775576,
      "step": 413,
      "train_runtime": 1254.4628,
      "train_tokens_per_second": 618.253
    },
    {
      "epoch": 0.4184988627748294,
      "grad_norm": 1.718621850013733,
      "learning_rate": 1.5868354430379747e-05,
      "loss": 0.11512285470962524,
      "num_input_tokens_seen": 777670,
      "step": 414,
      "train_runtime": 1256.7786,
      "train_tokens_per_second": 618.78
    },
    {
      "epoch": 0.4195097295931261,
      "grad_norm": 1.499655842781067,
      "learning_rate": 1.5858227848101266e-05,
      "loss": 0.10185996443033218,
      "num_input_tokens_seen": 779290,
      "step": 415,
      "train_runtime": 1259.0203,
      "train_tokens_per_second": 618.965
    },
    {
      "epoch": 0.4205205964114228,
      "grad_norm": 1.5286248922348022,
      "learning_rate": 1.5848101265822786e-05,
      "loss": 0.11060277372598648,
      "num_input_tokens_seen": 781116,
      "step": 416,
      "train_runtime": 1261.2613,
      "train_tokens_per_second": 619.313
    },
    {
      "epoch": 0.42153146322971946,
      "grad_norm": 1.2865464687347412,
      "learning_rate": 1.5837974683544305e-05,
      "loss": 0.10144440829753876,
      "num_input_tokens_seen": 783262,
      "step": 417,
      "train_runtime": 1263.5843,
      "train_tokens_per_second": 619.873
    },
    {
      "epoch": 0.42254233004801617,
      "grad_norm": 1.285799503326416,
      "learning_rate": 1.5827848101265824e-05,
      "loss": 0.09051541239023209,
      "num_input_tokens_seen": 785110,
      "step": 418,
      "train_runtime": 1265.9202,
      "train_tokens_per_second": 620.189
    },
    {
      "epoch": 0.4235531968663129,
      "grad_norm": 1.2414003610610962,
      "learning_rate": 1.5817721518987343e-05,
      "loss": 0.10315880924463272,
      "num_input_tokens_seen": 787038,
      "step": 419,
      "train_runtime": 1268.2454,
      "train_tokens_per_second": 620.572
    },
    {
      "epoch": 0.4245640636846095,
      "grad_norm": 1.362291932106018,
      "learning_rate": 1.5807594936708862e-05,
      "loss": 0.10962265729904175,
      "num_input_tokens_seen": 788866,
      "step": 420,
      "train_runtime": 1270.5292,
      "train_tokens_per_second": 620.896
    },
    {
      "epoch": 0.42557493050290623,
      "grad_norm": 1.3942428827285767,
      "learning_rate": 1.579746835443038e-05,
      "loss": 0.08523203432559967,
      "num_input_tokens_seen": 790578,
      "step": 421,
      "train_runtime": 1287.8445,
      "train_tokens_per_second": 613.877
    },
    {
      "epoch": 0.42658579732120294,
      "grad_norm": 1.325069785118103,
      "learning_rate": 1.57873417721519e-05,
      "loss": 0.08947727084159851,
      "num_input_tokens_seen": 792632,
      "step": 422,
      "train_runtime": 1290.1336,
      "train_tokens_per_second": 614.38
    },
    {
      "epoch": 0.4275966641394996,
      "grad_norm": 1.4082978963851929,
      "learning_rate": 1.577721518987342e-05,
      "loss": 0.07078568637371063,
      "num_input_tokens_seen": 794344,
      "step": 423,
      "train_runtime": 1292.3695,
      "train_tokens_per_second": 614.642
    },
    {
      "epoch": 0.4286075309577963,
      "grad_norm": 1.1040624380111694,
      "learning_rate": 1.5767088607594938e-05,
      "loss": 0.0761384516954422,
      "num_input_tokens_seen": 796212,
      "step": 424,
      "train_runtime": 1294.6402,
      "train_tokens_per_second": 615.006
    },
    {
      "epoch": 0.429618397776093,
      "grad_norm": 1.1366462707519531,
      "learning_rate": 1.5756962025316457e-05,
      "loss": 0.09637749940156937,
      "num_input_tokens_seen": 798458,
      "step": 425,
      "train_runtime": 1297.0632,
      "train_tokens_per_second": 615.589
    },
    {
      "epoch": 0.4306292645943897,
      "grad_norm": 1.388405203819275,
      "learning_rate": 1.5746835443037976e-05,
      "loss": 0.10852111876010895,
      "num_input_tokens_seen": 800190,
      "step": 426,
      "train_runtime": 1299.3023,
      "train_tokens_per_second": 615.861
    },
    {
      "epoch": 0.43164013141268637,
      "grad_norm": 1.279977798461914,
      "learning_rate": 1.5736708860759495e-05,
      "loss": 0.08166302740573883,
      "num_input_tokens_seen": 801656,
      "step": 427,
      "train_runtime": 1301.4619,
      "train_tokens_per_second": 615.966
    },
    {
      "epoch": 0.4326509982309831,
      "grad_norm": 1.1389224529266357,
      "learning_rate": 1.5726582278481015e-05,
      "loss": 0.06751921772956848,
      "num_input_tokens_seen": 803426,
      "step": 428,
      "train_runtime": 1303.8705,
      "train_tokens_per_second": 616.185
    },
    {
      "epoch": 0.4336618650492798,
      "grad_norm": 1.2209196090698242,
      "learning_rate": 1.5716455696202534e-05,
      "loss": 0.07853812724351883,
      "num_input_tokens_seen": 805396,
      "step": 429,
      "train_runtime": 1306.1459,
      "train_tokens_per_second": 616.62
    },
    {
      "epoch": 0.43467273186757643,
      "grad_norm": 1.1403597593307495,
      "learning_rate": 1.5706329113924053e-05,
      "loss": 0.10636825859546661,
      "num_input_tokens_seen": 807228,
      "step": 430,
      "train_runtime": 1308.3976,
      "train_tokens_per_second": 616.959
    },
    {
      "epoch": 0.43568359868587314,
      "grad_norm": 1.4978351593017578,
      "learning_rate": 1.5696202531645572e-05,
      "loss": 0.11419200897216797,
      "num_input_tokens_seen": 808880,
      "step": 431,
      "train_runtime": 1310.5998,
      "train_tokens_per_second": 617.183
    },
    {
      "epoch": 0.43669446550416985,
      "grad_norm": 1.1914385557174683,
      "learning_rate": 1.568607594936709e-05,
      "loss": 0.09804163128137589,
      "num_input_tokens_seen": 811672,
      "step": 432,
      "train_runtime": 1313.1271,
      "train_tokens_per_second": 618.121
    },
    {
      "epoch": 0.4377053323224665,
      "grad_norm": 1.1298398971557617,
      "learning_rate": 1.567594936708861e-05,
      "loss": 0.07625232636928558,
      "num_input_tokens_seen": 813512,
      "step": 433,
      "train_runtime": 1315.4346,
      "train_tokens_per_second": 618.436
    },
    {
      "epoch": 0.4387161991407632,
      "grad_norm": 1.5358973741531372,
      "learning_rate": 1.566582278481013e-05,
      "loss": 0.10365542024374008,
      "num_input_tokens_seen": 815038,
      "step": 434,
      "train_runtime": 1317.6541,
      "train_tokens_per_second": 618.552
    },
    {
      "epoch": 0.4397270659590599,
      "grad_norm": 1.553647756576538,
      "learning_rate": 1.5655696202531648e-05,
      "loss": 0.1037188470363617,
      "num_input_tokens_seen": 816976,
      "step": 435,
      "train_runtime": 1319.9001,
      "train_tokens_per_second": 618.968
    },
    {
      "epoch": 0.44073793277735657,
      "grad_norm": 1.0904991626739502,
      "learning_rate": 1.5645569620253164e-05,
      "loss": 0.0808977261185646,
      "num_input_tokens_seen": 818870,
      "step": 436,
      "train_runtime": 1322.1532,
      "train_tokens_per_second": 619.346
    },
    {
      "epoch": 0.4417487995956533,
      "grad_norm": 1.5115450620651245,
      "learning_rate": 1.5635443037974686e-05,
      "loss": 0.11773766577243805,
      "num_input_tokens_seen": 821004,
      "step": 437,
      "train_runtime": 1324.509,
      "train_tokens_per_second": 619.855
    },
    {
      "epoch": 0.44275966641395,
      "grad_norm": 1.0825086832046509,
      "learning_rate": 1.5625316455696205e-05,
      "loss": 0.07502110302448273,
      "num_input_tokens_seen": 823306,
      "step": 438,
      "train_runtime": 1326.9529,
      "train_tokens_per_second": 620.449
    },
    {
      "epoch": 0.44377053323224663,
      "grad_norm": 1.4008469581604004,
      "learning_rate": 1.561518987341772e-05,
      "loss": 0.12766876816749573,
      "num_input_tokens_seen": 825712,
      "step": 439,
      "train_runtime": 1329.4654,
      "train_tokens_per_second": 621.086
    },
    {
      "epoch": 0.44478140005054334,
      "grad_norm": 1.4241653680801392,
      "learning_rate": 1.5605063291139244e-05,
      "loss": 0.09557639807462692,
      "num_input_tokens_seen": 827680,
      "step": 440,
      "train_runtime": 1331.8762,
      "train_tokens_per_second": 621.439
    },
    {
      "epoch": 0.44579226686884005,
      "grad_norm": 1.3624467849731445,
      "learning_rate": 1.5594936708860763e-05,
      "loss": 0.10515819489955902,
      "num_input_tokens_seen": 829328,
      "step": 441,
      "train_runtime": 1334.1661,
      "train_tokens_per_second": 621.608
    },
    {
      "epoch": 0.4468031336871367,
      "grad_norm": 1.4202001094818115,
      "learning_rate": 1.558481012658228e-05,
      "loss": 0.09325391799211502,
      "num_input_tokens_seen": 831254,
      "step": 442,
      "train_runtime": 1336.4782,
      "train_tokens_per_second": 621.973
    },
    {
      "epoch": 0.4478140005054334,
      "grad_norm": 1.4952328205108643,
      "learning_rate": 1.55746835443038e-05,
      "loss": 0.12340879440307617,
      "num_input_tokens_seen": 833076,
      "step": 443,
      "train_runtime": 1338.8319,
      "train_tokens_per_second": 622.241
    },
    {
      "epoch": 0.4488248673237301,
      "grad_norm": 1.1978174448013306,
      "learning_rate": 1.5564556962025317e-05,
      "loss": 0.10327230393886566,
      "num_input_tokens_seen": 835272,
      "step": 444,
      "train_runtime": 1341.2874,
      "train_tokens_per_second": 622.739
    },
    {
      "epoch": 0.44983573414202677,
      "grad_norm": 1.4203438758850098,
      "learning_rate": 1.5554430379746836e-05,
      "loss": 0.1020071730017662,
      "num_input_tokens_seen": 837422,
      "step": 445,
      "train_runtime": 1343.6062,
      "train_tokens_per_second": 623.264
    },
    {
      "epoch": 0.4508466009603235,
      "grad_norm": 1.1363866329193115,
      "learning_rate": 1.5544303797468358e-05,
      "loss": 0.08220135420560837,
      "num_input_tokens_seen": 839660,
      "step": 446,
      "train_runtime": 1345.9437,
      "train_tokens_per_second": 623.845
    },
    {
      "epoch": 0.4518574677786202,
      "grad_norm": 1.6075685024261475,
      "learning_rate": 1.5534177215189874e-05,
      "loss": 0.12566004693508148,
      "num_input_tokens_seen": 841170,
      "step": 447,
      "train_runtime": 1348.1186,
      "train_tokens_per_second": 623.958
    },
    {
      "epoch": 0.45286833459691683,
      "grad_norm": 1.448835015296936,
      "learning_rate": 1.5524050632911393e-05,
      "loss": 0.12754669785499573,
      "num_input_tokens_seen": 843086,
      "step": 448,
      "train_runtime": 1350.3928,
      "train_tokens_per_second": 624.326
    },
    {
      "epoch": 0.45387920141521354,
      "grad_norm": 1.1091268062591553,
      "learning_rate": 1.5513924050632912e-05,
      "loss": 0.06876929104328156,
      "num_input_tokens_seen": 845328,
      "step": 449,
      "train_runtime": 1352.7205,
      "train_tokens_per_second": 624.91
    },
    {
      "epoch": 0.45489006823351025,
      "grad_norm": 1.4727420806884766,
      "learning_rate": 1.550379746835443e-05,
      "loss": 0.1233140230178833,
      "num_input_tokens_seen": 847832,
      "step": 450,
      "train_runtime": 1355.1187,
      "train_tokens_per_second": 625.651
    },
    {
      "epoch": 0.4559009350518069,
      "grad_norm": 0.9638090133666992,
      "learning_rate": 1.549367088607595e-05,
      "loss": 0.062319010496139526,
      "num_input_tokens_seen": 849910,
      "step": 451,
      "train_runtime": 1371.6498,
      "train_tokens_per_second": 619.626
    },
    {
      "epoch": 0.4569118018701036,
      "grad_norm": 1.1417428255081177,
      "learning_rate": 1.548354430379747e-05,
      "loss": 0.10173853486776352,
      "num_input_tokens_seen": 852294,
      "step": 452,
      "train_runtime": 1374.1042,
      "train_tokens_per_second": 620.254
    },
    {
      "epoch": 0.4579226686884003,
      "grad_norm": 1.3086588382720947,
      "learning_rate": 1.5473417721518988e-05,
      "loss": 0.09739860892295837,
      "num_input_tokens_seen": 854262,
      "step": 453,
      "train_runtime": 1376.4508,
      "train_tokens_per_second": 620.627
    },
    {
      "epoch": 0.45893353550669697,
      "grad_norm": 1.3375917673110962,
      "learning_rate": 1.5463291139240507e-05,
      "loss": 0.1066465973854065,
      "num_input_tokens_seen": 856256,
      "step": 454,
      "train_runtime": 1378.7841,
      "train_tokens_per_second": 621.023
    },
    {
      "epoch": 0.4599444023249937,
      "grad_norm": 1.4359407424926758,
      "learning_rate": 1.5453164556962026e-05,
      "loss": 0.08826154470443726,
      "num_input_tokens_seen": 857960,
      "step": 455,
      "train_runtime": 1381.0232,
      "train_tokens_per_second": 621.25
    },
    {
      "epoch": 0.4609552691432904,
      "grad_norm": 1.2359154224395752,
      "learning_rate": 1.5443037974683546e-05,
      "loss": 0.07979286462068558,
      "num_input_tokens_seen": 859844,
      "step": 456,
      "train_runtime": 1383.2948,
      "train_tokens_per_second": 621.591
    },
    {
      "epoch": 0.4619661359615871,
      "grad_norm": 2.229543685913086,
      "learning_rate": 1.5432911392405065e-05,
      "loss": 0.12227626144886017,
      "num_input_tokens_seen": 862234,
      "step": 457,
      "train_runtime": 1385.7057,
      "train_tokens_per_second": 622.235
    },
    {
      "epoch": 0.46297700277988374,
      "grad_norm": 1.5136470794677734,
      "learning_rate": 1.5422784810126584e-05,
      "loss": 0.1630091667175293,
      "num_input_tokens_seen": 864526,
      "step": 458,
      "train_runtime": 1388.0672,
      "train_tokens_per_second": 622.827
    },
    {
      "epoch": 0.46398786959818045,
      "grad_norm": 1.4026036262512207,
      "learning_rate": 1.5412658227848103e-05,
      "loss": 0.08755923062562943,
      "num_input_tokens_seen": 866254,
      "step": 459,
      "train_runtime": 1390.2948,
      "train_tokens_per_second": 623.072
    },
    {
      "epoch": 0.46499873641647715,
      "grad_norm": 1.5655118227005005,
      "learning_rate": 1.5402531645569622e-05,
      "loss": 0.11648698896169662,
      "num_input_tokens_seen": 868044,
      "step": 460,
      "train_runtime": 1392.5821,
      "train_tokens_per_second": 623.334
    },
    {
      "epoch": 0.4660096032347738,
      "grad_norm": 1.6040105819702148,
      "learning_rate": 1.539240506329114e-05,
      "loss": 0.12708653509616852,
      "num_input_tokens_seen": 869820,
      "step": 461,
      "train_runtime": 1394.8295,
      "train_tokens_per_second": 623.603
    },
    {
      "epoch": 0.4670204700530705,
      "grad_norm": 1.1650915145874023,
      "learning_rate": 1.538227848101266e-05,
      "loss": 0.0768815353512764,
      "num_input_tokens_seen": 871334,
      "step": 462,
      "train_runtime": 1397.0258,
      "train_tokens_per_second": 623.706
    },
    {
      "epoch": 0.4680313368713672,
      "grad_norm": 1.2267590761184692,
      "learning_rate": 1.537215189873418e-05,
      "loss": 0.11306999623775482,
      "num_input_tokens_seen": 873256,
      "step": 463,
      "train_runtime": 1399.3227,
      "train_tokens_per_second": 624.056
    },
    {
      "epoch": 0.4690422036896639,
      "grad_norm": 1.8214051723480225,
      "learning_rate": 1.5362025316455698e-05,
      "loss": 0.1086057648062706,
      "num_input_tokens_seen": 875286,
      "step": 464,
      "train_runtime": 1401.5966,
      "train_tokens_per_second": 624.492
    },
    {
      "epoch": 0.4700530705079606,
      "grad_norm": 1.5773112773895264,
      "learning_rate": 1.5351898734177217e-05,
      "loss": 0.09585914015769958,
      "num_input_tokens_seen": 876870,
      "step": 465,
      "train_runtime": 1403.8159,
      "train_tokens_per_second": 624.633
    },
    {
      "epoch": 0.4710639373262573,
      "grad_norm": 1.3764119148254395,
      "learning_rate": 1.5341772151898736e-05,
      "loss": 0.08842549473047256,
      "num_input_tokens_seen": 878840,
      "step": 466,
      "train_runtime": 1406.0904,
      "train_tokens_per_second": 625.024
    },
    {
      "epoch": 0.47207480414455394,
      "grad_norm": 1.25182044506073,
      "learning_rate": 1.5331645569620255e-05,
      "loss": 0.0800933912396431,
      "num_input_tokens_seen": 880468,
      "step": 467,
      "train_runtime": 1408.2757,
      "train_tokens_per_second": 625.21
    },
    {
      "epoch": 0.47308567096285065,
      "grad_norm": 1.4983059167861938,
      "learning_rate": 1.532151898734177e-05,
      "loss": 0.08812262117862701,
      "num_input_tokens_seen": 882314,
      "step": 468,
      "train_runtime": 1410.5973,
      "train_tokens_per_second": 625.49
    },
    {
      "epoch": 0.47409653778114735,
      "grad_norm": 1.5544148683547974,
      "learning_rate": 1.5311392405063294e-05,
      "loss": 0.09277161955833435,
      "num_input_tokens_seen": 884162,
      "step": 469,
      "train_runtime": 1412.8327,
      "train_tokens_per_second": 625.808
    },
    {
      "epoch": 0.475107404599444,
      "grad_norm": 1.2759485244750977,
      "learning_rate": 1.5301265822784813e-05,
      "loss": 0.08562381565570831,
      "num_input_tokens_seen": 885872,
      "step": 470,
      "train_runtime": 1415.1272,
      "train_tokens_per_second": 626.002
    },
    {
      "epoch": 0.4761182714177407,
      "grad_norm": 1.410587191581726,
      "learning_rate": 1.529113924050633e-05,
      "loss": 0.10296129435300827,
      "num_input_tokens_seen": 887478,
      "step": 471,
      "train_runtime": 1417.34,
      "train_tokens_per_second": 626.157
    },
    {
      "epoch": 0.4771291382360374,
      "grad_norm": 1.3727208375930786,
      "learning_rate": 1.528101265822785e-05,
      "loss": 0.10469543188810349,
      "num_input_tokens_seen": 889464,
      "step": 472,
      "train_runtime": 1419.682,
      "train_tokens_per_second": 626.523
    },
    {
      "epoch": 0.47814000505433407,
      "grad_norm": 1.1912351846694946,
      "learning_rate": 1.527088607594937e-05,
      "loss": 0.08748653531074524,
      "num_input_tokens_seen": 890990,
      "step": 473,
      "train_runtime": 1421.8167,
      "train_tokens_per_second": 626.656
    },
    {
      "epoch": 0.4791508718726308,
      "grad_norm": 1.2304508686065674,
      "learning_rate": 1.5260759493670886e-05,
      "loss": 0.08261626958847046,
      "num_input_tokens_seen": 892672,
      "step": 474,
      "train_runtime": 1424.0693,
      "train_tokens_per_second": 626.846
    },
    {
      "epoch": 0.4801617386909275,
      "grad_norm": 1.2530360221862793,
      "learning_rate": 1.5250632911392406e-05,
      "loss": 0.0887748971581459,
      "num_input_tokens_seen": 894660,
      "step": 475,
      "train_runtime": 1426.5372,
      "train_tokens_per_second": 627.155
    },
    {
      "epoch": 0.48117260550922414,
      "grad_norm": 1.5004278421401978,
      "learning_rate": 1.5240506329113924e-05,
      "loss": 0.10353885591030121,
      "num_input_tokens_seen": 896764,
      "step": 476,
      "train_runtime": 1428.9466,
      "train_tokens_per_second": 627.57
    },
    {
      "epoch": 0.48218347232752085,
      "grad_norm": 1.3240872621536255,
      "learning_rate": 1.5230379746835445e-05,
      "loss": 0.09488523751497269,
      "num_input_tokens_seen": 898270,
      "step": 477,
      "train_runtime": 1431.1262,
      "train_tokens_per_second": 627.667
    },
    {
      "epoch": 0.48319433914581755,
      "grad_norm": 1.6358325481414795,
      "learning_rate": 1.5220253164556964e-05,
      "loss": 0.10431206226348877,
      "num_input_tokens_seen": 900776,
      "step": 478,
      "train_runtime": 1433.5816,
      "train_tokens_per_second": 628.34
    },
    {
      "epoch": 0.4842052059641142,
      "grad_norm": 1.6957182884216309,
      "learning_rate": 1.5210126582278481e-05,
      "loss": 0.0877433717250824,
      "num_input_tokens_seen": 902464,
      "step": 479,
      "train_runtime": 1435.7843,
      "train_tokens_per_second": 628.551
    },
    {
      "epoch": 0.4852160727824109,
      "grad_norm": 1.5795923471450806,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 0.11310522258281708,
      "num_input_tokens_seen": 904536,
      "step": 480,
      "train_runtime": 1438.2063,
      "train_tokens_per_second": 628.933
    },
    {
      "epoch": 0.4862269396007076,
      "grad_norm": 1.4067100286483765,
      "learning_rate": 1.5189873417721521e-05,
      "loss": 0.1028033122420311,
      "num_input_tokens_seen": 906342,
      "step": 481,
      "train_runtime": 1454.2403,
      "train_tokens_per_second": 623.241
    },
    {
      "epoch": 0.48723780641900427,
      "grad_norm": 1.2052048444747925,
      "learning_rate": 1.5179746835443038e-05,
      "loss": 0.08187657594680786,
      "num_input_tokens_seen": 908624,
      "step": 482,
      "train_runtime": 1456.6482,
      "train_tokens_per_second": 623.777
    },
    {
      "epoch": 0.488248673237301,
      "grad_norm": 1.2163141965866089,
      "learning_rate": 1.5169620253164559e-05,
      "loss": 0.09129151701927185,
      "num_input_tokens_seen": 910822,
      "step": 483,
      "train_runtime": 1458.9901,
      "train_tokens_per_second": 624.282
    },
    {
      "epoch": 0.4892595400555977,
      "grad_norm": 1.49179208278656,
      "learning_rate": 1.5159493670886076e-05,
      "loss": 0.11724289506673813,
      "num_input_tokens_seen": 912538,
      "step": 484,
      "train_runtime": 1461.3004,
      "train_tokens_per_second": 624.47
    },
    {
      "epoch": 0.49027040687389434,
      "grad_norm": 1.0660570859909058,
      "learning_rate": 1.5149367088607596e-05,
      "loss": 0.07163531333208084,
      "num_input_tokens_seen": 914398,
      "step": 485,
      "train_runtime": 1463.609,
      "train_tokens_per_second": 624.756
    },
    {
      "epoch": 0.49128127369219105,
      "grad_norm": 1.3062098026275635,
      "learning_rate": 1.5139240506329116e-05,
      "loss": 0.08330444991588593,
      "num_input_tokens_seen": 915784,
      "step": 486,
      "train_runtime": 1465.7715,
      "train_tokens_per_second": 624.78
    },
    {
      "epoch": 0.49229214051048775,
      "grad_norm": 1.5049551725387573,
      "learning_rate": 1.5129113924050634e-05,
      "loss": 0.09343305230140686,
      "num_input_tokens_seen": 917832,
      "step": 487,
      "train_runtime": 1468.2228,
      "train_tokens_per_second": 625.131
    },
    {
      "epoch": 0.49330300732878446,
      "grad_norm": 1.3915646076202393,
      "learning_rate": 1.5118987341772153e-05,
      "loss": 0.14162439107894897,
      "num_input_tokens_seen": 919822,
      "step": 488,
      "train_runtime": 1470.5186,
      "train_tokens_per_second": 625.509
    },
    {
      "epoch": 0.4943138741470811,
      "grad_norm": 1.029426097869873,
      "learning_rate": 1.5108860759493674e-05,
      "loss": 0.06789886206388474,
      "num_input_tokens_seen": 921798,
      "step": 489,
      "train_runtime": 1472.8087,
      "train_tokens_per_second": 625.878
    },
    {
      "epoch": 0.4953247409653778,
      "grad_norm": 1.1193761825561523,
      "learning_rate": 1.5098734177215191e-05,
      "loss": 0.08151046931743622,
      "num_input_tokens_seen": 923560,
      "step": 490,
      "train_runtime": 1475.0875,
      "train_tokens_per_second": 626.105
    },
    {
      "epoch": 0.4963356077836745,
      "grad_norm": 1.4701436758041382,
      "learning_rate": 1.508860759493671e-05,
      "loss": 0.08011287450790405,
      "num_input_tokens_seen": 925226,
      "step": 491,
      "train_runtime": 1477.3851,
      "train_tokens_per_second": 626.259
    },
    {
      "epoch": 0.4973464746019712,
      "grad_norm": 1.2409087419509888,
      "learning_rate": 1.5078481012658227e-05,
      "loss": 0.08883355557918549,
      "num_input_tokens_seen": 926804,
      "step": 492,
      "train_runtime": 1479.8328,
      "train_tokens_per_second": 626.29
    },
    {
      "epoch": 0.4983573414202679,
      "grad_norm": 1.2009402513504028,
      "learning_rate": 1.5068354430379748e-05,
      "loss": 0.10840057581663132,
      "num_input_tokens_seen": 928740,
      "step": 493,
      "train_runtime": 1482.1156,
      "train_tokens_per_second": 626.631
    },
    {
      "epoch": 0.4993682082385646,
      "grad_norm": 1.3997632265090942,
      "learning_rate": 1.5058227848101267e-05,
      "loss": 0.10527746379375458,
      "num_input_tokens_seen": 930264,
      "step": 494,
      "train_runtime": 1484.3311,
      "train_tokens_per_second": 626.723
    },
    {
      "epoch": 0.5003790750568613,
      "grad_norm": 1.5624264478683472,
      "learning_rate": 1.5048101265822785e-05,
      "loss": 0.09936410188674927,
      "num_input_tokens_seen": 931754,
      "step": 495,
      "train_runtime": 1486.507,
      "train_tokens_per_second": 626.808
    },
    {
      "epoch": 0.501389941875158,
      "grad_norm": 1.3960294723510742,
      "learning_rate": 1.5037974683544306e-05,
      "loss": 0.12032736837863922,
      "num_input_tokens_seen": 933608,
      "step": 496,
      "train_runtime": 1488.7905,
      "train_tokens_per_second": 627.092
    },
    {
      "epoch": 0.5024008086934546,
      "grad_norm": 1.1936594247817993,
      "learning_rate": 1.5027848101265825e-05,
      "loss": 0.10966365039348602,
      "num_input_tokens_seen": 935584,
      "step": 497,
      "train_runtime": 1491.0642,
      "train_tokens_per_second": 627.461
    },
    {
      "epoch": 0.5034116755117514,
      "grad_norm": 1.394915223121643,
      "learning_rate": 1.5017721518987342e-05,
      "loss": 0.0936424508690834,
      "num_input_tokens_seen": 937248,
      "step": 498,
      "train_runtime": 1493.2837,
      "train_tokens_per_second": 627.642
    },
    {
      "epoch": 0.504422542330048,
      "grad_norm": 1.1654257774353027,
      "learning_rate": 1.5007594936708863e-05,
      "loss": 0.08217742294073105,
      "num_input_tokens_seen": 939192,
      "step": 499,
      "train_runtime": 1495.558,
      "train_tokens_per_second": 627.988
    },
    {
      "epoch": 0.5054334091483447,
      "grad_norm": 1.341120719909668,
      "learning_rate": 1.499746835443038e-05,
      "loss": 0.09140294790267944,
      "num_input_tokens_seen": 941142,
      "step": 500,
      "train_runtime": 1497.8542,
      "train_tokens_per_second": 628.327
    },
    {
      "epoch": 0.5064442759666414,
      "grad_norm": 1.223242163658142,
      "learning_rate": 1.49873417721519e-05,
      "loss": 0.07253067195415497,
      "num_input_tokens_seen": 942764,
      "step": 501,
      "train_runtime": 1500.1332,
      "train_tokens_per_second": 628.454
    },
    {
      "epoch": 0.5074551427849381,
      "grad_norm": 1.4446712732315063,
      "learning_rate": 1.497721518987342e-05,
      "loss": 0.11088109761476517,
      "num_input_tokens_seen": 944456,
      "step": 502,
      "train_runtime": 1502.3381,
      "train_tokens_per_second": 628.657
    },
    {
      "epoch": 0.5084660096032347,
      "grad_norm": 1.3642338514328003,
      "learning_rate": 1.4967088607594937e-05,
      "loss": 0.0994197428226471,
      "num_input_tokens_seen": 946438,
      "step": 503,
      "train_runtime": 1504.6147,
      "train_tokens_per_second": 629.023
    },
    {
      "epoch": 0.5094768764215315,
      "grad_norm": 1.311281681060791,
      "learning_rate": 1.4956962025316456e-05,
      "loss": 0.07897685468196869,
      "num_input_tokens_seen": 947942,
      "step": 504,
      "train_runtime": 1506.826,
      "train_tokens_per_second": 629.099
    },
    {
      "epoch": 0.5104877432398282,
      "grad_norm": 1.1747621297836304,
      "learning_rate": 1.4946835443037977e-05,
      "loss": 0.0812889114022255,
      "num_input_tokens_seen": 950170,
      "step": 505,
      "train_runtime": 1509.1522,
      "train_tokens_per_second": 629.605
    },
    {
      "epoch": 0.5114986100581248,
      "grad_norm": 1.2405896186828613,
      "learning_rate": 1.4936708860759495e-05,
      "loss": 0.09990272670984268,
      "num_input_tokens_seen": 952068,
      "step": 506,
      "train_runtime": 1511.3884,
      "train_tokens_per_second": 629.929
    },
    {
      "epoch": 0.5125094768764216,
      "grad_norm": 1.1583400964736938,
      "learning_rate": 1.4926582278481014e-05,
      "loss": 0.08129554241895676,
      "num_input_tokens_seen": 954106,
      "step": 507,
      "train_runtime": 1513.6533,
      "train_tokens_per_second": 630.333
    },
    {
      "epoch": 0.5135203436947182,
      "grad_norm": 1.3690828084945679,
      "learning_rate": 1.4916455696202531e-05,
      "loss": 0.09495596587657928,
      "num_input_tokens_seen": 956278,
      "step": 508,
      "train_runtime": 1515.997,
      "train_tokens_per_second": 630.791
    },
    {
      "epoch": 0.5145312105130149,
      "grad_norm": 1.5271061658859253,
      "learning_rate": 1.4906329113924052e-05,
      "loss": 0.1311858743429184,
      "num_input_tokens_seen": 957986,
      "step": 509,
      "train_runtime": 1518.2556,
      "train_tokens_per_second": 630.978
    },
    {
      "epoch": 0.5155420773313116,
      "grad_norm": 1.2158321142196655,
      "learning_rate": 1.4896202531645571e-05,
      "loss": 0.07281412184238434,
      "num_input_tokens_seen": 960114,
      "step": 510,
      "train_runtime": 1520.5902,
      "train_tokens_per_second": 631.409
    },
    {
      "epoch": 0.5165529441496083,
      "grad_norm": 1.267778992652893,
      "learning_rate": 1.4886075949367088e-05,
      "loss": 0.07296498119831085,
      "num_input_tokens_seen": 961990,
      "step": 511,
      "train_runtime": 1536.972,
      "train_tokens_per_second": 625.9
    },
    {
      "epoch": 0.5175638109679049,
      "grad_norm": 1.342843770980835,
      "learning_rate": 1.487594936708861e-05,
      "loss": 0.0883895754814148,
      "num_input_tokens_seen": 964296,
      "step": 512,
      "train_runtime": 1539.3376,
      "train_tokens_per_second": 626.436
    },
    {
      "epoch": 0.5185746777862017,
      "grad_norm": 1.561376690864563,
      "learning_rate": 1.4865822784810128e-05,
      "loss": 0.11057661473751068,
      "num_input_tokens_seen": 965920,
      "step": 513,
      "train_runtime": 1541.5702,
      "train_tokens_per_second": 626.582
    },
    {
      "epoch": 0.5195855446044984,
      "grad_norm": 1.4014407396316528,
      "learning_rate": 1.4855696202531646e-05,
      "loss": 0.10055877268314362,
      "num_input_tokens_seen": 967894,
      "step": 514,
      "train_runtime": 1543.9261,
      "train_tokens_per_second": 626.904
    },
    {
      "epoch": 0.520596411422795,
      "grad_norm": 1.510994553565979,
      "learning_rate": 1.4845569620253166e-05,
      "loss": 0.09872489422559738,
      "num_input_tokens_seen": 970006,
      "step": 515,
      "train_runtime": 1546.2716,
      "train_tokens_per_second": 627.319
    },
    {
      "epoch": 0.5216072782410918,
      "grad_norm": 1.2100088596343994,
      "learning_rate": 1.4835443037974684e-05,
      "loss": 0.08809270709753036,
      "num_input_tokens_seen": 971940,
      "step": 516,
      "train_runtime": 1548.5422,
      "train_tokens_per_second": 627.648
    },
    {
      "epoch": 0.5226181450593884,
      "grad_norm": 1.283185362815857,
      "learning_rate": 1.4825316455696203e-05,
      "loss": 0.09899348020553589,
      "num_input_tokens_seen": 973842,
      "step": 517,
      "train_runtime": 1550.8105,
      "train_tokens_per_second": 627.957
    },
    {
      "epoch": 0.5236290118776851,
      "grad_norm": 1.7970081567764282,
      "learning_rate": 1.4815189873417724e-05,
      "loss": 0.13198940455913544,
      "num_input_tokens_seen": 975408,
      "step": 518,
      "train_runtime": 1553.0622,
      "train_tokens_per_second": 628.055
    },
    {
      "epoch": 0.5246398786959818,
      "grad_norm": 2.1203253269195557,
      "learning_rate": 1.4805063291139241e-05,
      "loss": 0.10351548343896866,
      "num_input_tokens_seen": 976992,
      "step": 519,
      "train_runtime": 1555.2869,
      "train_tokens_per_second": 628.175
    },
    {
      "epoch": 0.5256507455142785,
      "grad_norm": 1.3503495454788208,
      "learning_rate": 1.479493670886076e-05,
      "loss": 0.1093158945441246,
      "num_input_tokens_seen": 979234,
      "step": 520,
      "train_runtime": 1557.6262,
      "train_tokens_per_second": 628.671
    },
    {
      "epoch": 0.5266616123325751,
      "grad_norm": 1.8989427089691162,
      "learning_rate": 1.4784810126582281e-05,
      "loss": 0.08838169276714325,
      "num_input_tokens_seen": 980844,
      "step": 521,
      "train_runtime": 1559.8377,
      "train_tokens_per_second": 628.812
    },
    {
      "epoch": 0.5276724791508719,
      "grad_norm": 1.437752604484558,
      "learning_rate": 1.4774683544303798e-05,
      "loss": 0.12611472606658936,
      "num_input_tokens_seen": 982764,
      "step": 522,
      "train_runtime": 1562.1103,
      "train_tokens_per_second": 629.126
    },
    {
      "epoch": 0.5286833459691685,
      "grad_norm": 1.428272008895874,
      "learning_rate": 1.4764556962025317e-05,
      "loss": 0.11297731846570969,
      "num_input_tokens_seen": 984420,
      "step": 523,
      "train_runtime": 1564.3837,
      "train_tokens_per_second": 629.27
    },
    {
      "epoch": 0.5296942127874652,
      "grad_norm": 1.6326502561569214,
      "learning_rate": 1.4754430379746836e-05,
      "loss": 0.12765304744243622,
      "num_input_tokens_seen": 986348,
      "step": 524,
      "train_runtime": 1566.6967,
      "train_tokens_per_second": 629.572
    },
    {
      "epoch": 0.530705079605762,
      "grad_norm": 1.2742489576339722,
      "learning_rate": 1.4744303797468356e-05,
      "loss": 0.08136743307113647,
      "num_input_tokens_seen": 988886,
      "step": 525,
      "train_runtime": 1569.2091,
      "train_tokens_per_second": 630.181
    },
    {
      "epoch": 0.5317159464240586,
      "grad_norm": 1.4503283500671387,
      "learning_rate": 1.4734177215189875e-05,
      "loss": 0.11254356801509857,
      "num_input_tokens_seen": 991078,
      "step": 526,
      "train_runtime": 1571.597,
      "train_tokens_per_second": 630.618
    },
    {
      "epoch": 0.5327268132423553,
      "grad_norm": 1.2532612085342407,
      "learning_rate": 1.4724050632911394e-05,
      "loss": 0.06500692665576935,
      "num_input_tokens_seen": 992786,
      "step": 527,
      "train_runtime": 1573.8847,
      "train_tokens_per_second": 630.787
    },
    {
      "epoch": 0.533737680060652,
      "grad_norm": 1.1754192113876343,
      "learning_rate": 1.4713924050632913e-05,
      "loss": 0.07045131176710129,
      "num_input_tokens_seen": 994364,
      "step": 528,
      "train_runtime": 1576.0595,
      "train_tokens_per_second": 630.918
    },
    {
      "epoch": 0.5347485468789487,
      "grad_norm": 1.5875356197357178,
      "learning_rate": 1.4703797468354432e-05,
      "loss": 0.08082123845815659,
      "num_input_tokens_seen": 996310,
      "step": 529,
      "train_runtime": 1578.3601,
      "train_tokens_per_second": 631.231
    },
    {
      "epoch": 0.5357594136972453,
      "grad_norm": 1.5012823343276978,
      "learning_rate": 1.4693670886075951e-05,
      "loss": 0.10939884185791016,
      "num_input_tokens_seen": 998072,
      "step": 530,
      "train_runtime": 1580.6071,
      "train_tokens_per_second": 631.449
    },
    {
      "epoch": 0.5367702805155421,
      "grad_norm": 1.4216855764389038,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.08538244664669037,
      "num_input_tokens_seen": 999422,
      "step": 531,
      "train_runtime": 1582.7805,
      "train_tokens_per_second": 631.434
    },
    {
      "epoch": 0.5377811473338387,
      "grad_norm": 1.4364869594573975,
      "learning_rate": 1.4673417721518987e-05,
      "loss": 0.10728795826435089,
      "num_input_tokens_seen": 1001204,
      "step": 532,
      "train_runtime": 1584.9948,
      "train_tokens_per_second": 631.677
    },
    {
      "epoch": 0.5387920141521354,
      "grad_norm": 1.5693148374557495,
      "learning_rate": 1.4663291139240508e-05,
      "loss": 0.08584736287593842,
      "num_input_tokens_seen": 1002712,
      "step": 533,
      "train_runtime": 1587.1717,
      "train_tokens_per_second": 631.76
    },
    {
      "epoch": 0.5398028809704322,
      "grad_norm": 1.8310790061950684,
      "learning_rate": 1.4653164556962027e-05,
      "loss": 0.14106503129005432,
      "num_input_tokens_seen": 1004918,
      "step": 534,
      "train_runtime": 1589.5276,
      "train_tokens_per_second": 632.212
    },
    {
      "epoch": 0.5408137477887288,
      "grad_norm": 1.524724006652832,
      "learning_rate": 1.4643037974683545e-05,
      "loss": 0.07863933593034744,
      "num_input_tokens_seen": 1006830,
      "step": 535,
      "train_runtime": 1591.8305,
      "train_tokens_per_second": 632.498
    },
    {
      "epoch": 0.5418246146070256,
      "grad_norm": 1.793192744255066,
      "learning_rate": 1.4632911392405065e-05,
      "loss": 0.0889945775270462,
      "num_input_tokens_seen": 1008674,
      "step": 536,
      "train_runtime": 1594.062,
      "train_tokens_per_second": 632.77
    },
    {
      "epoch": 0.5428354814253222,
      "grad_norm": 1.2465921640396118,
      "learning_rate": 1.4622784810126585e-05,
      "loss": 0.09050318598747253,
      "num_input_tokens_seen": 1010454,
      "step": 537,
      "train_runtime": 1596.3341,
      "train_tokens_per_second": 632.984
    },
    {
      "epoch": 0.5438463482436189,
      "grad_norm": 1.1289308071136475,
      "learning_rate": 1.4612658227848102e-05,
      "loss": 0.0836988016963005,
      "num_input_tokens_seen": 1012450,
      "step": 538,
      "train_runtime": 1598.612,
      "train_tokens_per_second": 633.331
    },
    {
      "epoch": 0.5448572150619156,
      "grad_norm": 1.6175777912139893,
      "learning_rate": 1.4602531645569623e-05,
      "loss": 0.09603976458311081,
      "num_input_tokens_seen": 1014482,
      "step": 539,
      "train_runtime": 1600.9696,
      "train_tokens_per_second": 633.667
    },
    {
      "epoch": 0.5458680818802123,
      "grad_norm": 1.4792476892471313,
      "learning_rate": 1.459240506329114e-05,
      "loss": 0.09882150590419769,
      "num_input_tokens_seen": 1016358,
      "step": 540,
      "train_runtime": 1603.2078,
      "train_tokens_per_second": 633.953
    },
    {
      "epoch": 0.546878948698509,
      "grad_norm": 1.4835591316223145,
      "learning_rate": 1.458227848101266e-05,
      "loss": 0.09315601736307144,
      "num_input_tokens_seen": 1018128,
      "step": 541,
      "train_runtime": 1619.5066,
      "train_tokens_per_second": 628.666
    },
    {
      "epoch": 0.5478898155168057,
      "grad_norm": 1.0326731204986572,
      "learning_rate": 1.457215189873418e-05,
      "loss": 0.08888642489910126,
      "num_input_tokens_seen": 1020734,
      "step": 542,
      "train_runtime": 1621.9786,
      "train_tokens_per_second": 629.314
    },
    {
      "epoch": 0.5489006823351024,
      "grad_norm": 1.6520812511444092,
      "learning_rate": 1.4562025316455697e-05,
      "loss": 0.11820662766695023,
      "num_input_tokens_seen": 1022694,
      "step": 543,
      "train_runtime": 1624.2661,
      "train_tokens_per_second": 629.634
    },
    {
      "epoch": 0.549911549153399,
      "grad_norm": 1.2735785245895386,
      "learning_rate": 1.4551898734177216e-05,
      "loss": 0.08194012939929962,
      "num_input_tokens_seen": 1024474,
      "step": 544,
      "train_runtime": 1626.5106,
      "train_tokens_per_second": 629.86
    },
    {
      "epoch": 0.5509224159716958,
      "grad_norm": 1.289209246635437,
      "learning_rate": 1.4541772151898734e-05,
      "loss": 0.10879933834075928,
      "num_input_tokens_seen": 1026564,
      "step": 545,
      "train_runtime": 1628.8592,
      "train_tokens_per_second": 630.235
    },
    {
      "epoch": 0.5519332827899924,
      "grad_norm": 1.1850883960723877,
      "learning_rate": 1.4531645569620255e-05,
      "loss": 0.08871644735336304,
      "num_input_tokens_seen": 1028902,
      "step": 546,
      "train_runtime": 1631.3021,
      "train_tokens_per_second": 630.724
    },
    {
      "epoch": 0.5529441496082891,
      "grad_norm": 1.2800348997116089,
      "learning_rate": 1.4521518987341774e-05,
      "loss": 0.09548143297433853,
      "num_input_tokens_seen": 1030790,
      "step": 547,
      "train_runtime": 1633.5656,
      "train_tokens_per_second": 631.006
    },
    {
      "epoch": 0.5539550164265858,
      "grad_norm": 1.6444133520126343,
      "learning_rate": 1.4511392405063291e-05,
      "loss": 0.09822962433099747,
      "num_input_tokens_seen": 1032664,
      "step": 548,
      "train_runtime": 1635.8492,
      "train_tokens_per_second": 631.271
    },
    {
      "epoch": 0.5549658832448825,
      "grad_norm": 1.2001855373382568,
      "learning_rate": 1.4501265822784812e-05,
      "loss": 0.0989033579826355,
      "num_input_tokens_seen": 1035148,
      "step": 549,
      "train_runtime": 1638.3639,
      "train_tokens_per_second": 631.818
    },
    {
      "epoch": 0.5559767500631791,
      "grad_norm": 1.5036348104476929,
      "learning_rate": 1.4491139240506331e-05,
      "loss": 0.11629313230514526,
      "num_input_tokens_seen": 1037144,
      "step": 550,
      "train_runtime": 1640.6717,
      "train_tokens_per_second": 632.146
    },
    {
      "epoch": 0.5569876168814759,
      "grad_norm": 1.3959003686904907,
      "learning_rate": 1.4481012658227848e-05,
      "loss": 0.08654140681028366,
      "num_input_tokens_seen": 1038914,
      "step": 551,
      "train_runtime": 1642.9292,
      "train_tokens_per_second": 632.355
    },
    {
      "epoch": 0.5579984836997726,
      "grad_norm": 1.6200852394104004,
      "learning_rate": 1.4470886075949369e-05,
      "loss": 0.07833883911371231,
      "num_input_tokens_seen": 1040926,
      "step": 552,
      "train_runtime": 1645.262,
      "train_tokens_per_second": 632.681
    },
    {
      "epoch": 0.5590093505180692,
      "grad_norm": 1.2970929145812988,
      "learning_rate": 1.4460759493670887e-05,
      "loss": 0.07217128574848175,
      "num_input_tokens_seen": 1042626,
      "step": 553,
      "train_runtime": 1647.4423,
      "train_tokens_per_second": 632.876
    },
    {
      "epoch": 0.560020217336366,
      "grad_norm": 1.4967849254608154,
      "learning_rate": 1.4450632911392406e-05,
      "loss": 0.09161016345024109,
      "num_input_tokens_seen": 1044082,
      "step": 554,
      "train_runtime": 1649.6274,
      "train_tokens_per_second": 632.92
    },
    {
      "epoch": 0.5610310841546626,
      "grad_norm": 1.1432979106903076,
      "learning_rate": 1.4440506329113926e-05,
      "loss": 0.075189508497715,
      "num_input_tokens_seen": 1045858,
      "step": 555,
      "train_runtime": 1651.8681,
      "train_tokens_per_second": 633.137
    },
    {
      "epoch": 0.5620419509729593,
      "grad_norm": 1.1668124198913574,
      "learning_rate": 1.4430379746835444e-05,
      "loss": 0.08752522617578506,
      "num_input_tokens_seen": 1047734,
      "step": 556,
      "train_runtime": 1654.143,
      "train_tokens_per_second": 633.4
    },
    {
      "epoch": 0.563052817791256,
      "grad_norm": 1.1975582838058472,
      "learning_rate": 1.4420253164556963e-05,
      "loss": 0.08159657567739487,
      "num_input_tokens_seen": 1049074,
      "step": 557,
      "train_runtime": 1656.2978,
      "train_tokens_per_second": 633.385
    },
    {
      "epoch": 0.5640636846095527,
      "grad_norm": 1.4059653282165527,
      "learning_rate": 1.4410126582278484e-05,
      "loss": 0.11405330151319504,
      "num_input_tokens_seen": 1050914,
      "step": 558,
      "train_runtime": 1658.6954,
      "train_tokens_per_second": 633.579
    },
    {
      "epoch": 0.5650745514278493,
      "grad_norm": 1.376706600189209,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.10115694999694824,
      "num_input_tokens_seen": 1052764,
      "step": 559,
      "train_runtime": 1660.9996,
      "train_tokens_per_second": 633.814
    },
    {
      "epoch": 0.5660854182461461,
      "grad_norm": 1.294089674949646,
      "learning_rate": 1.438987341772152e-05,
      "loss": 0.102215975522995,
      "num_input_tokens_seen": 1054490,
      "step": 560,
      "train_runtime": 1663.3142,
      "train_tokens_per_second": 633.969
    },
    {
      "epoch": 0.5670962850644428,
      "grad_norm": 1.1829617023468018,
      "learning_rate": 1.4379746835443038e-05,
      "loss": 0.08898695558309555,
      "num_input_tokens_seen": 1056460,
      "step": 561,
      "train_runtime": 1665.6588,
      "train_tokens_per_second": 634.26
    },
    {
      "epoch": 0.5681071518827394,
      "grad_norm": 1.4848825931549072,
      "learning_rate": 1.4369620253164558e-05,
      "loss": 0.11753303557634354,
      "num_input_tokens_seen": 1058084,
      "step": 562,
      "train_runtime": 1667.8735,
      "train_tokens_per_second": 634.391
    },
    {
      "epoch": 0.5691180187010362,
      "grad_norm": 1.299782633781433,
      "learning_rate": 1.4359493670886077e-05,
      "loss": 0.10481813549995422,
      "num_input_tokens_seen": 1060308,
      "step": 563,
      "train_runtime": 1670.2485,
      "train_tokens_per_second": 634.821
    },
    {
      "epoch": 0.5701288855193328,
      "grad_norm": 1.2719480991363525,
      "learning_rate": 1.4349367088607595e-05,
      "loss": 0.08594390004873276,
      "num_input_tokens_seen": 1061998,
      "step": 564,
      "train_runtime": 1672.4751,
      "train_tokens_per_second": 634.986
    },
    {
      "epoch": 0.5711397523376295,
      "grad_norm": 1.4669873714447021,
      "learning_rate": 1.4339240506329116e-05,
      "loss": 0.11037176847457886,
      "num_input_tokens_seen": 1063624,
      "step": 565,
      "train_runtime": 1674.7872,
      "train_tokens_per_second": 635.08
    },
    {
      "epoch": 0.5721506191559262,
      "grad_norm": 1.2808465957641602,
      "learning_rate": 1.4329113924050635e-05,
      "loss": 0.0842655748128891,
      "num_input_tokens_seen": 1065408,
      "step": 566,
      "train_runtime": 1677.1211,
      "train_tokens_per_second": 635.26
    },
    {
      "epoch": 0.5731614859742229,
      "grad_norm": 1.0893831253051758,
      "learning_rate": 1.4318987341772152e-05,
      "loss": 0.09920214116573334,
      "num_input_tokens_seen": 1067676,
      "step": 567,
      "train_runtime": 1679.5113,
      "train_tokens_per_second": 635.706
    },
    {
      "epoch": 0.5741723527925195,
      "grad_norm": 1.3617602586746216,
      "learning_rate": 1.4308860759493673e-05,
      "loss": 0.08810853958129883,
      "num_input_tokens_seen": 1069738,
      "step": 568,
      "train_runtime": 1681.8336,
      "train_tokens_per_second": 636.055
    },
    {
      "epoch": 0.5751832196108163,
      "grad_norm": 1.481034517288208,
      "learning_rate": 1.429873417721519e-05,
      "loss": 0.09026068449020386,
      "num_input_tokens_seen": 1071286,
      "step": 569,
      "train_runtime": 1684.0444,
      "train_tokens_per_second": 636.139
    },
    {
      "epoch": 0.576194086429113,
      "grad_norm": 1.343260645866394,
      "learning_rate": 1.428860759493671e-05,
      "loss": 0.08098727464675903,
      "num_input_tokens_seen": 1073526,
      "step": 570,
      "train_runtime": 1686.3793,
      "train_tokens_per_second": 636.586
    },
    {
      "epoch": 0.5772049532474096,
      "grad_norm": 1.2387670278549194,
      "learning_rate": 1.427848101265823e-05,
      "loss": 0.129585862159729,
      "num_input_tokens_seen": 1076090,
      "step": 571,
      "train_runtime": 1703.5398,
      "train_tokens_per_second": 631.679
    },
    {
      "epoch": 0.5782158200657064,
      "grad_norm": 1.4559392929077148,
      "learning_rate": 1.4268354430379747e-05,
      "loss": 0.11523658037185669,
      "num_input_tokens_seen": 1078008,
      "step": 572,
      "train_runtime": 1705.7843,
      "train_tokens_per_second": 631.972
    },
    {
      "epoch": 0.579226686884003,
      "grad_norm": 1.7048641443252563,
      "learning_rate": 1.4258227848101267e-05,
      "loss": 0.12695106863975525,
      "num_input_tokens_seen": 1079674,
      "step": 573,
      "train_runtime": 1708.1147,
      "train_tokens_per_second": 632.085
    },
    {
      "epoch": 0.5802375537022997,
      "grad_norm": 1.0525803565979004,
      "learning_rate": 1.4248101265822787e-05,
      "loss": 0.07779115438461304,
      "num_input_tokens_seen": 1081812,
      "step": 574,
      "train_runtime": 1710.4142,
      "train_tokens_per_second": 632.485
    },
    {
      "epoch": 0.5812484205205964,
      "grad_norm": 1.0833847522735596,
      "learning_rate": 1.4237974683544305e-05,
      "loss": 0.06608449667692184,
      "num_input_tokens_seen": 1084224,
      "step": 575,
      "train_runtime": 1712.7785,
      "train_tokens_per_second": 633.021
    },
    {
      "epoch": 0.5822592873388931,
      "grad_norm": 1.3155786991119385,
      "learning_rate": 1.4227848101265824e-05,
      "loss": 0.10153818130493164,
      "num_input_tokens_seen": 1085852,
      "step": 576,
      "train_runtime": 1714.9937,
      "train_tokens_per_second": 633.152
    },
    {
      "epoch": 0.5832701541571897,
      "grad_norm": 1.1174075603485107,
      "learning_rate": 1.4217721518987341e-05,
      "loss": 0.09092947840690613,
      "num_input_tokens_seen": 1088162,
      "step": 577,
      "train_runtime": 1717.4203,
      "train_tokens_per_second": 633.603
    },
    {
      "epoch": 0.5842810209754865,
      "grad_norm": 1.3650635480880737,
      "learning_rate": 1.4207594936708862e-05,
      "loss": 0.08384772390127182,
      "num_input_tokens_seen": 1089830,
      "step": 578,
      "train_runtime": 1719.6325,
      "train_tokens_per_second": 633.758
    },
    {
      "epoch": 0.5852918877937832,
      "grad_norm": 1.4608466625213623,
      "learning_rate": 1.4197468354430381e-05,
      "loss": 0.09117217361927032,
      "num_input_tokens_seen": 1091450,
      "step": 579,
      "train_runtime": 1721.8127,
      "train_tokens_per_second": 633.896
    },
    {
      "epoch": 0.5863027546120798,
      "grad_norm": 1.2263563871383667,
      "learning_rate": 1.4187341772151898e-05,
      "loss": 0.09236721694469452,
      "num_input_tokens_seen": 1093596,
      "step": 580,
      "train_runtime": 1724.1406,
      "train_tokens_per_second": 634.285
    },
    {
      "epoch": 0.5873136214303766,
      "grad_norm": 1.2389177083969116,
      "learning_rate": 1.417721518987342e-05,
      "loss": 0.1001046821475029,
      "num_input_tokens_seen": 1095482,
      "step": 581,
      "train_runtime": 1726.3722,
      "train_tokens_per_second": 634.557
    },
    {
      "epoch": 0.5883244882486732,
      "grad_norm": 1.4099202156066895,
      "learning_rate": 1.4167088607594938e-05,
      "loss": 0.10450707376003265,
      "num_input_tokens_seen": 1097222,
      "step": 582,
      "train_runtime": 1728.6093,
      "train_tokens_per_second": 634.743
    },
    {
      "epoch": 0.5893353550669699,
      "grad_norm": 1.3976120948791504,
      "learning_rate": 1.4156962025316456e-05,
      "loss": 0.0986606627702713,
      "num_input_tokens_seen": 1099366,
      "step": 583,
      "train_runtime": 1730.9042,
      "train_tokens_per_second": 635.14
    },
    {
      "epoch": 0.5903462218852666,
      "grad_norm": 1.5271971225738525,
      "learning_rate": 1.4146835443037976e-05,
      "loss": 0.11914996802806854,
      "num_input_tokens_seen": 1101232,
      "step": 584,
      "train_runtime": 1733.1427,
      "train_tokens_per_second": 635.396
    },
    {
      "epoch": 0.5913570887035633,
      "grad_norm": 1.394169807434082,
      "learning_rate": 1.4136708860759494e-05,
      "loss": 0.13228291273117065,
      "num_input_tokens_seen": 1103144,
      "step": 585,
      "train_runtime": 1735.4054,
      "train_tokens_per_second": 635.669
    },
    {
      "epoch": 0.59236795552186,
      "grad_norm": 1.1584802865982056,
      "learning_rate": 1.4126582278481013e-05,
      "loss": 0.09044644981622696,
      "num_input_tokens_seen": 1105134,
      "step": 586,
      "train_runtime": 1737.6833,
      "train_tokens_per_second": 635.981
    },
    {
      "epoch": 0.5933788223401567,
      "grad_norm": 1.2238107919692993,
      "learning_rate": 1.4116455696202534e-05,
      "loss": 0.10371394455432892,
      "num_input_tokens_seen": 1106752,
      "step": 587,
      "train_runtime": 1739.8836,
      "train_tokens_per_second": 636.107
    },
    {
      "epoch": 0.5943896891584534,
      "grad_norm": 1.5565840005874634,
      "learning_rate": 1.4106329113924051e-05,
      "loss": 0.093437559902668,
      "num_input_tokens_seen": 1108152,
      "step": 588,
      "train_runtime": 1742.0736,
      "train_tokens_per_second": 636.111
    },
    {
      "epoch": 0.59540055597675,
      "grad_norm": 1.2015377283096313,
      "learning_rate": 1.409620253164557e-05,
      "loss": 0.08773557841777802,
      "num_input_tokens_seen": 1109994,
      "step": 589,
      "train_runtime": 1744.3725,
      "train_tokens_per_second": 636.329
    },
    {
      "epoch": 0.5964114227950468,
      "grad_norm": 1.4085197448730469,
      "learning_rate": 1.4086075949367091e-05,
      "loss": 0.09602763503789902,
      "num_input_tokens_seen": 1112366,
      "step": 590,
      "train_runtime": 1747.0116,
      "train_tokens_per_second": 636.725
    },
    {
      "epoch": 0.5974222896133434,
      "grad_norm": 1.3660284280776978,
      "learning_rate": 1.4075949367088608e-05,
      "loss": 0.11296717822551727,
      "num_input_tokens_seen": 1114230,
      "step": 591,
      "train_runtime": 1749.2776,
      "train_tokens_per_second": 636.966
    },
    {
      "epoch": 0.5984331564316401,
      "grad_norm": 1.5200594663619995,
      "learning_rate": 1.4065822784810127e-05,
      "loss": 0.11103707551956177,
      "num_input_tokens_seen": 1115886,
      "step": 592,
      "train_runtime": 1751.5377,
      "train_tokens_per_second": 637.089
    },
    {
      "epoch": 0.5994440232499368,
      "grad_norm": 1.2209599018096924,
      "learning_rate": 1.4055696202531647e-05,
      "loss": 0.08207274973392487,
      "num_input_tokens_seen": 1117622,
      "step": 593,
      "train_runtime": 1753.7926,
      "train_tokens_per_second": 637.26
    },
    {
      "epoch": 0.6004548900682335,
      "grad_norm": 1.2323637008666992,
      "learning_rate": 1.4045569620253166e-05,
      "loss": 0.08164481818675995,
      "num_input_tokens_seen": 1119962,
      "step": 594,
      "train_runtime": 1756.1607,
      "train_tokens_per_second": 637.733
    },
    {
      "epoch": 0.6004548900682335,
      "eval_loss": 0.3728827238082886,
      "eval_runtime": 54.7389,
      "eval_samples_per_second": 16.058,
      "eval_steps_per_second": 8.038,
      "num_input_tokens_seen": 1119962,
      "step": 594
    },
    {
      "epoch": 0.6014657568865301,
      "grad_norm": 1.4974766969680786,
      "learning_rate": 1.4035443037974685e-05,
      "loss": 0.11136750131845474,
      "num_input_tokens_seen": 1121502,
      "step": 595,
      "train_runtime": 1813.2392,
      "train_tokens_per_second": 618.507
    },
    {
      "epoch": 0.6024766237048269,
      "grad_norm": 1.230702519416809,
      "learning_rate": 1.4025316455696202e-05,
      "loss": 0.07754892855882645,
      "num_input_tokens_seen": 1123712,
      "step": 596,
      "train_runtime": 1815.5803,
      "train_tokens_per_second": 618.927
    },
    {
      "epoch": 0.6034874905231236,
      "grad_norm": 1.3749258518218994,
      "learning_rate": 1.4015189873417723e-05,
      "loss": 0.08618313074111938,
      "num_input_tokens_seen": 1125700,
      "step": 597,
      "train_runtime": 1817.8863,
      "train_tokens_per_second": 619.236
    },
    {
      "epoch": 0.6044983573414203,
      "grad_norm": 1.4063817262649536,
      "learning_rate": 1.4005063291139242e-05,
      "loss": 0.12821811437606812,
      "num_input_tokens_seen": 1127428,
      "step": 598,
      "train_runtime": 1820.1038,
      "train_tokens_per_second": 619.431
    },
    {
      "epoch": 0.605509224159717,
      "grad_norm": 1.7265965938568115,
      "learning_rate": 1.399493670886076e-05,
      "loss": 0.09498798847198486,
      "num_input_tokens_seen": 1128956,
      "step": 599,
      "train_runtime": 1822.2867,
      "train_tokens_per_second": 619.527
    },
    {
      "epoch": 0.6065200909780136,
      "grad_norm": 1.4627681970596313,
      "learning_rate": 1.398481012658228e-05,
      "loss": 0.11261933296918869,
      "num_input_tokens_seen": 1130624,
      "step": 600,
      "train_runtime": 1824.4907,
      "train_tokens_per_second": 619.693
    },
    {
      "epoch": 0.6075309577963104,
      "grad_norm": 1.4472676515579224,
      "learning_rate": 1.3974683544303798e-05,
      "loss": 0.10068300366401672,
      "num_input_tokens_seen": 1132692,
      "step": 601,
      "train_runtime": 1840.2127,
      "train_tokens_per_second": 615.522
    },
    {
      "epoch": 0.608541824614607,
      "grad_norm": 1.8458853960037231,
      "learning_rate": 1.3964556962025317e-05,
      "loss": 0.1499527394771576,
      "num_input_tokens_seen": 1134506,
      "step": 602,
      "train_runtime": 1842.4982,
      "train_tokens_per_second": 615.743
    },
    {
      "epoch": 0.6095526914329037,
      "grad_norm": 1.1479370594024658,
      "learning_rate": 1.3954430379746837e-05,
      "loss": 0.07514189928770065,
      "num_input_tokens_seen": 1136336,
      "step": 603,
      "train_runtime": 1844.7626,
      "train_tokens_per_second": 615.98
    },
    {
      "epoch": 0.6105635582512005,
      "grad_norm": 1.1275358200073242,
      "learning_rate": 1.3944303797468355e-05,
      "loss": 0.07794931530952454,
      "num_input_tokens_seen": 1138190,
      "step": 604,
      "train_runtime": 1847.0661,
      "train_tokens_per_second": 616.215
    },
    {
      "epoch": 0.6115744250694971,
      "grad_norm": 1.5748790502548218,
      "learning_rate": 1.3934177215189874e-05,
      "loss": 0.11096584796905518,
      "num_input_tokens_seen": 1140056,
      "step": 605,
      "train_runtime": 1849.3173,
      "train_tokens_per_second": 616.474
    },
    {
      "epoch": 0.6125852918877938,
      "grad_norm": 1.0246809720993042,
      "learning_rate": 1.3924050632911395e-05,
      "loss": 0.07092858850955963,
      "num_input_tokens_seen": 1141870,
      "step": 606,
      "train_runtime": 1851.5985,
      "train_tokens_per_second": 616.694
    },
    {
      "epoch": 0.6135961587060905,
      "grad_norm": 1.459824562072754,
      "learning_rate": 1.3913924050632912e-05,
      "loss": 0.10918493568897247,
      "num_input_tokens_seen": 1143626,
      "step": 607,
      "train_runtime": 1853.8778,
      "train_tokens_per_second": 616.883
    },
    {
      "epoch": 0.6146070255243872,
      "grad_norm": 1.285499930381775,
      "learning_rate": 1.3903797468354431e-05,
      "loss": 0.11168704926967621,
      "num_input_tokens_seen": 1145798,
      "step": 608,
      "train_runtime": 1856.239,
      "train_tokens_per_second": 617.269
    },
    {
      "epoch": 0.6156178923426838,
      "grad_norm": 1.3664791584014893,
      "learning_rate": 1.389367088607595e-05,
      "loss": 0.09251128137111664,
      "num_input_tokens_seen": 1147482,
      "step": 609,
      "train_runtime": 1858.4667,
      "train_tokens_per_second": 617.435
    },
    {
      "epoch": 0.6166287591609806,
      "grad_norm": 1.304483413696289,
      "learning_rate": 1.388354430379747e-05,
      "loss": 0.08640559017658234,
      "num_input_tokens_seen": 1149396,
      "step": 610,
      "train_runtime": 1860.7328,
      "train_tokens_per_second": 617.711
    },
    {
      "epoch": 0.6176396259792772,
      "grad_norm": 1.1270325183868408,
      "learning_rate": 1.3873417721518988e-05,
      "loss": 0.07292294502258301,
      "num_input_tokens_seen": 1150896,
      "step": 611,
      "train_runtime": 1863.0463,
      "train_tokens_per_second": 617.75
    },
    {
      "epoch": 0.6186504927975739,
      "grad_norm": 1.4046927690505981,
      "learning_rate": 1.3863291139240507e-05,
      "loss": 0.14309613406658173,
      "num_input_tokens_seen": 1153316,
      "step": 612,
      "train_runtime": 1865.5701,
      "train_tokens_per_second": 618.211
    },
    {
      "epoch": 0.6196613596158707,
      "grad_norm": 1.4617398977279663,
      "learning_rate": 1.3853164556962027e-05,
      "loss": 0.10099618881940842,
      "num_input_tokens_seen": 1155122,
      "step": 613,
      "train_runtime": 1867.807,
      "train_tokens_per_second": 618.438
    },
    {
      "epoch": 0.6206722264341673,
      "grad_norm": 1.148921012878418,
      "learning_rate": 1.3843037974683546e-05,
      "loss": 0.07941463589668274,
      "num_input_tokens_seen": 1156720,
      "step": 614,
      "train_runtime": 1869.983,
      "train_tokens_per_second": 618.572
    },
    {
      "epoch": 0.621683093252464,
      "grad_norm": 1.072580099105835,
      "learning_rate": 1.3832911392405065e-05,
      "loss": 0.07684888690710068,
      "num_input_tokens_seen": 1158748,
      "step": 615,
      "train_runtime": 1872.2934,
      "train_tokens_per_second": 618.892
    },
    {
      "epoch": 0.6226939600707607,
      "grad_norm": 1.4846950769424438,
      "learning_rate": 1.3822784810126584e-05,
      "loss": 0.12174379825592041,
      "num_input_tokens_seen": 1160814,
      "step": 616,
      "train_runtime": 1874.6157,
      "train_tokens_per_second": 619.228
    },
    {
      "epoch": 0.6237048268890574,
      "grad_norm": 1.6231915950775146,
      "learning_rate": 1.3812658227848101e-05,
      "loss": 0.13438279926776886,
      "num_input_tokens_seen": 1162850,
      "step": 617,
      "train_runtime": 1876.9235,
      "train_tokens_per_second": 619.551
    },
    {
      "epoch": 0.624715693707354,
      "grad_norm": 1.226564884185791,
      "learning_rate": 1.3802531645569622e-05,
      "loss": 0.0925150066614151,
      "num_input_tokens_seen": 1165406,
      "step": 618,
      "train_runtime": 1879.3431,
      "train_tokens_per_second": 620.113
    },
    {
      "epoch": 0.6257265605256508,
      "grad_norm": 1.6471208333969116,
      "learning_rate": 1.3792405063291141e-05,
      "loss": 0.1100998967885971,
      "num_input_tokens_seen": 1167150,
      "step": 619,
      "train_runtime": 1881.5797,
      "train_tokens_per_second": 620.303
    },
    {
      "epoch": 0.6267374273439474,
      "grad_norm": 1.223799467086792,
      "learning_rate": 1.3782278481012658e-05,
      "loss": 0.10464879870414734,
      "num_input_tokens_seen": 1169178,
      "step": 620,
      "train_runtime": 1883.8878,
      "train_tokens_per_second": 620.62
    },
    {
      "epoch": 0.6277482941622441,
      "grad_norm": 1.2662726640701294,
      "learning_rate": 1.377215189873418e-05,
      "loss": 0.09581077098846436,
      "num_input_tokens_seen": 1170776,
      "step": 621,
      "train_runtime": 1886.1062,
      "train_tokens_per_second": 620.737
    },
    {
      "epoch": 0.6287591609805409,
      "grad_norm": 1.2782937288284302,
      "learning_rate": 1.3762025316455698e-05,
      "loss": 0.07302618026733398,
      "num_input_tokens_seen": 1172508,
      "step": 622,
      "train_runtime": 1888.3373,
      "train_tokens_per_second": 620.921
    },
    {
      "epoch": 0.6297700277988375,
      "grad_norm": 1.15910005569458,
      "learning_rate": 1.3751898734177216e-05,
      "loss": 0.09295112639665604,
      "num_input_tokens_seen": 1174694,
      "step": 623,
      "train_runtime": 1890.6942,
      "train_tokens_per_second": 621.303
    },
    {
      "epoch": 0.6307808946171342,
      "grad_norm": 1.6547151803970337,
      "learning_rate": 1.3741772151898736e-05,
      "loss": 0.08670621365308762,
      "num_input_tokens_seen": 1176490,
      "step": 624,
      "train_runtime": 1892.9669,
      "train_tokens_per_second": 621.506
    },
    {
      "epoch": 0.6317917614354309,
      "grad_norm": 1.1826273202896118,
      "learning_rate": 1.3731645569620254e-05,
      "loss": 0.07737353444099426,
      "num_input_tokens_seen": 1178784,
      "step": 625,
      "train_runtime": 1895.3074,
      "train_tokens_per_second": 621.949
    },
    {
      "epoch": 0.6328026282537276,
      "grad_norm": 1.4511724710464478,
      "learning_rate": 1.3721518987341773e-05,
      "loss": 0.09621845930814743,
      "num_input_tokens_seen": 1180080,
      "step": 626,
      "train_runtime": 1897.5065,
      "train_tokens_per_second": 621.911
    },
    {
      "epoch": 0.6338134950720242,
      "grad_norm": 1.4344146251678467,
      "learning_rate": 1.3711392405063294e-05,
      "loss": 0.07546381652355194,
      "num_input_tokens_seen": 1181322,
      "step": 627,
      "train_runtime": 1899.716,
      "train_tokens_per_second": 621.841
    },
    {
      "epoch": 0.634824361890321,
      "grad_norm": 1.3891464471817017,
      "learning_rate": 1.3701265822784811e-05,
      "loss": 0.13230165839195251,
      "num_input_tokens_seen": 1183382,
      "step": 628,
      "train_runtime": 1902.0587,
      "train_tokens_per_second": 622.159
    },
    {
      "epoch": 0.6358352287086176,
      "grad_norm": 1.2920548915863037,
      "learning_rate": 1.369113924050633e-05,
      "loss": 0.10276541858911514,
      "num_input_tokens_seen": 1185392,
      "step": 629,
      "train_runtime": 1904.3721,
      "train_tokens_per_second": 622.458
    },
    {
      "epoch": 0.6368460955269143,
      "grad_norm": 1.4265518188476562,
      "learning_rate": 1.3681012658227851e-05,
      "loss": 0.1239503026008606,
      "num_input_tokens_seen": 1187446,
      "step": 630,
      "train_runtime": 1906.7148,
      "train_tokens_per_second": 622.771
    },
    {
      "epoch": 0.637856962345211,
      "grad_norm": 1.3055717945098877,
      "learning_rate": 1.3670886075949368e-05,
      "loss": 0.09938111156225204,
      "num_input_tokens_seen": 1189472,
      "step": 631,
      "train_runtime": 1922.3408,
      "train_tokens_per_second": 618.762
    },
    {
      "epoch": 0.6388678291635077,
      "grad_norm": 1.341225266456604,
      "learning_rate": 1.3660759493670887e-05,
      "loss": 0.10700959712266922,
      "num_input_tokens_seen": 1191462,
      "step": 632,
      "train_runtime": 1924.6437,
      "train_tokens_per_second": 619.056
    },
    {
      "epoch": 0.6398786959818044,
      "grad_norm": 1.4074451923370361,
      "learning_rate": 1.3650632911392405e-05,
      "loss": 0.1088184043765068,
      "num_input_tokens_seen": 1193630,
      "step": 633,
      "train_runtime": 1926.982,
      "train_tokens_per_second": 619.43
    },
    {
      "epoch": 0.6408895628001011,
      "grad_norm": 1.350207805633545,
      "learning_rate": 1.3640506329113926e-05,
      "loss": 0.09866318106651306,
      "num_input_tokens_seen": 1195248,
      "step": 634,
      "train_runtime": 1929.25,
      "train_tokens_per_second": 619.54
    },
    {
      "epoch": 0.6419004296183978,
      "grad_norm": 1.2634080648422241,
      "learning_rate": 1.3630379746835445e-05,
      "loss": 0.10193081945180893,
      "num_input_tokens_seen": 1197508,
      "step": 635,
      "train_runtime": 1931.638,
      "train_tokens_per_second": 619.944
    },
    {
      "epoch": 0.6429112964366944,
      "grad_norm": 1.1352425813674927,
      "learning_rate": 1.3620253164556962e-05,
      "loss": 0.09427931904792786,
      "num_input_tokens_seen": 1199538,
      "step": 636,
      "train_runtime": 1934.0314,
      "train_tokens_per_second": 620.227
    },
    {
      "epoch": 0.6439221632549912,
      "grad_norm": 1.3848373889923096,
      "learning_rate": 1.3610126582278483e-05,
      "loss": 0.10558828711509705,
      "num_input_tokens_seen": 1201214,
      "step": 637,
      "train_runtime": 1936.3585,
      "train_tokens_per_second": 620.347
    },
    {
      "epoch": 0.6449330300732878,
      "grad_norm": 1.2026723623275757,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 0.08852043747901917,
      "num_input_tokens_seen": 1202928,
      "step": 638,
      "train_runtime": 1938.6621,
      "train_tokens_per_second": 620.494
    },
    {
      "epoch": 0.6459438968915845,
      "grad_norm": 1.1012513637542725,
      "learning_rate": 1.358987341772152e-05,
      "loss": 0.08230563998222351,
      "num_input_tokens_seen": 1205536,
      "step": 639,
      "train_runtime": 1941.1067,
      "train_tokens_per_second": 621.056
    },
    {
      "epoch": 0.6469547637098813,
      "grad_norm": 1.008036732673645,
      "learning_rate": 1.357974683544304e-05,
      "loss": 0.09450970590114594,
      "num_input_tokens_seen": 1208210,
      "step": 640,
      "train_runtime": 1943.6072,
      "train_tokens_per_second": 621.633
    },
    {
      "epoch": 0.6479656305281779,
      "grad_norm": 1.4536775350570679,
      "learning_rate": 1.3569620253164557e-05,
      "loss": 0.10127629339694977,
      "num_input_tokens_seen": 1209926,
      "step": 641,
      "train_runtime": 1945.9397,
      "train_tokens_per_second": 621.77
    },
    {
      "epoch": 0.6489764973464746,
      "grad_norm": 1.9795511960983276,
      "learning_rate": 1.3559493670886077e-05,
      "loss": 0.15318067371845245,
      "num_input_tokens_seen": 1211520,
      "step": 642,
      "train_runtime": 1948.1584,
      "train_tokens_per_second": 621.88
    },
    {
      "epoch": 0.6499873641647713,
      "grad_norm": 1.412170648574829,
      "learning_rate": 1.3549367088607597e-05,
      "loss": 0.09511595219373703,
      "num_input_tokens_seen": 1213840,
      "step": 643,
      "train_runtime": 1950.588,
      "train_tokens_per_second": 622.294
    },
    {
      "epoch": 0.650998230983068,
      "grad_norm": 1.7817212343215942,
      "learning_rate": 1.3539240506329115e-05,
      "loss": 0.14267957210540771,
      "num_input_tokens_seen": 1215450,
      "step": 644,
      "train_runtime": 1952.8277,
      "train_tokens_per_second": 622.405
    },
    {
      "epoch": 0.6520090978013646,
      "grad_norm": 1.1762109994888306,
      "learning_rate": 1.3529113924050634e-05,
      "loss": 0.10678732395172119,
      "num_input_tokens_seen": 1217814,
      "step": 645,
      "train_runtime": 1955.1947,
      "train_tokens_per_second": 622.861
    },
    {
      "epoch": 0.6530199646196614,
      "grad_norm": 1.2385965585708618,
      "learning_rate": 1.3518987341772155e-05,
      "loss": 0.11040400713682175,
      "num_input_tokens_seen": 1220000,
      "step": 646,
      "train_runtime": 1957.5921,
      "train_tokens_per_second": 623.215
    },
    {
      "epoch": 0.654030831437958,
      "grad_norm": 1.7378219366073608,
      "learning_rate": 1.3508860759493672e-05,
      "loss": 0.09476306289434433,
      "num_input_tokens_seen": 1221386,
      "step": 647,
      "train_runtime": 1959.7685,
      "train_tokens_per_second": 623.23
    },
    {
      "epoch": 0.6550416982562547,
      "grad_norm": 1.2294435501098633,
      "learning_rate": 1.3498734177215191e-05,
      "loss": 0.08711254596710205,
      "num_input_tokens_seen": 1222972,
      "step": 648,
      "train_runtime": 1962.0276,
      "train_tokens_per_second": 623.32
    },
    {
      "epoch": 0.6560525650745515,
      "grad_norm": 1.0858039855957031,
      "learning_rate": 1.3488607594936708e-05,
      "loss": 0.08094838261604309,
      "num_input_tokens_seen": 1224776,
      "step": 649,
      "train_runtime": 1964.3018,
      "train_tokens_per_second": 623.517
    },
    {
      "epoch": 0.6570634318928481,
      "grad_norm": 1.4281175136566162,
      "learning_rate": 1.347848101265823e-05,
      "loss": 0.10982443392276764,
      "num_input_tokens_seen": 1226784,
      "step": 650,
      "train_runtime": 1966.5943,
      "train_tokens_per_second": 623.811
    },
    {
      "epoch": 0.6580742987111448,
      "grad_norm": 1.8370027542114258,
      "learning_rate": 1.3468354430379748e-05,
      "loss": 0.1082611083984375,
      "num_input_tokens_seen": 1228484,
      "step": 651,
      "train_runtime": 1968.8016,
      "train_tokens_per_second": 623.976
    },
    {
      "epoch": 0.6590851655294415,
      "grad_norm": 1.2840733528137207,
      "learning_rate": 1.3458227848101266e-05,
      "loss": 0.08515030145645142,
      "num_input_tokens_seen": 1230210,
      "step": 652,
      "train_runtime": 1970.9987,
      "train_tokens_per_second": 624.156
    },
    {
      "epoch": 0.6600960323477382,
      "grad_norm": 1.6538041830062866,
      "learning_rate": 1.3448101265822786e-05,
      "loss": 0.14575600624084473,
      "num_input_tokens_seen": 1232078,
      "step": 653,
      "train_runtime": 1973.2675,
      "train_tokens_per_second": 624.385
    },
    {
      "epoch": 0.6611068991660348,
      "grad_norm": 1.4147297143936157,
      "learning_rate": 1.3437974683544306e-05,
      "loss": 0.09866475313901901,
      "num_input_tokens_seen": 1233970,
      "step": 654,
      "train_runtime": 1975.5126,
      "train_tokens_per_second": 624.633
    },
    {
      "epoch": 0.6621177659843316,
      "grad_norm": 1.150806188583374,
      "learning_rate": 1.3427848101265823e-05,
      "loss": 0.09834861755371094,
      "num_input_tokens_seen": 1235770,
      "step": 655,
      "train_runtime": 1977.7801,
      "train_tokens_per_second": 624.827
    },
    {
      "epoch": 0.6631286328026282,
      "grad_norm": 1.2239781618118286,
      "learning_rate": 1.3417721518987344e-05,
      "loss": 0.09434303641319275,
      "num_input_tokens_seen": 1237816,
      "step": 656,
      "train_runtime": 1980.0888,
      "train_tokens_per_second": 625.132
    },
    {
      "epoch": 0.6641394996209249,
      "grad_norm": 1.286088228225708,
      "learning_rate": 1.3407594936708861e-05,
      "loss": 0.08801987767219543,
      "num_input_tokens_seen": 1240466,
      "step": 657,
      "train_runtime": 1982.5757,
      "train_tokens_per_second": 625.684
    },
    {
      "epoch": 0.6651503664392217,
      "grad_norm": 1.1865568161010742,
      "learning_rate": 1.339746835443038e-05,
      "loss": 0.08435536921024323,
      "num_input_tokens_seen": 1242766,
      "step": 658,
      "train_runtime": 1984.9526,
      "train_tokens_per_second": 626.094
    },
    {
      "epoch": 0.6661612332575183,
      "grad_norm": 1.3380271196365356,
      "learning_rate": 1.3387341772151901e-05,
      "loss": 0.08816331624984741,
      "num_input_tokens_seen": 1244288,
      "step": 659,
      "train_runtime": 1987.1526,
      "train_tokens_per_second": 626.166
    },
    {
      "epoch": 0.6671721000758151,
      "grad_norm": 1.245286464691162,
      "learning_rate": 1.3377215189873418e-05,
      "loss": 0.063364677131176,
      "num_input_tokens_seen": 1245836,
      "step": 660,
      "train_runtime": 1989.3989,
      "train_tokens_per_second": 626.237
    },
    {
      "epoch": 0.6681829668941117,
      "grad_norm": 1.1613205671310425,
      "learning_rate": 1.3367088607594937e-05,
      "loss": 0.0773451030254364,
      "num_input_tokens_seen": 1247956,
      "step": 661,
      "train_runtime": 2005.1443,
      "train_tokens_per_second": 622.377
    },
    {
      "epoch": 0.6691938337124084,
      "grad_norm": 1.5810867547988892,
      "learning_rate": 1.3356962025316458e-05,
      "loss": 0.09334032237529755,
      "num_input_tokens_seen": 1249858,
      "step": 662,
      "train_runtime": 2007.4394,
      "train_tokens_per_second": 622.613
    },
    {
      "epoch": 0.6702047005307051,
      "grad_norm": 1.2277162075042725,
      "learning_rate": 1.3346835443037976e-05,
      "loss": 0.11391384154558182,
      "num_input_tokens_seen": 1252000,
      "step": 663,
      "train_runtime": 2009.8695,
      "train_tokens_per_second": 622.926
    },
    {
      "epoch": 0.6712155673490018,
      "grad_norm": 1.1767802238464355,
      "learning_rate": 1.3336708860759495e-05,
      "loss": 0.06979891657829285,
      "num_input_tokens_seen": 1253574,
      "step": 664,
      "train_runtime": 2012.0865,
      "train_tokens_per_second": 623.022
    },
    {
      "epoch": 0.6722264341672984,
      "grad_norm": 1.2678321599960327,
      "learning_rate": 1.3326582278481012e-05,
      "loss": 0.07842098921537399,
      "num_input_tokens_seen": 1255260,
      "step": 665,
      "train_runtime": 2014.2821,
      "train_tokens_per_second": 623.18
    },
    {
      "epoch": 0.6732373009855952,
      "grad_norm": 1.0496454238891602,
      "learning_rate": 1.3316455696202533e-05,
      "loss": 0.10145942866802216,
      "num_input_tokens_seen": 1257338,
      "step": 666,
      "train_runtime": 2016.6438,
      "train_tokens_per_second": 623.48
    },
    {
      "epoch": 0.6742481678038919,
      "grad_norm": 1.7913352251052856,
      "learning_rate": 1.3306329113924052e-05,
      "loss": 0.10923612862825394,
      "num_input_tokens_seen": 1259314,
      "step": 667,
      "train_runtime": 2018.9446,
      "train_tokens_per_second": 623.749
    },
    {
      "epoch": 0.6752590346221885,
      "grad_norm": 1.2996304035186768,
      "learning_rate": 1.329620253164557e-05,
      "loss": 0.08440631628036499,
      "num_input_tokens_seen": 1261426,
      "step": 668,
      "train_runtime": 2021.2342,
      "train_tokens_per_second": 624.087
    },
    {
      "epoch": 0.6762699014404853,
      "grad_norm": 1.1881675720214844,
      "learning_rate": 1.328607594936709e-05,
      "loss": 0.08615678548812866,
      "num_input_tokens_seen": 1263376,
      "step": 669,
      "train_runtime": 2023.5579,
      "train_tokens_per_second": 624.334
    },
    {
      "epoch": 0.6772807682587819,
      "grad_norm": 1.4807711839675903,
      "learning_rate": 1.327594936708861e-05,
      "loss": 0.09940747916698456,
      "num_input_tokens_seen": 1265248,
      "step": 670,
      "train_runtime": 2025.958,
      "train_tokens_per_second": 624.518
    },
    {
      "epoch": 0.6782916350770786,
      "grad_norm": 1.29509699344635,
      "learning_rate": 1.3265822784810127e-05,
      "loss": 0.07858435809612274,
      "num_input_tokens_seen": 1267244,
      "step": 671,
      "train_runtime": 2028.3122,
      "train_tokens_per_second": 624.778
    },
    {
      "epoch": 0.6793025018953753,
      "grad_norm": 1.6262776851654053,
      "learning_rate": 1.3255696202531647e-05,
      "loss": 0.09985235333442688,
      "num_input_tokens_seen": 1269154,
      "step": 672,
      "train_runtime": 2030.6226,
      "train_tokens_per_second": 625.007
    },
    {
      "epoch": 0.680313368713672,
      "grad_norm": 1.2777587175369263,
      "learning_rate": 1.3245569620253165e-05,
      "loss": 0.05365104600787163,
      "num_input_tokens_seen": 1270498,
      "step": 673,
      "train_runtime": 2032.7512,
      "train_tokens_per_second": 625.014
    },
    {
      "epoch": 0.6813242355319686,
      "grad_norm": 1.5250898599624634,
      "learning_rate": 1.3235443037974684e-05,
      "loss": 0.12705999612808228,
      "num_input_tokens_seen": 1272808,
      "step": 674,
      "train_runtime": 2035.1808,
      "train_tokens_per_second": 625.403
    },
    {
      "epoch": 0.6823351023502654,
      "grad_norm": 1.2472586631774902,
      "learning_rate": 1.3225316455696205e-05,
      "loss": 0.08404464274644852,
      "num_input_tokens_seen": 1274488,
      "step": 675,
      "train_runtime": 2037.445,
      "train_tokens_per_second": 625.532
    },
    {
      "epoch": 0.683345969168562,
      "grad_norm": 1.4096324443817139,
      "learning_rate": 1.3215189873417722e-05,
      "loss": 0.09273739159107208,
      "num_input_tokens_seen": 1275734,
      "step": 676,
      "train_runtime": 2039.7406,
      "train_tokens_per_second": 625.439
    },
    {
      "epoch": 0.6843568359868587,
      "grad_norm": 1.3529670238494873,
      "learning_rate": 1.3205063291139241e-05,
      "loss": 0.0949849858880043,
      "num_input_tokens_seen": 1278096,
      "step": 677,
      "train_runtime": 2042.2298,
      "train_tokens_per_second": 625.834
    },
    {
      "epoch": 0.6853677028051555,
      "grad_norm": 1.4302928447723389,
      "learning_rate": 1.3194936708860762e-05,
      "loss": 0.0908072367310524,
      "num_input_tokens_seen": 1279612,
      "step": 678,
      "train_runtime": 2044.4393,
      "train_tokens_per_second": 625.899
    },
    {
      "epoch": 0.6863785696234521,
      "grad_norm": 1.5077518224716187,
      "learning_rate": 1.318481012658228e-05,
      "loss": 0.12817761301994324,
      "num_input_tokens_seen": 1281712,
      "step": 679,
      "train_runtime": 2046.7869,
      "train_tokens_per_second": 626.207
    },
    {
      "epoch": 0.6873894364417488,
      "grad_norm": 1.6469041109085083,
      "learning_rate": 1.3174683544303798e-05,
      "loss": 0.11668027937412262,
      "num_input_tokens_seen": 1284012,
      "step": 680,
      "train_runtime": 2049.1493,
      "train_tokens_per_second": 626.607
    },
    {
      "epoch": 0.6884003032600455,
      "grad_norm": 1.4139769077301025,
      "learning_rate": 1.3164556962025317e-05,
      "loss": 0.06819451600313187,
      "num_input_tokens_seen": 1285832,
      "step": 681,
      "train_runtime": 2051.4208,
      "train_tokens_per_second": 626.801
    },
    {
      "epoch": 0.6894111700783422,
      "grad_norm": 1.3623279333114624,
      "learning_rate": 1.3154430379746837e-05,
      "loss": 0.1149786114692688,
      "num_input_tokens_seen": 1287984,
      "step": 682,
      "train_runtime": 2053.8291,
      "train_tokens_per_second": 627.114
    },
    {
      "epoch": 0.6904220368966388,
      "grad_norm": 1.9275081157684326,
      "learning_rate": 1.3144303797468356e-05,
      "loss": 0.1380566954612732,
      "num_input_tokens_seen": 1289848,
      "step": 683,
      "train_runtime": 2056.1008,
      "train_tokens_per_second": 627.327
    },
    {
      "epoch": 0.6914329037149356,
      "grad_norm": 1.4662679433822632,
      "learning_rate": 1.3134177215189875e-05,
      "loss": 0.11257638037204742,
      "num_input_tokens_seen": 1291486,
      "step": 684,
      "train_runtime": 2058.342,
      "train_tokens_per_second": 627.44
    },
    {
      "epoch": 0.6924437705332323,
      "grad_norm": 1.4109159708023071,
      "learning_rate": 1.3124050632911394e-05,
      "loss": 0.11249294131994247,
      "num_input_tokens_seen": 1293226,
      "step": 685,
      "train_runtime": 2060.5782,
      "train_tokens_per_second": 627.603
    },
    {
      "epoch": 0.6934546373515289,
      "grad_norm": 1.3101881742477417,
      "learning_rate": 1.3113924050632913e-05,
      "loss": 0.08162788301706314,
      "num_input_tokens_seen": 1294980,
      "step": 686,
      "train_runtime": 2062.8207,
      "train_tokens_per_second": 627.771
    },
    {
      "epoch": 0.6944655041698257,
      "grad_norm": 1.2676407098770142,
      "learning_rate": 1.310379746835443e-05,
      "loss": 0.07914060354232788,
      "num_input_tokens_seen": 1296604,
      "step": 687,
      "train_runtime": 2065.0577,
      "train_tokens_per_second": 627.878
    },
    {
      "epoch": 0.6954763709881223,
      "grad_norm": 1.3755985498428345,
      "learning_rate": 1.3093670886075951e-05,
      "loss": 0.08941030502319336,
      "num_input_tokens_seen": 1298266,
      "step": 688,
      "train_runtime": 2067.348,
      "train_tokens_per_second": 627.986
    },
    {
      "epoch": 0.696487237806419,
      "grad_norm": 1.135475993156433,
      "learning_rate": 1.3083544303797468e-05,
      "loss": 0.08094897866249084,
      "num_input_tokens_seen": 1300102,
      "step": 689,
      "train_runtime": 2069.7822,
      "train_tokens_per_second": 628.135
    },
    {
      "epoch": 0.6974981046247157,
      "grad_norm": 1.3128398656845093,
      "learning_rate": 1.3073417721518988e-05,
      "loss": 0.10910716652870178,
      "num_input_tokens_seen": 1301776,
      "step": 690,
      "train_runtime": 2071.9913,
      "train_tokens_per_second": 628.273
    },
    {
      "epoch": 0.6985089714430124,
      "grad_norm": 1.333283543586731,
      "learning_rate": 1.3063291139240508e-05,
      "loss": 0.07274118065834045,
      "num_input_tokens_seen": 1303442,
      "step": 691,
      "train_runtime": 2087.6634,
      "train_tokens_per_second": 624.354
    },
    {
      "epoch": 0.699519838261309,
      "grad_norm": 1.4998698234558105,
      "learning_rate": 1.3053164556962026e-05,
      "loss": 0.11137869209051132,
      "num_input_tokens_seen": 1305156,
      "step": 692,
      "train_runtime": 2089.8745,
      "train_tokens_per_second": 624.514
    },
    {
      "epoch": 0.7005307050796058,
      "grad_norm": 1.498834252357483,
      "learning_rate": 1.3043037974683545e-05,
      "loss": 0.13288113474845886,
      "num_input_tokens_seen": 1306902,
      "step": 693,
      "train_runtime": 2092.1851,
      "train_tokens_per_second": 624.659
    },
    {
      "epoch": 0.7015415718979024,
      "grad_norm": 1.2488044500350952,
      "learning_rate": 1.3032911392405066e-05,
      "loss": 0.0887034684419632,
      "num_input_tokens_seen": 1308664,
      "step": 694,
      "train_runtime": 2094.4446,
      "train_tokens_per_second": 624.826
    },
    {
      "epoch": 0.7025524387161991,
      "grad_norm": 1.2471626996994019,
      "learning_rate": 1.3022784810126583e-05,
      "loss": 0.09268858283758163,
      "num_input_tokens_seen": 1310814,
      "step": 695,
      "train_runtime": 2096.7758,
      "train_tokens_per_second": 625.157
    },
    {
      "epoch": 0.7035633055344959,
      "grad_norm": 1.370259404182434,
      "learning_rate": 1.3012658227848102e-05,
      "loss": 0.09940025955438614,
      "num_input_tokens_seen": 1312786,
      "step": 696,
      "train_runtime": 2099.1036,
      "train_tokens_per_second": 625.403
    },
    {
      "epoch": 0.7045741723527925,
      "grad_norm": 1.0626189708709717,
      "learning_rate": 1.3002531645569621e-05,
      "loss": 0.07294227927923203,
      "num_input_tokens_seen": 1315162,
      "step": 697,
      "train_runtime": 2101.5378,
      "train_tokens_per_second": 625.809
    },
    {
      "epoch": 0.7055850391710892,
      "grad_norm": 1.2751500606536865,
      "learning_rate": 1.299240506329114e-05,
      "loss": 0.10228762775659561,
      "num_input_tokens_seen": 1316704,
      "step": 698,
      "train_runtime": 2103.8802,
      "train_tokens_per_second": 625.846
    },
    {
      "epoch": 0.7065959059893859,
      "grad_norm": 1.4122878313064575,
      "learning_rate": 1.298227848101266e-05,
      "loss": 0.09768881648778915,
      "num_input_tokens_seen": 1318362,
      "step": 699,
      "train_runtime": 2106.117,
      "train_tokens_per_second": 625.968
    },
    {
      "epoch": 0.7076067728076826,
      "grad_norm": 1.5043821334838867,
      "learning_rate": 1.2972151898734178e-05,
      "loss": 0.11154785752296448,
      "num_input_tokens_seen": 1319922,
      "step": 700,
      "train_runtime": 2108.342,
      "train_tokens_per_second": 626.047
    },
    {
      "epoch": 0.7086176396259792,
      "grad_norm": 1.0825846195220947,
      "learning_rate": 1.2962025316455697e-05,
      "loss": 0.08886873722076416,
      "num_input_tokens_seen": 1322010,
      "step": 701,
      "train_runtime": 2110.684,
      "train_tokens_per_second": 626.342
    },
    {
      "epoch": 0.709628506444276,
      "grad_norm": 1.2912213802337646,
      "learning_rate": 1.2951898734177217e-05,
      "loss": 0.10815602540969849,
      "num_input_tokens_seen": 1324172,
      "step": 702,
      "train_runtime": 2113.0478,
      "train_tokens_per_second": 626.664
    },
    {
      "epoch": 0.7106393732625726,
      "grad_norm": 1.1536753177642822,
      "learning_rate": 1.2941772151898736e-05,
      "loss": 0.07558616250753403,
      "num_input_tokens_seen": 1325838,
      "step": 703,
      "train_runtime": 2115.2376,
      "train_tokens_per_second": 626.803
    },
    {
      "epoch": 0.7116502400808693,
      "grad_norm": 1.50513756275177,
      "learning_rate": 1.2931645569620255e-05,
      "loss": 0.10250692069530487,
      "num_input_tokens_seen": 1327436,
      "step": 704,
      "train_runtime": 2117.4109,
      "train_tokens_per_second": 626.915
    },
    {
      "epoch": 0.7126611068991661,
      "grad_norm": 1.2506285905838013,
      "learning_rate": 1.2921518987341772e-05,
      "loss": 0.08238071203231812,
      "num_input_tokens_seen": 1329294,
      "step": 705,
      "train_runtime": 2119.6936,
      "train_tokens_per_second": 627.116
    },
    {
      "epoch": 0.7136719737174627,
      "grad_norm": 1.3678765296936035,
      "learning_rate": 1.2911392405063293e-05,
      "loss": 0.08024592697620392,
      "num_input_tokens_seen": 1331232,
      "step": 706,
      "train_runtime": 2121.9591,
      "train_tokens_per_second": 627.36
    },
    {
      "epoch": 0.7146828405357594,
      "grad_norm": 1.443300724029541,
      "learning_rate": 1.2901265822784812e-05,
      "loss": 0.12251617014408112,
      "num_input_tokens_seen": 1333056,
      "step": 707,
      "train_runtime": 2124.2655,
      "train_tokens_per_second": 627.537
    },
    {
      "epoch": 0.7156937073540561,
      "grad_norm": 1.2339009046554565,
      "learning_rate": 1.289113924050633e-05,
      "loss": 0.10755623877048492,
      "num_input_tokens_seen": 1334582,
      "step": 708,
      "train_runtime": 2126.5397,
      "train_tokens_per_second": 627.584
    },
    {
      "epoch": 0.7167045741723528,
      "grad_norm": 1.3110127449035645,
      "learning_rate": 1.288101265822785e-05,
      "loss": 0.08711139112710953,
      "num_input_tokens_seen": 1336404,
      "step": 709,
      "train_runtime": 2128.8117,
      "train_tokens_per_second": 627.77
    },
    {
      "epoch": 0.7177154409906494,
      "grad_norm": 1.476033091545105,
      "learning_rate": 1.287088607594937e-05,
      "loss": 0.12103165686130524,
      "num_input_tokens_seen": 1338188,
      "step": 710,
      "train_runtime": 2131.0517,
      "train_tokens_per_second": 627.947
    },
    {
      "epoch": 0.7187263078089462,
      "grad_norm": 1.1368860006332397,
      "learning_rate": 1.2860759493670887e-05,
      "loss": 0.09072631597518921,
      "num_input_tokens_seen": 1340470,
      "step": 711,
      "train_runtime": 2133.428,
      "train_tokens_per_second": 628.317
    },
    {
      "epoch": 0.7197371746272428,
      "grad_norm": 1.4660931825637817,
      "learning_rate": 1.2850632911392407e-05,
      "loss": 0.10700499266386032,
      "num_input_tokens_seen": 1342374,
      "step": 712,
      "train_runtime": 2135.6679,
      "train_tokens_per_second": 628.55
    },
    {
      "epoch": 0.7207480414455395,
      "grad_norm": 1.401763916015625,
      "learning_rate": 1.2840506329113925e-05,
      "loss": 0.08156513422727585,
      "num_input_tokens_seen": 1344294,
      "step": 713,
      "train_runtime": 2137.9525,
      "train_tokens_per_second": 628.776
    },
    {
      "epoch": 0.7217589082638363,
      "grad_norm": 1.135650873184204,
      "learning_rate": 1.2830379746835444e-05,
      "loss": 0.08099284768104553,
      "num_input_tokens_seen": 1346268,
      "step": 714,
      "train_runtime": 2140.2283,
      "train_tokens_per_second": 629.03
    },
    {
      "epoch": 0.7227697750821329,
      "grad_norm": 1.2015876770019531,
      "learning_rate": 1.2820253164556965e-05,
      "loss": 0.08257974684238434,
      "num_input_tokens_seen": 1347826,
      "step": 715,
      "train_runtime": 2142.438,
      "train_tokens_per_second": 629.109
    },
    {
      "epoch": 0.7237806419004296,
      "grad_norm": 1.2310470342636108,
      "learning_rate": 1.2810126582278482e-05,
      "loss": 0.08701344579458237,
      "num_input_tokens_seen": 1349612,
      "step": 716,
      "train_runtime": 2144.6719,
      "train_tokens_per_second": 629.286
    },
    {
      "epoch": 0.7247915087187263,
      "grad_norm": 1.2671425342559814,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.10148347914218903,
      "num_input_tokens_seen": 1351630,
      "step": 717,
      "train_runtime": 2146.9435,
      "train_tokens_per_second": 629.56
    },
    {
      "epoch": 0.725802375537023,
      "grad_norm": 1.3022403717041016,
      "learning_rate": 1.2789873417721522e-05,
      "loss": 0.08409610390663147,
      "num_input_tokens_seen": 1353802,
      "step": 718,
      "train_runtime": 2149.2954,
      "train_tokens_per_second": 629.882
    },
    {
      "epoch": 0.7268132423553196,
      "grad_norm": 1.1092630624771118,
      "learning_rate": 1.277974683544304e-05,
      "loss": 0.07556605339050293,
      "num_input_tokens_seen": 1355654,
      "step": 719,
      "train_runtime": 2151.5615,
      "train_tokens_per_second": 630.079
    },
    {
      "epoch": 0.7278241091736164,
      "grad_norm": 1.5835394859313965,
      "learning_rate": 1.2769620253164558e-05,
      "loss": 0.1381828486919403,
      "num_input_tokens_seen": 1357500,
      "step": 720,
      "train_runtime": 2153.8352,
      "train_tokens_per_second": 630.271
    },
    {
      "epoch": 0.728834975991913,
      "grad_norm": 1.4468761682510376,
      "learning_rate": 1.2759493670886076e-05,
      "loss": 0.09593550115823746,
      "num_input_tokens_seen": 1359508,
      "step": 721,
      "train_runtime": 2169.5334,
      "train_tokens_per_second": 626.636
    },
    {
      "epoch": 0.7298458428102097,
      "grad_norm": 1.1962016820907593,
      "learning_rate": 1.2749367088607597e-05,
      "loss": 0.0873030573129654,
      "num_input_tokens_seen": 1361332,
      "step": 722,
      "train_runtime": 2171.7711,
      "train_tokens_per_second": 626.83
    },
    {
      "epoch": 0.7308567096285065,
      "grad_norm": 1.1502801179885864,
      "learning_rate": 1.2739240506329116e-05,
      "loss": 0.0930904671549797,
      "num_input_tokens_seen": 1363258,
      "step": 723,
      "train_runtime": 2174.085,
      "train_tokens_per_second": 627.049
    },
    {
      "epoch": 0.7318675764468031,
      "grad_norm": 0.8665573596954346,
      "learning_rate": 1.2729113924050633e-05,
      "loss": 0.06266863644123077,
      "num_input_tokens_seen": 1365702,
      "step": 724,
      "train_runtime": 2176.4881,
      "train_tokens_per_second": 627.48
    },
    {
      "epoch": 0.7328784432650999,
      "grad_norm": 1.243510127067566,
      "learning_rate": 1.2718987341772154e-05,
      "loss": 0.07042838633060455,
      "num_input_tokens_seen": 1367666,
      "step": 725,
      "train_runtime": 2178.7487,
      "train_tokens_per_second": 627.73
    },
    {
      "epoch": 0.7338893100833965,
      "grad_norm": 1.488826036453247,
      "learning_rate": 1.2708860759493671e-05,
      "loss": 0.0845520943403244,
      "num_input_tokens_seen": 1369268,
      "step": 726,
      "train_runtime": 2180.9313,
      "train_tokens_per_second": 627.836
    },
    {
      "epoch": 0.7349001769016932,
      "grad_norm": 1.3990442752838135,
      "learning_rate": 1.269873417721519e-05,
      "loss": 0.08900083601474762,
      "num_input_tokens_seen": 1370694,
      "step": 727,
      "train_runtime": 2183.1088,
      "train_tokens_per_second": 627.863
    },
    {
      "epoch": 0.7359110437199899,
      "grad_norm": 1.2960129976272583,
      "learning_rate": 1.2688607594936711e-05,
      "loss": 0.10376360267400742,
      "num_input_tokens_seen": 1372434,
      "step": 728,
      "train_runtime": 2185.3092,
      "train_tokens_per_second": 628.027
    },
    {
      "epoch": 0.7369219105382866,
      "grad_norm": 1.9930686950683594,
      "learning_rate": 1.2678481012658228e-05,
      "loss": 0.07666580379009247,
      "num_input_tokens_seen": 1374256,
      "step": 729,
      "train_runtime": 2187.5515,
      "train_tokens_per_second": 628.217
    },
    {
      "epoch": 0.7379327773565832,
      "grad_norm": 1.2373098134994507,
      "learning_rate": 1.2668354430379748e-05,
      "loss": 0.09075476974248886,
      "num_input_tokens_seen": 1376118,
      "step": 730,
      "train_runtime": 2189.8706,
      "train_tokens_per_second": 628.402
    },
    {
      "epoch": 0.73894364417488,
      "grad_norm": 1.373629093170166,
      "learning_rate": 1.2658227848101268e-05,
      "loss": 0.08953611552715302,
      "num_input_tokens_seen": 1377884,
      "step": 731,
      "train_runtime": 2192.128,
      "train_tokens_per_second": 628.56
    },
    {
      "epoch": 0.7399545109931767,
      "grad_norm": 1.2163437604904175,
      "learning_rate": 1.2648101265822786e-05,
      "loss": 0.09008659422397614,
      "num_input_tokens_seen": 1379496,
      "step": 732,
      "train_runtime": 2194.3577,
      "train_tokens_per_second": 628.656
    },
    {
      "epoch": 0.7409653778114733,
      "grad_norm": 1.2478456497192383,
      "learning_rate": 1.2637974683544305e-05,
      "loss": 0.07520309090614319,
      "num_input_tokens_seen": 1381244,
      "step": 733,
      "train_runtime": 2196.6255,
      "train_tokens_per_second": 628.803
    },
    {
      "epoch": 0.7419762446297701,
      "grad_norm": 1.6541633605957031,
      "learning_rate": 1.2627848101265822e-05,
      "loss": 0.09266624599695206,
      "num_input_tokens_seen": 1383306,
      "step": 734,
      "train_runtime": 2198.9875,
      "train_tokens_per_second": 629.065
    },
    {
      "epoch": 0.7429871114480667,
      "grad_norm": 1.1022968292236328,
      "learning_rate": 1.2617721518987343e-05,
      "loss": 0.08509687334299088,
      "num_input_tokens_seen": 1385492,
      "step": 735,
      "train_runtime": 2201.3099,
      "train_tokens_per_second": 629.394
    },
    {
      "epoch": 0.7439979782663634,
      "grad_norm": 1.5268213748931885,
      "learning_rate": 1.2607594936708862e-05,
      "loss": 0.11774347722530365,
      "num_input_tokens_seen": 1387608,
      "step": 736,
      "train_runtime": 2203.6344,
      "train_tokens_per_second": 629.691
    },
    {
      "epoch": 0.7450088450846601,
      "grad_norm": 1.312606930732727,
      "learning_rate": 1.259746835443038e-05,
      "loss": 0.10731050372123718,
      "num_input_tokens_seen": 1389814,
      "step": 737,
      "train_runtime": 2206.0149,
      "train_tokens_per_second": 630.011
    },
    {
      "epoch": 0.7460197119029568,
      "grad_norm": 1.449189305305481,
      "learning_rate": 1.25873417721519e-05,
      "loss": 0.11829297244548798,
      "num_input_tokens_seen": 1391662,
      "step": 738,
      "train_runtime": 2208.2625,
      "train_tokens_per_second": 630.207
    },
    {
      "epoch": 0.7470305787212534,
      "grad_norm": 1.2229044437408447,
      "learning_rate": 1.257721518987342e-05,
      "loss": 0.09924839437007904,
      "num_input_tokens_seen": 1393230,
      "step": 739,
      "train_runtime": 2210.5093,
      "train_tokens_per_second": 630.276
    },
    {
      "epoch": 0.7480414455395502,
      "grad_norm": 0.9675111174583435,
      "learning_rate": 1.2567088607594937e-05,
      "loss": 0.06866496056318283,
      "num_input_tokens_seen": 1395328,
      "step": 740,
      "train_runtime": 2212.9001,
      "train_tokens_per_second": 630.543
    },
    {
      "epoch": 0.7490523123578469,
      "grad_norm": 1.437273621559143,
      "learning_rate": 1.2556962025316457e-05,
      "loss": 0.12035974115133286,
      "num_input_tokens_seen": 1397110,
      "step": 741,
      "train_runtime": 2215.1286,
      "train_tokens_per_second": 630.713
    },
    {
      "epoch": 0.7500631791761435,
      "grad_norm": 1.3278621435165405,
      "learning_rate": 1.2546835443037975e-05,
      "loss": 0.11160625517368317,
      "num_input_tokens_seen": 1399078,
      "step": 742,
      "train_runtime": 2217.4451,
      "train_tokens_per_second": 630.941
    },
    {
      "epoch": 0.7510740459944403,
      "grad_norm": 1.2631326913833618,
      "learning_rate": 1.2536708860759494e-05,
      "loss": 0.072649747133255,
      "num_input_tokens_seen": 1400736,
      "step": 743,
      "train_runtime": 2219.6523,
      "train_tokens_per_second": 631.061
    },
    {
      "epoch": 0.7520849128127369,
      "grad_norm": 1.2321404218673706,
      "learning_rate": 1.2526582278481015e-05,
      "loss": 0.09959045052528381,
      "num_input_tokens_seen": 1403160,
      "step": 744,
      "train_runtime": 2222.062,
      "train_tokens_per_second": 631.468
    },
    {
      "epoch": 0.7530957796310336,
      "grad_norm": 1.221240758895874,
      "learning_rate": 1.2516455696202532e-05,
      "loss": 0.08668084442615509,
      "num_input_tokens_seen": 1405066,
      "step": 745,
      "train_runtime": 2224.2903,
      "train_tokens_per_second": 631.692
    },
    {
      "epoch": 0.7541066464493303,
      "grad_norm": 1.5003905296325684,
      "learning_rate": 1.2506329113924051e-05,
      "loss": 0.11735235154628754,
      "num_input_tokens_seen": 1406536,
      "step": 746,
      "train_runtime": 2226.4237,
      "train_tokens_per_second": 631.747
    },
    {
      "epoch": 0.755117513267627,
      "grad_norm": 1.13472318649292,
      "learning_rate": 1.2496202531645572e-05,
      "loss": 0.07171247154474258,
      "num_input_tokens_seen": 1408360,
      "step": 747,
      "train_runtime": 2228.6579,
      "train_tokens_per_second": 631.932
    },
    {
      "epoch": 0.7561283800859236,
      "grad_norm": 1.063692569732666,
      "learning_rate": 1.248607594936709e-05,
      "loss": 0.07593874633312225,
      "num_input_tokens_seen": 1410444,
      "step": 748,
      "train_runtime": 2230.9177,
      "train_tokens_per_second": 632.226
    },
    {
      "epoch": 0.7571392469042204,
      "grad_norm": 0.9933139681816101,
      "learning_rate": 1.2475949367088608e-05,
      "loss": 0.07491488009691238,
      "num_input_tokens_seen": 1412882,
      "step": 749,
      "train_runtime": 2233.319,
      "train_tokens_per_second": 632.638
    },
    {
      "epoch": 0.7581501137225171,
      "grad_norm": 1.3128093481063843,
      "learning_rate": 1.2465822784810126e-05,
      "loss": 0.0925944596529007,
      "num_input_tokens_seen": 1415136,
      "step": 750,
      "train_runtime": 2235.65,
      "train_tokens_per_second": 632.986
    },
    {
      "epoch": 0.7591609805408137,
      "grad_norm": 1.1090009212493896,
      "learning_rate": 1.2455696202531647e-05,
      "loss": 0.08612965792417526,
      "num_input_tokens_seen": 1416954,
      "step": 751,
      "train_runtime": 2251.2547,
      "train_tokens_per_second": 629.406
    },
    {
      "epoch": 0.7601718473591105,
      "grad_norm": 1.309633493423462,
      "learning_rate": 1.2445569620253166e-05,
      "loss": 0.08517396450042725,
      "num_input_tokens_seen": 1419028,
      "step": 752,
      "train_runtime": 2253.5806,
      "train_tokens_per_second": 629.677
    },
    {
      "epoch": 0.7611827141774071,
      "grad_norm": 1.451980710029602,
      "learning_rate": 1.2435443037974683e-05,
      "loss": 0.12312909960746765,
      "num_input_tokens_seen": 1421050,
      "step": 753,
      "train_runtime": 2255.8839,
      "train_tokens_per_second": 629.93
    },
    {
      "epoch": 0.7621935809957038,
      "grad_norm": 1.2391310930252075,
      "learning_rate": 1.2425316455696204e-05,
      "loss": 0.08031059056520462,
      "num_input_tokens_seen": 1422814,
      "step": 754,
      "train_runtime": 2258.1535,
      "train_tokens_per_second": 630.079
    },
    {
      "epoch": 0.7632044478140005,
      "grad_norm": 1.2742424011230469,
      "learning_rate": 1.2415189873417723e-05,
      "loss": 0.0937035083770752,
      "num_input_tokens_seen": 1424398,
      "step": 755,
      "train_runtime": 2260.3586,
      "train_tokens_per_second": 630.165
    },
    {
      "epoch": 0.7642153146322972,
      "grad_norm": 1.4075407981872559,
      "learning_rate": 1.240506329113924e-05,
      "loss": 0.09099723398685455,
      "num_input_tokens_seen": 1426688,
      "step": 756,
      "train_runtime": 2262.7488,
      "train_tokens_per_second": 630.511
    },
    {
      "epoch": 0.7652261814505938,
      "grad_norm": 1.7033988237380981,
      "learning_rate": 1.2394936708860761e-05,
      "loss": 0.1424482762813568,
      "num_input_tokens_seen": 1428828,
      "step": 757,
      "train_runtime": 2265.0958,
      "train_tokens_per_second": 630.802
    },
    {
      "epoch": 0.7662370482688906,
      "grad_norm": 1.0469303131103516,
      "learning_rate": 1.2384810126582278e-05,
      "loss": 0.07260406017303467,
      "num_input_tokens_seen": 1430676,
      "step": 758,
      "train_runtime": 2267.3597,
      "train_tokens_per_second": 630.988
    },
    {
      "epoch": 0.7672479150871873,
      "grad_norm": 1.3124287128448486,
      "learning_rate": 1.2374683544303798e-05,
      "loss": 0.11914030462503433,
      "num_input_tokens_seen": 1432730,
      "step": 759,
      "train_runtime": 2269.7143,
      "train_tokens_per_second": 631.238
    },
    {
      "epoch": 0.7682587819054839,
      "grad_norm": 1.3265185356140137,
      "learning_rate": 1.2364556962025318e-05,
      "loss": 0.09607601165771484,
      "num_input_tokens_seen": 1434948,
      "step": 760,
      "train_runtime": 2272.1137,
      "train_tokens_per_second": 631.548
    },
    {
      "epoch": 0.7692696487237807,
      "grad_norm": 1.3244144916534424,
      "learning_rate": 1.2354430379746836e-05,
      "loss": 0.09496047347784042,
      "num_input_tokens_seen": 1436938,
      "step": 761,
      "train_runtime": 2274.5501,
      "train_tokens_per_second": 631.746
    },
    {
      "epoch": 0.7702805155420773,
      "grad_norm": 1.0441399812698364,
      "learning_rate": 1.2344303797468355e-05,
      "loss": 0.07020121812820435,
      "num_input_tokens_seen": 1439072,
      "step": 762,
      "train_runtime": 2276.9142,
      "train_tokens_per_second": 632.027
    },
    {
      "epoch": 0.771291382360374,
      "grad_norm": 1.4196510314941406,
      "learning_rate": 1.2334177215189876e-05,
      "loss": 0.08475708961486816,
      "num_input_tokens_seen": 1440696,
      "step": 763,
      "train_runtime": 2279.1738,
      "train_tokens_per_second": 632.113
    },
    {
      "epoch": 0.7723022491786707,
      "grad_norm": 1.190209984779358,
      "learning_rate": 1.2324050632911393e-05,
      "loss": 0.09173276275396347,
      "num_input_tokens_seen": 1442820,
      "step": 764,
      "train_runtime": 2281.4728,
      "train_tokens_per_second": 632.407
    },
    {
      "epoch": 0.7733131159969674,
      "grad_norm": 1.1727373600006104,
      "learning_rate": 1.2313924050632912e-05,
      "loss": 0.089374840259552,
      "num_input_tokens_seen": 1445002,
      "step": 765,
      "train_runtime": 2283.7845,
      "train_tokens_per_second": 632.723
    },
    {
      "epoch": 0.774323982815264,
      "grad_norm": 1.1790472269058228,
      "learning_rate": 1.2303797468354431e-05,
      "loss": 0.08570949733257294,
      "num_input_tokens_seen": 1446852,
      "step": 766,
      "train_runtime": 2286.0611,
      "train_tokens_per_second": 632.902
    },
    {
      "epoch": 0.7753348496335608,
      "grad_norm": 1.1178817749023438,
      "learning_rate": 1.229367088607595e-05,
      "loss": 0.05503210052847862,
      "num_input_tokens_seen": 1448588,
      "step": 767,
      "train_runtime": 2288.2661,
      "train_tokens_per_second": 633.051
    },
    {
      "epoch": 0.7763457164518575,
      "grad_norm": 1.3860853910446167,
      "learning_rate": 1.228354430379747e-05,
      "loss": 0.0975511372089386,
      "num_input_tokens_seen": 1450580,
      "step": 768,
      "train_runtime": 2290.685,
      "train_tokens_per_second": 633.252
    },
    {
      "epoch": 0.7773565832701541,
      "grad_norm": 1.205965518951416,
      "learning_rate": 1.2273417721518988e-05,
      "loss": 0.08186589181423187,
      "num_input_tokens_seen": 1452580,
      "step": 769,
      "train_runtime": 2292.9926,
      "train_tokens_per_second": 633.487
    },
    {
      "epoch": 0.7783674500884509,
      "grad_norm": 1.3132317066192627,
      "learning_rate": 1.2263291139240508e-05,
      "loss": 0.11491493880748749,
      "num_input_tokens_seen": 1454246,
      "step": 770,
      "train_runtime": 2295.2388,
      "train_tokens_per_second": 633.592
    },
    {
      "epoch": 0.7793783169067475,
      "grad_norm": 1.4057326316833496,
      "learning_rate": 1.2253164556962027e-05,
      "loss": 0.09340424835681915,
      "num_input_tokens_seen": 1456172,
      "step": 771,
      "train_runtime": 2297.5182,
      "train_tokens_per_second": 633.802
    },
    {
      "epoch": 0.7803891837250442,
      "grad_norm": 1.348260760307312,
      "learning_rate": 1.2243037974683546e-05,
      "loss": 0.09945379197597504,
      "num_input_tokens_seen": 1458110,
      "step": 772,
      "train_runtime": 2299.8038,
      "train_tokens_per_second": 634.015
    },
    {
      "epoch": 0.7814000505433409,
      "grad_norm": 1.357066035270691,
      "learning_rate": 1.2232911392405065e-05,
      "loss": 0.10925261676311493,
      "num_input_tokens_seen": 1459878,
      "step": 773,
      "train_runtime": 2302.0426,
      "train_tokens_per_second": 634.166
    },
    {
      "epoch": 0.7824109173616376,
      "grad_norm": 1.5371371507644653,
      "learning_rate": 1.2222784810126582e-05,
      "loss": 0.09402672946453094,
      "num_input_tokens_seen": 1461484,
      "step": 774,
      "train_runtime": 2304.2211,
      "train_tokens_per_second": 634.264
    },
    {
      "epoch": 0.7834217841799342,
      "grad_norm": 1.3534631729125977,
      "learning_rate": 1.2212658227848101e-05,
      "loss": 0.09957920014858246,
      "num_input_tokens_seen": 1463336,
      "step": 775,
      "train_runtime": 2306.5036,
      "train_tokens_per_second": 634.439
    },
    {
      "epoch": 0.784432650998231,
      "grad_norm": 1.579530119895935,
      "learning_rate": 1.2202531645569622e-05,
      "loss": 0.07522060722112656,
      "num_input_tokens_seen": 1464984,
      "step": 776,
      "train_runtime": 2308.7105,
      "train_tokens_per_second": 634.546
    },
    {
      "epoch": 0.7854435178165277,
      "grad_norm": 1.7755144834518433,
      "learning_rate": 1.219240506329114e-05,
      "loss": 0.1491493433713913,
      "num_input_tokens_seen": 1466894,
      "step": 777,
      "train_runtime": 2310.9641,
      "train_tokens_per_second": 634.754
    },
    {
      "epoch": 0.7864543846348243,
      "grad_norm": 1.397637963294983,
      "learning_rate": 1.2182278481012658e-05,
      "loss": 0.09773722290992737,
      "num_input_tokens_seen": 1468682,
      "step": 778,
      "train_runtime": 2313.2023,
      "train_tokens_per_second": 634.913
    },
    {
      "epoch": 0.7874652514531211,
      "grad_norm": 1.133230447769165,
      "learning_rate": 1.217215189873418e-05,
      "loss": 0.08112357556819916,
      "num_input_tokens_seen": 1470236,
      "step": 779,
      "train_runtime": 2315.3877,
      "train_tokens_per_second": 634.985
    },
    {
      "epoch": 0.7884761182714177,
      "grad_norm": 1.238479733467102,
      "learning_rate": 1.2162025316455697e-05,
      "loss": 0.07440964132547379,
      "num_input_tokens_seen": 1472040,
      "step": 780,
      "train_runtime": 2317.6707,
      "train_tokens_per_second": 635.138
    },
    {
      "epoch": 0.7894869850897144,
      "grad_norm": 1.3582795858383179,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.08965818583965302,
      "num_input_tokens_seen": 1474404,
      "step": 781,
      "train_runtime": 2333.4251,
      "train_tokens_per_second": 631.863
    },
    {
      "epoch": 0.7904978519080111,
      "grad_norm": 1.657088041305542,
      "learning_rate": 1.2141772151898735e-05,
      "loss": 0.11930066347122192,
      "num_input_tokens_seen": 1476702,
      "step": 782,
      "train_runtime": 2335.8046,
      "train_tokens_per_second": 632.203
    },
    {
      "epoch": 0.7915087187263078,
      "grad_norm": 1.2233569622039795,
      "learning_rate": 1.2131645569620254e-05,
      "loss": 0.09705591201782227,
      "num_input_tokens_seen": 1478698,
      "step": 783,
      "train_runtime": 2338.0877,
      "train_tokens_per_second": 632.439
    },
    {
      "epoch": 0.7925195855446044,
      "grad_norm": 1.1697092056274414,
      "learning_rate": 1.2121518987341773e-05,
      "loss": 0.092095747590065,
      "num_input_tokens_seen": 1480804,
      "step": 784,
      "train_runtime": 2340.3968,
      "train_tokens_per_second": 632.715
    },
    {
      "epoch": 0.7935304523629012,
      "grad_norm": 1.7770445346832275,
      "learning_rate": 1.2111392405063292e-05,
      "loss": 0.13721917569637299,
      "num_input_tokens_seen": 1482470,
      "step": 785,
      "train_runtime": 2342.615,
      "train_tokens_per_second": 632.827
    },
    {
      "epoch": 0.7945413191811979,
      "grad_norm": 1.1368803977966309,
      "learning_rate": 1.2101265822784811e-05,
      "loss": 0.08487553894519806,
      "num_input_tokens_seen": 1484010,
      "step": 786,
      "train_runtime": 2344.7976,
      "train_tokens_per_second": 632.895
    },
    {
      "epoch": 0.7955521859994946,
      "grad_norm": 1.4755949974060059,
      "learning_rate": 1.209113924050633e-05,
      "loss": 0.11078151315450668,
      "num_input_tokens_seen": 1485590,
      "step": 787,
      "train_runtime": 2346.9978,
      "train_tokens_per_second": 632.975
    },
    {
      "epoch": 0.7965630528177913,
      "grad_norm": 1.3972913026809692,
      "learning_rate": 1.208101265822785e-05,
      "loss": 0.10825448483228683,
      "num_input_tokens_seen": 1487662,
      "step": 788,
      "train_runtime": 2349.3566,
      "train_tokens_per_second": 633.221
    },
    {
      "epoch": 0.7975739196360879,
      "grad_norm": 1.3730190992355347,
      "learning_rate": 1.2070886075949368e-05,
      "loss": 0.13756567239761353,
      "num_input_tokens_seen": 1490132,
      "step": 789,
      "train_runtime": 2351.765,
      "train_tokens_per_second": 633.623
    },
    {
      "epoch": 0.7985847864543847,
      "grad_norm": 1.1280244588851929,
      "learning_rate": 1.2060759493670886e-05,
      "loss": 0.08637134730815887,
      "num_input_tokens_seen": 1491952,
      "step": 790,
      "train_runtime": 2354.2732,
      "train_tokens_per_second": 633.721
    },
    {
      "epoch": 0.7995956532726813,
      "grad_norm": 1.4997068643569946,
      "learning_rate": 1.2050632911392407e-05,
      "loss": 0.08552324771881104,
      "num_input_tokens_seen": 1493938,
      "step": 791,
      "train_runtime": 2356.7144,
      "train_tokens_per_second": 633.907
    },
    {
      "epoch": 0.800606520090978,
      "grad_norm": 2.009636640548706,
      "learning_rate": 1.2040506329113926e-05,
      "loss": 0.10743894428014755,
      "num_input_tokens_seen": 1495656,
      "step": 792,
      "train_runtime": 2359.0196,
      "train_tokens_per_second": 634.016
    },
    {
      "epoch": 0.800606520090978,
      "eval_loss": 0.35930678248405457,
      "eval_runtime": 54.3304,
      "eval_samples_per_second": 16.179,
      "eval_steps_per_second": 8.099,
      "num_input_tokens_seen": 1495656,
      "step": 792
    },
    {
      "epoch": 0.8016173869092748,
      "grad_norm": 1.7928344011306763,
      "learning_rate": 1.2030379746835443e-05,
      "loss": 0.11549399048089981,
      "num_input_tokens_seen": 1497184,
      "step": 793,
      "train_runtime": 2415.5783,
      "train_tokens_per_second": 619.804
    },
    {
      "epoch": 0.8026282537275714,
      "grad_norm": 1.2222858667373657,
      "learning_rate": 1.2020253164556964e-05,
      "loss": 0.05913369357585907,
      "num_input_tokens_seen": 1498650,
      "step": 794,
      "train_runtime": 2417.7861,
      "train_tokens_per_second": 619.844
    },
    {
      "epoch": 0.8036391205458681,
      "grad_norm": 1.4649543762207031,
      "learning_rate": 1.2010126582278483e-05,
      "loss": 0.09922467172145844,
      "num_input_tokens_seen": 1500418,
      "step": 795,
      "train_runtime": 2420.03,
      "train_tokens_per_second": 620.0
    },
    {
      "epoch": 0.8046499873641648,
      "grad_norm": 1.232645869255066,
      "learning_rate": 1.2e-05,
      "loss": 0.08798468858003616,
      "num_input_tokens_seen": 1502404,
      "step": 796,
      "train_runtime": 2422.3368,
      "train_tokens_per_second": 620.229
    },
    {
      "epoch": 0.8056608541824615,
      "grad_norm": 1.0265458822250366,
      "learning_rate": 1.1989873417721521e-05,
      "loss": 0.077436164021492,
      "num_input_tokens_seen": 1504506,
      "step": 797,
      "train_runtime": 2424.6461,
      "train_tokens_per_second": 620.505
    },
    {
      "epoch": 0.8066717210007581,
      "grad_norm": 1.3002969026565552,
      "learning_rate": 1.1979746835443038e-05,
      "loss": 0.08807089924812317,
      "num_input_tokens_seen": 1506334,
      "step": 798,
      "train_runtime": 2426.9357,
      "train_tokens_per_second": 620.673
    },
    {
      "epoch": 0.8076825878190549,
      "grad_norm": 1.5173736810684204,
      "learning_rate": 1.1969620253164558e-05,
      "loss": 0.11598929017782211,
      "num_input_tokens_seen": 1508208,
      "step": 799,
      "train_runtime": 2429.2135,
      "train_tokens_per_second": 620.863
    },
    {
      "epoch": 0.8086934546373515,
      "grad_norm": 1.0948872566223145,
      "learning_rate": 1.1959493670886078e-05,
      "loss": 0.07768983393907547,
      "num_input_tokens_seen": 1509974,
      "step": 800,
      "train_runtime": 2431.474,
      "train_tokens_per_second": 621.012
    },
    {
      "epoch": 0.8097043214556482,
      "grad_norm": 1.5887296199798584,
      "learning_rate": 1.1949367088607596e-05,
      "loss": 0.13126257061958313,
      "num_input_tokens_seen": 1511792,
      "step": 801,
      "train_runtime": 2433.8018,
      "train_tokens_per_second": 621.165
    },
    {
      "epoch": 0.810715188273945,
      "grad_norm": 1.3212714195251465,
      "learning_rate": 1.1939240506329115e-05,
      "loss": 0.10201811790466309,
      "num_input_tokens_seen": 1513666,
      "step": 802,
      "train_runtime": 2436.0742,
      "train_tokens_per_second": 621.355
    },
    {
      "epoch": 0.8117260550922416,
      "grad_norm": 1.2121456861495972,
      "learning_rate": 1.1929113924050636e-05,
      "loss": 0.09122280031442642,
      "num_input_tokens_seen": 1516078,
      "step": 803,
      "train_runtime": 2438.4704,
      "train_tokens_per_second": 621.733
    },
    {
      "epoch": 0.8127369219105383,
      "grad_norm": 1.2374351024627686,
      "learning_rate": 1.1918987341772153e-05,
      "loss": 0.07882948219776154,
      "num_input_tokens_seen": 1517678,
      "step": 804,
      "train_runtime": 2440.6457,
      "train_tokens_per_second": 621.835
    },
    {
      "epoch": 0.813747788728835,
      "grad_norm": 1.380174994468689,
      "learning_rate": 1.1908860759493672e-05,
      "loss": 0.09974582493305206,
      "num_input_tokens_seen": 1519652,
      "step": 805,
      "train_runtime": 2442.9211,
      "train_tokens_per_second": 622.063
    },
    {
      "epoch": 0.8147586555471317,
      "grad_norm": 1.328392744064331,
      "learning_rate": 1.189873417721519e-05,
      "loss": 0.09549365192651749,
      "num_input_tokens_seen": 1521208,
      "step": 806,
      "train_runtime": 2445.1007,
      "train_tokens_per_second": 622.145
    },
    {
      "epoch": 0.8157695223654283,
      "grad_norm": 1.6172430515289307,
      "learning_rate": 1.188860759493671e-05,
      "loss": 0.06688052415847778,
      "num_input_tokens_seen": 1523202,
      "step": 807,
      "train_runtime": 2447.3708,
      "train_tokens_per_second": 622.383
    },
    {
      "epoch": 0.8167803891837251,
      "grad_norm": 1.511674404144287,
      "learning_rate": 1.187848101265823e-05,
      "loss": 0.11953920125961304,
      "num_input_tokens_seen": 1524950,
      "step": 808,
      "train_runtime": 2449.6067,
      "train_tokens_per_second": 622.528
    },
    {
      "epoch": 0.8177912560020217,
      "grad_norm": 1.4442185163497925,
      "learning_rate": 1.1868354430379747e-05,
      "loss": 0.08532628417015076,
      "num_input_tokens_seen": 1526390,
      "step": 809,
      "train_runtime": 2451.7873,
      "train_tokens_per_second": 622.562
    },
    {
      "epoch": 0.8188021228203184,
      "grad_norm": 1.389597773551941,
      "learning_rate": 1.1858227848101267e-05,
      "loss": 0.09901273995637894,
      "num_input_tokens_seen": 1528232,
      "step": 810,
      "train_runtime": 2454.0735,
      "train_tokens_per_second": 622.733
    },
    {
      "epoch": 0.8198129896386152,
      "grad_norm": 1.3464856147766113,
      "learning_rate": 1.1848101265822787e-05,
      "loss": 0.10714782029390335,
      "num_input_tokens_seen": 1530056,
      "step": 811,
      "train_runtime": 2469.7011,
      "train_tokens_per_second": 619.531
    },
    {
      "epoch": 0.8208238564569118,
      "grad_norm": 1.619875431060791,
      "learning_rate": 1.1837974683544304e-05,
      "loss": 0.12249326705932617,
      "num_input_tokens_seen": 1532250,
      "step": 812,
      "train_runtime": 2472.0727,
      "train_tokens_per_second": 619.824
    },
    {
      "epoch": 0.8218347232752085,
      "grad_norm": 1.0970927476882935,
      "learning_rate": 1.1827848101265825e-05,
      "loss": 0.07444127649068832,
      "num_input_tokens_seen": 1534102,
      "step": 813,
      "train_runtime": 2474.3566,
      "train_tokens_per_second": 620.0
    },
    {
      "epoch": 0.8228455900935052,
      "grad_norm": 1.2683417797088623,
      "learning_rate": 1.1817721518987342e-05,
      "loss": 0.0832136869430542,
      "num_input_tokens_seen": 1535648,
      "step": 814,
      "train_runtime": 2476.5509,
      "train_tokens_per_second": 620.075
    },
    {
      "epoch": 0.8238564569118019,
      "grad_norm": 1.217896819114685,
      "learning_rate": 1.1807594936708861e-05,
      "loss": 0.086221843957901,
      "num_input_tokens_seen": 1537408,
      "step": 815,
      "train_runtime": 2478.8425,
      "train_tokens_per_second": 620.212
    },
    {
      "epoch": 0.8248673237300985,
      "grad_norm": 1.1209136247634888,
      "learning_rate": 1.1797468354430382e-05,
      "loss": 0.05607465282082558,
      "num_input_tokens_seen": 1539398,
      "step": 816,
      "train_runtime": 2481.1293,
      "train_tokens_per_second": 620.442
    },
    {
      "epoch": 0.8258781905483953,
      "grad_norm": 1.3270211219787598,
      "learning_rate": 1.17873417721519e-05,
      "loss": 0.09606637060642242,
      "num_input_tokens_seen": 1541054,
      "step": 817,
      "train_runtime": 2483.3588,
      "train_tokens_per_second": 620.552
    },
    {
      "epoch": 0.8268890573666919,
      "grad_norm": 1.0250391960144043,
      "learning_rate": 1.1777215189873418e-05,
      "loss": 0.08006328344345093,
      "num_input_tokens_seen": 1543138,
      "step": 818,
      "train_runtime": 2485.6833,
      "train_tokens_per_second": 620.81
    },
    {
      "epoch": 0.8278999241849886,
      "grad_norm": 1.1161867380142212,
      "learning_rate": 1.176708860759494e-05,
      "loss": 0.0717461109161377,
      "num_input_tokens_seen": 1544674,
      "step": 819,
      "train_runtime": 2487.9235,
      "train_tokens_per_second": 620.869
    },
    {
      "epoch": 0.8289107910032854,
      "grad_norm": 1.4846850633621216,
      "learning_rate": 1.1756962025316457e-05,
      "loss": 0.13073456287384033,
      "num_input_tokens_seen": 1546470,
      "step": 820,
      "train_runtime": 2490.1578,
      "train_tokens_per_second": 621.033
    },
    {
      "epoch": 0.829921657821582,
      "grad_norm": 1.3668291568756104,
      "learning_rate": 1.1746835443037976e-05,
      "loss": 0.10845942050218582,
      "num_input_tokens_seen": 1548530,
      "step": 821,
      "train_runtime": 2492.6282,
      "train_tokens_per_second": 621.244
    },
    {
      "epoch": 0.8309325246398787,
      "grad_norm": 1.655059814453125,
      "learning_rate": 1.1736708860759493e-05,
      "loss": 0.09784270823001862,
      "num_input_tokens_seen": 1549984,
      "step": 822,
      "train_runtime": 2495.0317,
      "train_tokens_per_second": 621.228
    },
    {
      "epoch": 0.8319433914581754,
      "grad_norm": 1.283531665802002,
      "learning_rate": 1.1726582278481014e-05,
      "loss": 0.0716037005186081,
      "num_input_tokens_seen": 1552118,
      "step": 823,
      "train_runtime": 2497.4111,
      "train_tokens_per_second": 621.491
    },
    {
      "epoch": 0.8329542582764721,
      "grad_norm": 1.3344287872314453,
      "learning_rate": 1.1716455696202533e-05,
      "loss": 0.10793325304985046,
      "num_input_tokens_seen": 1554850,
      "step": 824,
      "train_runtime": 2499.9058,
      "train_tokens_per_second": 621.963
    },
    {
      "epoch": 0.8339651250947687,
      "grad_norm": 1.2568620443344116,
      "learning_rate": 1.170632911392405e-05,
      "loss": 0.09416936337947845,
      "num_input_tokens_seen": 1556788,
      "step": 825,
      "train_runtime": 2502.1786,
      "train_tokens_per_second": 622.173
    },
    {
      "epoch": 0.8349759919130655,
      "grad_norm": 1.3056141138076782,
      "learning_rate": 1.1696202531645571e-05,
      "loss": 0.11997391283512115,
      "num_input_tokens_seen": 1558446,
      "step": 826,
      "train_runtime": 2504.4166,
      "train_tokens_per_second": 622.279
    },
    {
      "epoch": 0.8359868587313621,
      "grad_norm": 1.4461060762405396,
      "learning_rate": 1.168607594936709e-05,
      "loss": 0.11721283197402954,
      "num_input_tokens_seen": 1560354,
      "step": 827,
      "train_runtime": 2506.6819,
      "train_tokens_per_second": 622.478
    },
    {
      "epoch": 0.8369977255496588,
      "grad_norm": 1.654499888420105,
      "learning_rate": 1.1675949367088608e-05,
      "loss": 0.14523056149482727,
      "num_input_tokens_seen": 1562322,
      "step": 828,
      "train_runtime": 2508.9837,
      "train_tokens_per_second": 622.691
    },
    {
      "epoch": 0.8380085923679556,
      "grad_norm": 1.6324923038482666,
      "learning_rate": 1.1665822784810128e-05,
      "loss": 0.09989740699529648,
      "num_input_tokens_seen": 1564184,
      "step": 829,
      "train_runtime": 2511.2273,
      "train_tokens_per_second": 622.876
    },
    {
      "epoch": 0.8390194591862522,
      "grad_norm": 1.7122888565063477,
      "learning_rate": 1.1655696202531646e-05,
      "loss": 0.11083639413118362,
      "num_input_tokens_seen": 1565994,
      "step": 830,
      "train_runtime": 2513.5596,
      "train_tokens_per_second": 623.018
    },
    {
      "epoch": 0.8400303260045489,
      "grad_norm": 1.2459877729415894,
      "learning_rate": 1.1645569620253165e-05,
      "loss": 0.09095220267772675,
      "num_input_tokens_seen": 1567916,
      "step": 831,
      "train_runtime": 2515.8203,
      "train_tokens_per_second": 623.223
    },
    {
      "epoch": 0.8410411928228456,
      "grad_norm": 1.263240933418274,
      "learning_rate": 1.1635443037974686e-05,
      "loss": 0.10126728564500809,
      "num_input_tokens_seen": 1569890,
      "step": 832,
      "train_runtime": 2518.1751,
      "train_tokens_per_second": 623.424
    },
    {
      "epoch": 0.8420520596411423,
      "grad_norm": 1.4330466985702515,
      "learning_rate": 1.1625316455696203e-05,
      "loss": 0.11584387719631195,
      "num_input_tokens_seen": 1572012,
      "step": 833,
      "train_runtime": 2520.4854,
      "train_tokens_per_second": 623.694
    },
    {
      "epoch": 0.8430629264594389,
      "grad_norm": 1.1377328634262085,
      "learning_rate": 1.1615189873417722e-05,
      "loss": 0.06077614426612854,
      "num_input_tokens_seen": 1573832,
      "step": 834,
      "train_runtime": 2522.7251,
      "train_tokens_per_second": 623.862
    },
    {
      "epoch": 0.8440737932777357,
      "grad_norm": 1.0365792512893677,
      "learning_rate": 1.1605063291139243e-05,
      "loss": 0.0817495584487915,
      "num_input_tokens_seen": 1575922,
      "step": 835,
      "train_runtime": 2525.047,
      "train_tokens_per_second": 624.116
    },
    {
      "epoch": 0.8450846600960323,
      "grad_norm": 1.4109801054000854,
      "learning_rate": 1.159493670886076e-05,
      "loss": 0.11633533239364624,
      "num_input_tokens_seen": 1577630,
      "step": 836,
      "train_runtime": 2527.2722,
      "train_tokens_per_second": 624.242
    },
    {
      "epoch": 0.846095526914329,
      "grad_norm": 1.3445757627487183,
      "learning_rate": 1.158481012658228e-05,
      "loss": 0.08244745433330536,
      "num_input_tokens_seen": 1579326,
      "step": 837,
      "train_runtime": 2529.4949,
      "train_tokens_per_second": 624.364
    },
    {
      "epoch": 0.8471063937326258,
      "grad_norm": 1.3189071416854858,
      "learning_rate": 1.1574683544303797e-05,
      "loss": 0.09386275708675385,
      "num_input_tokens_seen": 1581406,
      "step": 838,
      "train_runtime": 2531.792,
      "train_tokens_per_second": 624.619
    },
    {
      "epoch": 0.8481172605509224,
      "grad_norm": 1.2236045598983765,
      "learning_rate": 1.1564556962025318e-05,
      "loss": 0.09385530650615692,
      "num_input_tokens_seen": 1583126,
      "step": 839,
      "train_runtime": 2534.0216,
      "train_tokens_per_second": 624.748
    },
    {
      "epoch": 0.849128127369219,
      "grad_norm": 1.2396245002746582,
      "learning_rate": 1.1554430379746837e-05,
      "loss": 0.105586938560009,
      "num_input_tokens_seen": 1585318,
      "step": 840,
      "train_runtime": 2536.3768,
      "train_tokens_per_second": 625.033
    },
    {
      "epoch": 0.8501389941875158,
      "grad_norm": 1.0138757228851318,
      "learning_rate": 1.1544303797468354e-05,
      "loss": 0.060473039746284485,
      "num_input_tokens_seen": 1586846,
      "step": 841,
      "train_runtime": 2551.8492,
      "train_tokens_per_second": 621.842
    },
    {
      "epoch": 0.8511498610058125,
      "grad_norm": 1.7759934663772583,
      "learning_rate": 1.1534177215189875e-05,
      "loss": 0.08443131297826767,
      "num_input_tokens_seen": 1588624,
      "step": 842,
      "train_runtime": 2554.0486,
      "train_tokens_per_second": 622.002
    },
    {
      "epoch": 0.8521607278241091,
      "grad_norm": 1.3102169036865234,
      "learning_rate": 1.1524050632911394e-05,
      "loss": 0.09713117778301239,
      "num_input_tokens_seen": 1590574,
      "step": 843,
      "train_runtime": 2556.3518,
      "train_tokens_per_second": 622.205
    },
    {
      "epoch": 0.8531715946424059,
      "grad_norm": 1.360823154449463,
      "learning_rate": 1.1513924050632911e-05,
      "loss": 0.09284176677465439,
      "num_input_tokens_seen": 1592676,
      "step": 844,
      "train_runtime": 2558.6917,
      "train_tokens_per_second": 622.457
    },
    {
      "epoch": 0.8541824614607025,
      "grad_norm": 1.1800224781036377,
      "learning_rate": 1.1503797468354432e-05,
      "loss": 0.09340135753154755,
      "num_input_tokens_seen": 1594634,
      "step": 845,
      "train_runtime": 2560.9627,
      "train_tokens_per_second": 622.67
    },
    {
      "epoch": 0.8551933282789992,
      "grad_norm": 1.7343326807022095,
      "learning_rate": 1.149367088607595e-05,
      "loss": 0.0815059244632721,
      "num_input_tokens_seen": 1596328,
      "step": 846,
      "train_runtime": 2563.2004,
      "train_tokens_per_second": 622.787
    },
    {
      "epoch": 0.856204195097296,
      "grad_norm": 1.2050825357437134,
      "learning_rate": 1.1483544303797469e-05,
      "loss": 0.11697637289762497,
      "num_input_tokens_seen": 1598918,
      "step": 847,
      "train_runtime": 2565.6376,
      "train_tokens_per_second": 623.205
    },
    {
      "epoch": 0.8572150619155926,
      "grad_norm": 1.4726566076278687,
      "learning_rate": 1.147341772151899e-05,
      "loss": 0.13028667867183685,
      "num_input_tokens_seen": 1600810,
      "step": 848,
      "train_runtime": 2567.9316,
      "train_tokens_per_second": 623.385
    },
    {
      "epoch": 0.8582259287338894,
      "grad_norm": 1.2934855222702026,
      "learning_rate": 1.1463291139240507e-05,
      "loss": 0.10013256967067719,
      "num_input_tokens_seen": 1602868,
      "step": 849,
      "train_runtime": 2570.2184,
      "train_tokens_per_second": 623.631
    },
    {
      "epoch": 0.859236795552186,
      "grad_norm": 1.3081752061843872,
      "learning_rate": 1.1453164556962026e-05,
      "loss": 0.08440461754798889,
      "num_input_tokens_seen": 1604334,
      "step": 850,
      "train_runtime": 2572.4376,
      "train_tokens_per_second": 623.663
    },
    {
      "epoch": 0.8602476623704827,
      "grad_norm": 1.536041498184204,
      "learning_rate": 1.1443037974683547e-05,
      "loss": 0.12332676351070404,
      "num_input_tokens_seen": 1606456,
      "step": 851,
      "train_runtime": 2574.9985,
      "train_tokens_per_second": 623.867
    },
    {
      "epoch": 0.8612585291887794,
      "grad_norm": 1.4961923360824585,
      "learning_rate": 1.1432911392405064e-05,
      "loss": 0.08930612355470657,
      "num_input_tokens_seen": 1608280,
      "step": 852,
      "train_runtime": 2577.2845,
      "train_tokens_per_second": 624.021
    },
    {
      "epoch": 0.8622693960070761,
      "grad_norm": 1.233816146850586,
      "learning_rate": 1.1422784810126583e-05,
      "loss": 0.07946805655956268,
      "num_input_tokens_seen": 1609990,
      "step": 853,
      "train_runtime": 2579.534,
      "train_tokens_per_second": 624.14
    },
    {
      "epoch": 0.8632802628253727,
      "grad_norm": 1.2792223691940308,
      "learning_rate": 1.1412658227848102e-05,
      "loss": 0.1130804792046547,
      "num_input_tokens_seen": 1611726,
      "step": 854,
      "train_runtime": 2581.7812,
      "train_tokens_per_second": 624.269
    },
    {
      "epoch": 0.8642911296436695,
      "grad_norm": 1.2561981678009033,
      "learning_rate": 1.1402531645569621e-05,
      "loss": 0.10342805087566376,
      "num_input_tokens_seen": 1613962,
      "step": 855,
      "train_runtime": 2584.1287,
      "train_tokens_per_second": 624.567
    },
    {
      "epoch": 0.8653019964619661,
      "grad_norm": 1.2710999250411987,
      "learning_rate": 1.139240506329114e-05,
      "loss": 0.08188050240278244,
      "num_input_tokens_seen": 1615632,
      "step": 856,
      "train_runtime": 2586.3626,
      "train_tokens_per_second": 624.673
    },
    {
      "epoch": 0.8663128632802628,
      "grad_norm": 1.585524320602417,
      "learning_rate": 1.138227848101266e-05,
      "loss": 0.08345942199230194,
      "num_input_tokens_seen": 1617550,
      "step": 857,
      "train_runtime": 2588.6582,
      "train_tokens_per_second": 624.86
    },
    {
      "epoch": 0.8673237300985596,
      "grad_norm": 1.2750364542007446,
      "learning_rate": 1.1372151898734178e-05,
      "loss": 0.09358559548854828,
      "num_input_tokens_seen": 1619282,
      "step": 858,
      "train_runtime": 2590.908,
      "train_tokens_per_second": 624.986
    },
    {
      "epoch": 0.8683345969168562,
      "grad_norm": 1.430489420890808,
      "learning_rate": 1.1362025316455698e-05,
      "loss": 0.10552196949720383,
      "num_input_tokens_seen": 1620816,
      "step": 859,
      "train_runtime": 2593.121,
      "train_tokens_per_second": 625.044
    },
    {
      "epoch": 0.8693454637351529,
      "grad_norm": 1.147262454032898,
      "learning_rate": 1.1351898734177217e-05,
      "loss": 0.09242111444473267,
      "num_input_tokens_seen": 1622940,
      "step": 860,
      "train_runtime": 2595.4643,
      "train_tokens_per_second": 625.299
    },
    {
      "epoch": 0.8703563305534496,
      "grad_norm": 1.1435247659683228,
      "learning_rate": 1.1341772151898736e-05,
      "loss": 0.08492071181535721,
      "num_input_tokens_seen": 1624928,
      "step": 861,
      "train_runtime": 2597.7423,
      "train_tokens_per_second": 625.515
    },
    {
      "epoch": 0.8713671973717463,
      "grad_norm": 1.2436879873275757,
      "learning_rate": 1.1331645569620253e-05,
      "loss": 0.10726184397935867,
      "num_input_tokens_seen": 1626842,
      "step": 862,
      "train_runtime": 2599.9931,
      "train_tokens_per_second": 625.71
    },
    {
      "epoch": 0.8723780641900429,
      "grad_norm": 1.2487578392028809,
      "learning_rate": 1.1321518987341772e-05,
      "loss": 0.09124457091093063,
      "num_input_tokens_seen": 1628466,
      "step": 863,
      "train_runtime": 2602.204,
      "train_tokens_per_second": 625.803
    },
    {
      "epoch": 0.8733889310083397,
      "grad_norm": 1.1977343559265137,
      "learning_rate": 1.1311392405063293e-05,
      "loss": 0.08865179866552353,
      "num_input_tokens_seen": 1630404,
      "step": 864,
      "train_runtime": 2604.4756,
      "train_tokens_per_second": 626.001
    },
    {
      "epoch": 0.8743997978266363,
      "grad_norm": 1.2378313541412354,
      "learning_rate": 1.130126582278481e-05,
      "loss": 0.10676542669534683,
      "num_input_tokens_seen": 1632534,
      "step": 865,
      "train_runtime": 2606.8878,
      "train_tokens_per_second": 626.239
    },
    {
      "epoch": 0.875410664644933,
      "grad_norm": 1.7088805437088013,
      "learning_rate": 1.129113924050633e-05,
      "loss": 0.08600839227437973,
      "num_input_tokens_seen": 1634250,
      "step": 866,
      "train_runtime": 2609.105,
      "train_tokens_per_second": 626.364
    },
    {
      "epoch": 0.8764215314632298,
      "grad_norm": 1.0192761421203613,
      "learning_rate": 1.128101265822785e-05,
      "loss": 0.07756936550140381,
      "num_input_tokens_seen": 1636256,
      "step": 867,
      "train_runtime": 2611.4178,
      "train_tokens_per_second": 626.578
    },
    {
      "epoch": 0.8774323982815264,
      "grad_norm": 1.2040374279022217,
      "learning_rate": 1.1270886075949368e-05,
      "loss": 0.10210748016834259,
      "num_input_tokens_seen": 1638308,
      "step": 868,
      "train_runtime": 2613.7515,
      "train_tokens_per_second": 626.803
    },
    {
      "epoch": 0.8784432650998231,
      "grad_norm": 1.1048957109451294,
      "learning_rate": 1.1260759493670887e-05,
      "loss": 0.05952364578843117,
      "num_input_tokens_seen": 1639966,
      "step": 869,
      "train_runtime": 2615.999,
      "train_tokens_per_second": 626.899
    },
    {
      "epoch": 0.8794541319181198,
      "grad_norm": 1.111210823059082,
      "learning_rate": 1.1250632911392406e-05,
      "loss": 0.09272414445877075,
      "num_input_tokens_seen": 1642326,
      "step": 870,
      "train_runtime": 2618.4432,
      "train_tokens_per_second": 627.215
    },
    {
      "epoch": 0.8804649987364165,
      "grad_norm": 1.1282939910888672,
      "learning_rate": 1.1240506329113925e-05,
      "loss": 0.0725017562508583,
      "num_input_tokens_seen": 1643814,
      "step": 871,
      "train_runtime": 2633.9133,
      "train_tokens_per_second": 624.096
    },
    {
      "epoch": 0.8814758655547131,
      "grad_norm": 1.5181385278701782,
      "learning_rate": 1.1230379746835444e-05,
      "loss": 0.07863800972700119,
      "num_input_tokens_seen": 1645608,
      "step": 872,
      "train_runtime": 2636.2032,
      "train_tokens_per_second": 624.234
    },
    {
      "epoch": 0.8824867323730099,
      "grad_norm": 1.4236536026000977,
      "learning_rate": 1.1220253164556963e-05,
      "loss": 0.12383362650871277,
      "num_input_tokens_seen": 1647372,
      "step": 873,
      "train_runtime": 2638.4522,
      "train_tokens_per_second": 624.371
    },
    {
      "epoch": 0.8834975991913065,
      "grad_norm": 1.9042201042175293,
      "learning_rate": 1.1210126582278482e-05,
      "loss": 0.13041570782661438,
      "num_input_tokens_seen": 1649480,
      "step": 874,
      "train_runtime": 2640.7905,
      "train_tokens_per_second": 624.616
    },
    {
      "epoch": 0.8845084660096032,
      "grad_norm": 1.4499646425247192,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 0.1207432895898819,
      "num_input_tokens_seen": 1651222,
      "step": 875,
      "train_runtime": 2643.0695,
      "train_tokens_per_second": 624.737
    },
    {
      "epoch": 0.8855193328279,
      "grad_norm": 1.054856777191162,
      "learning_rate": 1.118987341772152e-05,
      "loss": 0.07857292890548706,
      "num_input_tokens_seen": 1653380,
      "step": 876,
      "train_runtime": 2645.5031,
      "train_tokens_per_second": 624.978
    },
    {
      "epoch": 0.8865301996461966,
      "grad_norm": 1.4799036979675293,
      "learning_rate": 1.117974683544304e-05,
      "loss": 0.11102406680583954,
      "num_input_tokens_seen": 1654968,
      "step": 877,
      "train_runtime": 2647.6718,
      "train_tokens_per_second": 625.065
    },
    {
      "epoch": 0.8875410664644933,
      "grad_norm": 1.5008419752120972,
      "learning_rate": 1.1169620253164557e-05,
      "loss": 0.09687478840351105,
      "num_input_tokens_seen": 1656582,
      "step": 878,
      "train_runtime": 2649.8987,
      "train_tokens_per_second": 625.149
    },
    {
      "epoch": 0.88855193328279,
      "grad_norm": 1.3605046272277832,
      "learning_rate": 1.1159493670886078e-05,
      "loss": 0.09195833653211594,
      "num_input_tokens_seen": 1658438,
      "step": 879,
      "train_runtime": 2652.1339,
      "train_tokens_per_second": 625.322
    },
    {
      "epoch": 0.8895628001010867,
      "grad_norm": 1.4020121097564697,
      "learning_rate": 1.1149367088607597e-05,
      "loss": 0.11234365403652191,
      "num_input_tokens_seen": 1660068,
      "step": 880,
      "train_runtime": 2654.4921,
      "train_tokens_per_second": 625.381
    },
    {
      "epoch": 0.8905736669193833,
      "grad_norm": 1.208852767944336,
      "learning_rate": 1.1139240506329114e-05,
      "loss": 0.08281447738409042,
      "num_input_tokens_seen": 1661494,
      "step": 881,
      "train_runtime": 2656.8487,
      "train_tokens_per_second": 625.363
    },
    {
      "epoch": 0.8915845337376801,
      "grad_norm": 1.037976622581482,
      "learning_rate": 1.1129113924050635e-05,
      "loss": 0.05970768630504608,
      "num_input_tokens_seen": 1663150,
      "step": 882,
      "train_runtime": 2659.055,
      "train_tokens_per_second": 625.467
    },
    {
      "epoch": 0.8925954005559767,
      "grad_norm": 1.2863762378692627,
      "learning_rate": 1.1118987341772154e-05,
      "loss": 0.07820567488670349,
      "num_input_tokens_seen": 1664702,
      "step": 883,
      "train_runtime": 2661.2329,
      "train_tokens_per_second": 625.538
    },
    {
      "epoch": 0.8936062673742734,
      "grad_norm": 1.1576390266418457,
      "learning_rate": 1.1108860759493671e-05,
      "loss": 0.09026230871677399,
      "num_input_tokens_seen": 1667500,
      "step": 884,
      "train_runtime": 2663.722,
      "train_tokens_per_second": 626.004
    },
    {
      "epoch": 0.8946171341925702,
      "grad_norm": 1.0148956775665283,
      "learning_rate": 1.1098734177215192e-05,
      "loss": 0.07751452177762985,
      "num_input_tokens_seen": 1669594,
      "step": 885,
      "train_runtime": 2666.0191,
      "train_tokens_per_second": 626.25
    },
    {
      "epoch": 0.8956280010108668,
      "grad_norm": 1.491645336151123,
      "learning_rate": 1.108860759493671e-05,
      "loss": 0.09333303570747375,
      "num_input_tokens_seen": 1671040,
      "step": 886,
      "train_runtime": 2668.1599,
      "train_tokens_per_second": 626.289
    },
    {
      "epoch": 0.8966388678291635,
      "grad_norm": 1.2536612749099731,
      "learning_rate": 1.1078481012658229e-05,
      "loss": 0.09178593009710312,
      "num_input_tokens_seen": 1672500,
      "step": 887,
      "train_runtime": 2670.3255,
      "train_tokens_per_second": 626.328
    },
    {
      "epoch": 0.8976497346474602,
      "grad_norm": 1.4343655109405518,
      "learning_rate": 1.106835443037975e-05,
      "loss": 0.07966658473014832,
      "num_input_tokens_seen": 1674008,
      "step": 888,
      "train_runtime": 2672.4715,
      "train_tokens_per_second": 626.389
    },
    {
      "epoch": 0.8986606014657569,
      "grad_norm": 1.1468838453292847,
      "learning_rate": 1.1058227848101267e-05,
      "loss": 0.06319437175989151,
      "num_input_tokens_seen": 1676148,
      "step": 889,
      "train_runtime": 2674.8308,
      "train_tokens_per_second": 626.637
    },
    {
      "epoch": 0.8996714682840535,
      "grad_norm": 1.5938904285430908,
      "learning_rate": 1.1048101265822786e-05,
      "loss": 0.13954637944698334,
      "num_input_tokens_seen": 1677864,
      "step": 890,
      "train_runtime": 2677.0491,
      "train_tokens_per_second": 626.759
    },
    {
      "epoch": 0.9006823351023503,
      "grad_norm": 1.0170766115188599,
      "learning_rate": 1.1037974683544307e-05,
      "loss": 0.06047394126653671,
      "num_input_tokens_seen": 1679694,
      "step": 891,
      "train_runtime": 2679.3068,
      "train_tokens_per_second": 626.914
    },
    {
      "epoch": 0.901693201920647,
      "grad_norm": 1.4534028768539429,
      "learning_rate": 1.1027848101265824e-05,
      "loss": 0.11416472494602203,
      "num_input_tokens_seen": 1681590,
      "step": 892,
      "train_runtime": 2681.5788,
      "train_tokens_per_second": 627.09
    },
    {
      "epoch": 0.9027040687389436,
      "grad_norm": 0.9670836925506592,
      "learning_rate": 1.1017721518987343e-05,
      "loss": 0.05867832154035568,
      "num_input_tokens_seen": 1683350,
      "step": 893,
      "train_runtime": 2683.7967,
      "train_tokens_per_second": 627.227
    },
    {
      "epoch": 0.9037149355572404,
      "grad_norm": 1.0167635679244995,
      "learning_rate": 1.100759493670886e-05,
      "loss": 0.06254291534423828,
      "num_input_tokens_seen": 1685174,
      "step": 894,
      "train_runtime": 2686.0335,
      "train_tokens_per_second": 627.384
    },
    {
      "epoch": 0.904725802375537,
      "grad_norm": 1.2036210298538208,
      "learning_rate": 1.0997468354430381e-05,
      "loss": 0.09041282534599304,
      "num_input_tokens_seen": 1687326,
      "step": 895,
      "train_runtime": 2688.328,
      "train_tokens_per_second": 627.649
    },
    {
      "epoch": 0.9057366691938337,
      "grad_norm": 1.472809910774231,
      "learning_rate": 1.09873417721519e-05,
      "loss": 0.08746948838233948,
      "num_input_tokens_seen": 1689160,
      "step": 896,
      "train_runtime": 2690.6012,
      "train_tokens_per_second": 627.8
    },
    {
      "epoch": 0.9067475360121304,
      "grad_norm": 1.5001329183578491,
      "learning_rate": 1.0977215189873418e-05,
      "loss": 0.07501882314682007,
      "num_input_tokens_seen": 1690894,
      "step": 897,
      "train_runtime": 2692.8316,
      "train_tokens_per_second": 627.924
    },
    {
      "epoch": 0.9077584028304271,
      "grad_norm": 1.3893307447433472,
      "learning_rate": 1.0967088607594938e-05,
      "loss": 0.0938003659248352,
      "num_input_tokens_seen": 1692400,
      "step": 898,
      "train_runtime": 2695.078,
      "train_tokens_per_second": 627.96
    },
    {
      "epoch": 0.9087692696487237,
      "grad_norm": 1.2841914892196655,
      "learning_rate": 1.0956962025316458e-05,
      "loss": 0.09565150737762451,
      "num_input_tokens_seen": 1694112,
      "step": 899,
      "train_runtime": 2697.2872,
      "train_tokens_per_second": 628.08
    },
    {
      "epoch": 0.9097801364670205,
      "grad_norm": 1.5948413610458374,
      "learning_rate": 1.0946835443037975e-05,
      "loss": 0.12826338410377502,
      "num_input_tokens_seen": 1695600,
      "step": 900,
      "train_runtime": 2699.4302,
      "train_tokens_per_second": 628.133
    },
    {
      "epoch": 0.9107910032853171,
      "grad_norm": 1.1783521175384521,
      "learning_rate": 1.0936708860759496e-05,
      "loss": 0.09433984011411667,
      "num_input_tokens_seen": 1697734,
      "step": 901,
      "train_runtime": 2713.9562,
      "train_tokens_per_second": 625.557
    },
    {
      "epoch": 0.9118018701036138,
      "grad_norm": 1.1530615091323853,
      "learning_rate": 1.0926582278481013e-05,
      "loss": 0.08439808338880539,
      "num_input_tokens_seen": 1699566,
      "step": 902,
      "train_runtime": 2716.2257,
      "train_tokens_per_second": 625.709
    },
    {
      "epoch": 0.9128127369219106,
      "grad_norm": 1.3786677122116089,
      "learning_rate": 1.0916455696202532e-05,
      "loss": 0.08519135415554047,
      "num_input_tokens_seen": 1701632,
      "step": 903,
      "train_runtime": 2718.5394,
      "train_tokens_per_second": 625.936
    },
    {
      "epoch": 0.9138236037402072,
      "grad_norm": 1.207466959953308,
      "learning_rate": 1.0906329113924053e-05,
      "loss": 0.08704199641942978,
      "num_input_tokens_seen": 1703610,
      "step": 904,
      "train_runtime": 2720.8624,
      "train_tokens_per_second": 626.129
    },
    {
      "epoch": 0.9148344705585039,
      "grad_norm": 1.3876044750213623,
      "learning_rate": 1.089620253164557e-05,
      "loss": 0.08722815662622452,
      "num_input_tokens_seen": 1705760,
      "step": 905,
      "train_runtime": 2723.3105,
      "train_tokens_per_second": 626.355
    },
    {
      "epoch": 0.9158453373768006,
      "grad_norm": 1.3396273851394653,
      "learning_rate": 1.088607594936709e-05,
      "loss": 0.11136481910943985,
      "num_input_tokens_seen": 1708032,
      "step": 906,
      "train_runtime": 2725.6741,
      "train_tokens_per_second": 626.646
    },
    {
      "epoch": 0.9168562041950973,
      "grad_norm": 1.7690585851669312,
      "learning_rate": 1.0875949367088607e-05,
      "loss": 0.13523980975151062,
      "num_input_tokens_seen": 1709656,
      "step": 907,
      "train_runtime": 2727.9133,
      "train_tokens_per_second": 626.727
    },
    {
      "epoch": 0.9178670710133939,
      "grad_norm": 1.2731359004974365,
      "learning_rate": 1.0865822784810128e-05,
      "loss": 0.09120108187198639,
      "num_input_tokens_seen": 1711488,
      "step": 908,
      "train_runtime": 2730.1541,
      "train_tokens_per_second": 626.883
    },
    {
      "epoch": 0.9188779378316907,
      "grad_norm": 1.1677757501602173,
      "learning_rate": 1.0855696202531647e-05,
      "loss": 0.07026521861553192,
      "num_input_tokens_seen": 1713238,
      "step": 909,
      "train_runtime": 2732.4122,
      "train_tokens_per_second": 627.006
    },
    {
      "epoch": 0.9198888046499873,
      "grad_norm": 1.2155470848083496,
      "learning_rate": 1.0845569620253164e-05,
      "loss": 0.07531221210956573,
      "num_input_tokens_seen": 1714968,
      "step": 910,
      "train_runtime": 2734.7467,
      "train_tokens_per_second": 627.103
    },
    {
      "epoch": 0.9208996714682841,
      "grad_norm": 1.3503425121307373,
      "learning_rate": 1.0835443037974685e-05,
      "loss": 0.0716508999466896,
      "num_input_tokens_seen": 1716480,
      "step": 911,
      "train_runtime": 2737.0312,
      "train_tokens_per_second": 627.132
    },
    {
      "epoch": 0.9219105382865808,
      "grad_norm": 1.126657485961914,
      "learning_rate": 1.0825316455696204e-05,
      "loss": 0.06158575043082237,
      "num_input_tokens_seen": 1718392,
      "step": 912,
      "train_runtime": 2739.3199,
      "train_tokens_per_second": 627.306
    },
    {
      "epoch": 0.9229214051048774,
      "grad_norm": 1.3263458013534546,
      "learning_rate": 1.0815189873417721e-05,
      "loss": 0.09726858884096146,
      "num_input_tokens_seen": 1720378,
      "step": 913,
      "train_runtime": 2741.6101,
      "train_tokens_per_second": 627.506
    },
    {
      "epoch": 0.9239322719231742,
      "grad_norm": 1.4066224098205566,
      "learning_rate": 1.0805063291139242e-05,
      "loss": 0.11678944528102875,
      "num_input_tokens_seen": 1722602,
      "step": 914,
      "train_runtime": 2743.9506,
      "train_tokens_per_second": 627.782
    },
    {
      "epoch": 0.9249431387414708,
      "grad_norm": 1.2614234685897827,
      "learning_rate": 1.079493670886076e-05,
      "loss": 0.08505284041166306,
      "num_input_tokens_seen": 1724174,
      "step": 915,
      "train_runtime": 2746.1314,
      "train_tokens_per_second": 627.856
    },
    {
      "epoch": 0.9259540055597675,
      "grad_norm": 1.3592069149017334,
      "learning_rate": 1.0784810126582279e-05,
      "loss": 0.10287471860647202,
      "num_input_tokens_seen": 1726094,
      "step": 916,
      "train_runtime": 2748.3985,
      "train_tokens_per_second": 628.036
    },
    {
      "epoch": 0.9269648723780642,
      "grad_norm": 1.3442494869232178,
      "learning_rate": 1.07746835443038e-05,
      "loss": 0.10883361101150513,
      "num_input_tokens_seen": 1728142,
      "step": 917,
      "train_runtime": 2750.6957,
      "train_tokens_per_second": 628.256
    },
    {
      "epoch": 0.9279757391963609,
      "grad_norm": 1.4693677425384521,
      "learning_rate": 1.0764556962025317e-05,
      "loss": 0.08407851308584213,
      "num_input_tokens_seen": 1729944,
      "step": 918,
      "train_runtime": 2752.9092,
      "train_tokens_per_second": 628.406
    },
    {
      "epoch": 0.9289866060146575,
      "grad_norm": 1.1104780435562134,
      "learning_rate": 1.0754430379746836e-05,
      "loss": 0.06933184713125229,
      "num_input_tokens_seen": 1731584,
      "step": 919,
      "train_runtime": 2755.1449,
      "train_tokens_per_second": 628.491
    },
    {
      "epoch": 0.9299974728329543,
      "grad_norm": 1.067386507987976,
      "learning_rate": 1.0744303797468357e-05,
      "loss": 0.08393247425556183,
      "num_input_tokens_seen": 1733392,
      "step": 920,
      "train_runtime": 2757.4007,
      "train_tokens_per_second": 628.633
    },
    {
      "epoch": 0.931008339651251,
      "grad_norm": 1.0293340682983398,
      "learning_rate": 1.0734177215189874e-05,
      "loss": 0.07523457705974579,
      "num_input_tokens_seen": 1735478,
      "step": 921,
      "train_runtime": 2759.7052,
      "train_tokens_per_second": 628.864
    },
    {
      "epoch": 0.9320192064695476,
      "grad_norm": 1.4051296710968018,
      "learning_rate": 1.0724050632911393e-05,
      "loss": 0.10520265996456146,
      "num_input_tokens_seen": 1737024,
      "step": 922,
      "train_runtime": 2761.887,
      "train_tokens_per_second": 628.927
    },
    {
      "epoch": 0.9330300732878444,
      "grad_norm": 1.302826166152954,
      "learning_rate": 1.071392405063291e-05,
      "loss": 0.0870056301355362,
      "num_input_tokens_seen": 1738466,
      "step": 923,
      "train_runtime": 2764.0354,
      "train_tokens_per_second": 628.959
    },
    {
      "epoch": 0.934040940106141,
      "grad_norm": 1.1568562984466553,
      "learning_rate": 1.0703797468354431e-05,
      "loss": 0.08165904134511948,
      "num_input_tokens_seen": 1740182,
      "step": 924,
      "train_runtime": 2766.2755,
      "train_tokens_per_second": 629.07
    },
    {
      "epoch": 0.9350518069244377,
      "grad_norm": 1.498045802116394,
      "learning_rate": 1.069367088607595e-05,
      "loss": 0.07458744943141937,
      "num_input_tokens_seen": 1742024,
      "step": 925,
      "train_runtime": 2768.5295,
      "train_tokens_per_second": 629.224
    },
    {
      "epoch": 0.9360626737427344,
      "grad_norm": 1.5417124032974243,
      "learning_rate": 1.0683544303797468e-05,
      "loss": 0.08173717558383942,
      "num_input_tokens_seen": 1744238,
      "step": 926,
      "train_runtime": 2770.8653,
      "train_tokens_per_second": 629.492
    },
    {
      "epoch": 0.9370735405610311,
      "grad_norm": 1.4843844175338745,
      "learning_rate": 1.0673417721518988e-05,
      "loss": 0.09999965876340866,
      "num_input_tokens_seen": 1746118,
      "step": 927,
      "train_runtime": 2773.1104,
      "train_tokens_per_second": 629.66
    },
    {
      "epoch": 0.9380844073793277,
      "grad_norm": 1.3070656061172485,
      "learning_rate": 1.0663291139240508e-05,
      "loss": 0.08072533458471298,
      "num_input_tokens_seen": 1748128,
      "step": 928,
      "train_runtime": 2775.3922,
      "train_tokens_per_second": 629.867
    },
    {
      "epoch": 0.9390952741976245,
      "grad_norm": 1.1128119230270386,
      "learning_rate": 1.0653164556962025e-05,
      "loss": 0.07749972492456436,
      "num_input_tokens_seen": 1749964,
      "step": 929,
      "train_runtime": 2777.6304,
      "train_tokens_per_second": 630.02
    },
    {
      "epoch": 0.9401061410159212,
      "grad_norm": 1.4054396152496338,
      "learning_rate": 1.0643037974683546e-05,
      "loss": 0.10775762796401978,
      "num_input_tokens_seen": 1752108,
      "step": 930,
      "train_runtime": 2779.9497,
      "train_tokens_per_second": 630.266
    },
    {
      "epoch": 0.9411170078342178,
      "grad_norm": 1.2067521810531616,
      "learning_rate": 1.0632911392405063e-05,
      "loss": 0.061036814004182816,
      "num_input_tokens_seen": 1753674,
      "step": 931,
      "train_runtime": 2795.9291,
      "train_tokens_per_second": 627.224
    },
    {
      "epoch": 0.9421278746525146,
      "grad_norm": 1.5800102949142456,
      "learning_rate": 1.0622784810126582e-05,
      "loss": 0.13530917465686798,
      "num_input_tokens_seen": 1755348,
      "step": 932,
      "train_runtime": 2798.1598,
      "train_tokens_per_second": 627.322
    },
    {
      "epoch": 0.9431387414708112,
      "grad_norm": 1.1216083765029907,
      "learning_rate": 1.0612658227848103e-05,
      "loss": 0.06917855888605118,
      "num_input_tokens_seen": 1757138,
      "step": 933,
      "train_runtime": 2800.4535,
      "train_tokens_per_second": 627.448
    },
    {
      "epoch": 0.9441496082891079,
      "grad_norm": 1.2829315662384033,
      "learning_rate": 1.060253164556962e-05,
      "loss": 0.11217761039733887,
      "num_input_tokens_seen": 1759246,
      "step": 934,
      "train_runtime": 2802.7831,
      "train_tokens_per_second": 627.678
    },
    {
      "epoch": 0.9451604751074046,
      "grad_norm": 1.126326084136963,
      "learning_rate": 1.059240506329114e-05,
      "loss": 0.08738552033901215,
      "num_input_tokens_seen": 1760924,
      "step": 935,
      "train_runtime": 2805.0314,
      "train_tokens_per_second": 627.773
    },
    {
      "epoch": 0.9461713419257013,
      "grad_norm": 1.3069648742675781,
      "learning_rate": 1.058227848101266e-05,
      "loss": 0.09563612192869186,
      "num_input_tokens_seen": 1762612,
      "step": 936,
      "train_runtime": 2807.2487,
      "train_tokens_per_second": 627.879
    },
    {
      "epoch": 0.9471822087439979,
      "grad_norm": 1.8466509580612183,
      "learning_rate": 1.0572151898734178e-05,
      "loss": 0.09266931563615799,
      "num_input_tokens_seen": 1764110,
      "step": 937,
      "train_runtime": 2809.4139,
      "train_tokens_per_second": 627.928
    },
    {
      "epoch": 0.9481930755622947,
      "grad_norm": 1.0844366550445557,
      "learning_rate": 1.0562025316455697e-05,
      "loss": 0.0760909914970398,
      "num_input_tokens_seen": 1765750,
      "step": 938,
      "train_runtime": 2811.6021,
      "train_tokens_per_second": 628.023
    },
    {
      "epoch": 0.9492039423805914,
      "grad_norm": 1.2627480030059814,
      "learning_rate": 1.0551898734177216e-05,
      "loss": 0.08009866625070572,
      "num_input_tokens_seen": 1768042,
      "step": 939,
      "train_runtime": 2813.9799,
      "train_tokens_per_second": 628.307
    },
    {
      "epoch": 0.950214809198888,
      "grad_norm": 1.338079571723938,
      "learning_rate": 1.0541772151898735e-05,
      "loss": 0.09438227117061615,
      "num_input_tokens_seen": 1769708,
      "step": 940,
      "train_runtime": 2816.1705,
      "train_tokens_per_second": 628.409
    },
    {
      "epoch": 0.9512256760171848,
      "grad_norm": 1.1252024173736572,
      "learning_rate": 1.0531645569620254e-05,
      "loss": 0.07405475527048111,
      "num_input_tokens_seen": 1771470,
      "step": 941,
      "train_runtime": 2818.3813,
      "train_tokens_per_second": 628.542
    },
    {
      "epoch": 0.9522365428354814,
      "grad_norm": 1.2941299676895142,
      "learning_rate": 1.0521518987341773e-05,
      "loss": 0.1025523766875267,
      "num_input_tokens_seen": 1773556,
      "step": 942,
      "train_runtime": 2820.7599,
      "train_tokens_per_second": 628.751
    },
    {
      "epoch": 0.9532474096537781,
      "grad_norm": 1.3735921382904053,
      "learning_rate": 1.0511392405063292e-05,
      "loss": 0.10084748268127441,
      "num_input_tokens_seen": 1775544,
      "step": 943,
      "train_runtime": 2823.1414,
      "train_tokens_per_second": 628.925
    },
    {
      "epoch": 0.9542582764720748,
      "grad_norm": 1.2376788854599,
      "learning_rate": 1.0501265822784811e-05,
      "loss": 0.08542808145284653,
      "num_input_tokens_seen": 1777788,
      "step": 944,
      "train_runtime": 2825.491,
      "train_tokens_per_second": 629.196
    },
    {
      "epoch": 0.9552691432903715,
      "grad_norm": 1.7774964570999146,
      "learning_rate": 1.049113924050633e-05,
      "loss": 0.1164168193936348,
      "num_input_tokens_seen": 1779460,
      "step": 945,
      "train_runtime": 2827.6964,
      "train_tokens_per_second": 629.297
    },
    {
      "epoch": 0.9562800101086681,
      "grad_norm": 1.1214243173599243,
      "learning_rate": 1.048101265822785e-05,
      "loss": 0.10965441912412643,
      "num_input_tokens_seen": 1781276,
      "step": 946,
      "train_runtime": 2829.995,
      "train_tokens_per_second": 629.427
    },
    {
      "epoch": 0.9572908769269649,
      "grad_norm": 1.211427927017212,
      "learning_rate": 1.0470886075949367e-05,
      "loss": 0.07722736895084381,
      "num_input_tokens_seen": 1783112,
      "step": 947,
      "train_runtime": 2832.3556,
      "train_tokens_per_second": 629.551
    },
    {
      "epoch": 0.9583017437452616,
      "grad_norm": 1.2366421222686768,
      "learning_rate": 1.0460759493670888e-05,
      "loss": 0.06977664679288864,
      "num_input_tokens_seen": 1784986,
      "step": 948,
      "train_runtime": 2834.7276,
      "train_tokens_per_second": 629.685
    },
    {
      "epoch": 0.9593126105635582,
      "grad_norm": 1.3729935884475708,
      "learning_rate": 1.0450632911392407e-05,
      "loss": 0.12681029736995697,
      "num_input_tokens_seen": 1787244,
      "step": 949,
      "train_runtime": 2837.0511,
      "train_tokens_per_second": 629.965
    },
    {
      "epoch": 0.960323477381855,
      "grad_norm": 1.2358911037445068,
      "learning_rate": 1.0440506329113924e-05,
      "loss": 0.08143652230501175,
      "num_input_tokens_seen": 1788888,
      "step": 950,
      "train_runtime": 2839.3479,
      "train_tokens_per_second": 630.035
    },
    {
      "epoch": 0.9613343442001516,
      "grad_norm": 1.2744375467300415,
      "learning_rate": 1.0430379746835443e-05,
      "loss": 0.07948824018239975,
      "num_input_tokens_seen": 1790766,
      "step": 951,
      "train_runtime": 2841.7231,
      "train_tokens_per_second": 630.169
    },
    {
      "epoch": 0.9623452110184483,
      "grad_norm": 1.2582530975341797,
      "learning_rate": 1.0420253164556964e-05,
      "loss": 0.08794012665748596,
      "num_input_tokens_seen": 1792828,
      "step": 952,
      "train_runtime": 2844.0645,
      "train_tokens_per_second": 630.375
    },
    {
      "epoch": 0.963356077836745,
      "grad_norm": 1.4969607591629028,
      "learning_rate": 1.0410126582278481e-05,
      "loss": 0.09313049167394638,
      "num_input_tokens_seen": 1794592,
      "step": 953,
      "train_runtime": 2846.3064,
      "train_tokens_per_second": 630.499
    },
    {
      "epoch": 0.9643669446550417,
      "grad_norm": 1.5015640258789062,
      "learning_rate": 1.04e-05,
      "loss": 0.09978991001844406,
      "num_input_tokens_seen": 1796582,
      "step": 954,
      "train_runtime": 2848.5542,
      "train_tokens_per_second": 630.7
    },
    {
      "epoch": 0.9653778114733383,
      "grad_norm": 1.805250883102417,
      "learning_rate": 1.038987341772152e-05,
      "loss": 0.09670915454626083,
      "num_input_tokens_seen": 1798762,
      "step": 955,
      "train_runtime": 2850.8744,
      "train_tokens_per_second": 630.951
    },
    {
      "epoch": 0.9663886782916351,
      "grad_norm": 1.4795347452163696,
      "learning_rate": 1.0379746835443039e-05,
      "loss": 0.08996425569057465,
      "num_input_tokens_seen": 1800326,
      "step": 956,
      "train_runtime": 2853.084,
      "train_tokens_per_second": 631.011
    },
    {
      "epoch": 0.9673995451099318,
      "grad_norm": 0.9255862832069397,
      "learning_rate": 1.0369620253164558e-05,
      "loss": 0.058813609182834625,
      "num_input_tokens_seen": 1802300,
      "step": 957,
      "train_runtime": 2855.3629,
      "train_tokens_per_second": 631.198
    },
    {
      "epoch": 0.9684104119282284,
      "grad_norm": 1.4676817655563354,
      "learning_rate": 1.0359493670886077e-05,
      "loss": 0.09787975251674652,
      "num_input_tokens_seen": 1804308,
      "step": 958,
      "train_runtime": 2857.7397,
      "train_tokens_per_second": 631.376
    },
    {
      "epoch": 0.9694212787465252,
      "grad_norm": 1.2104109525680542,
      "learning_rate": 1.0349367088607596e-05,
      "loss": 0.08531485497951508,
      "num_input_tokens_seen": 1806076,
      "step": 959,
      "train_runtime": 2860.0443,
      "train_tokens_per_second": 631.485
    },
    {
      "epoch": 0.9704321455648218,
      "grad_norm": 1.3908240795135498,
      "learning_rate": 1.0339240506329115e-05,
      "loss": 0.06906469166278839,
      "num_input_tokens_seen": 1807730,
      "step": 960,
      "train_runtime": 2862.251,
      "train_tokens_per_second": 631.576
    },
    {
      "epoch": 0.9714430123831185,
      "grad_norm": 1.6727023124694824,
      "learning_rate": 1.0329113924050634e-05,
      "loss": 0.10353363305330276,
      "num_input_tokens_seen": 1809348,
      "step": 961,
      "train_runtime": 2877.8083,
      "train_tokens_per_second": 628.724
    },
    {
      "epoch": 0.9724538792014152,
      "grad_norm": 1.5217154026031494,
      "learning_rate": 1.0318987341772153e-05,
      "loss": 0.10294269770383835,
      "num_input_tokens_seen": 1811064,
      "step": 962,
      "train_runtime": 2880.0629,
      "train_tokens_per_second": 628.828
    },
    {
      "epoch": 0.9734647460197119,
      "grad_norm": 1.4586036205291748,
      "learning_rate": 1.030886075949367e-05,
      "loss": 0.1298624575138092,
      "num_input_tokens_seen": 1813668,
      "step": 963,
      "train_runtime": 2882.509,
      "train_tokens_per_second": 629.198
    },
    {
      "epoch": 0.9744756128380085,
      "grad_norm": 1.3519442081451416,
      "learning_rate": 1.0298734177215191e-05,
      "loss": 0.07855690270662308,
      "num_input_tokens_seen": 1815248,
      "step": 964,
      "train_runtime": 2884.6649,
      "train_tokens_per_second": 629.275
    },
    {
      "epoch": 0.9754864796563053,
      "grad_norm": 1.5139068365097046,
      "learning_rate": 1.028860759493671e-05,
      "loss": 0.07745806872844696,
      "num_input_tokens_seen": 1816818,
      "step": 965,
      "train_runtime": 2886.8776,
      "train_tokens_per_second": 629.337
    },
    {
      "epoch": 0.976497346474602,
      "grad_norm": 1.4229735136032104,
      "learning_rate": 1.0278481012658228e-05,
      "loss": 0.08527548611164093,
      "num_input_tokens_seen": 1818406,
      "step": 966,
      "train_runtime": 2889.1878,
      "train_tokens_per_second": 629.383
    },
    {
      "epoch": 0.9775082132928986,
      "grad_norm": 1.1302425861358643,
      "learning_rate": 1.0268354430379748e-05,
      "loss": 0.08704423904418945,
      "num_input_tokens_seen": 1820820,
      "step": 967,
      "train_runtime": 2891.6001,
      "train_tokens_per_second": 629.693
    },
    {
      "epoch": 0.9785190801111954,
      "grad_norm": 1.5682348012924194,
      "learning_rate": 1.0258227848101268e-05,
      "loss": 0.09315840899944305,
      "num_input_tokens_seen": 1822530,
      "step": 968,
      "train_runtime": 2893.8251,
      "train_tokens_per_second": 629.8
    },
    {
      "epoch": 0.979529946929492,
      "grad_norm": 1.3528974056243896,
      "learning_rate": 1.0248101265822785e-05,
      "loss": 0.07272883504629135,
      "num_input_tokens_seen": 1824318,
      "step": 969,
      "train_runtime": 2896.0731,
      "train_tokens_per_second": 629.928
    },
    {
      "epoch": 0.9805408137477887,
      "grad_norm": 1.1246165037155151,
      "learning_rate": 1.0237974683544306e-05,
      "loss": 0.08196530491113663,
      "num_input_tokens_seen": 1826020,
      "step": 970,
      "train_runtime": 2898.2925,
      "train_tokens_per_second": 630.033
    },
    {
      "epoch": 0.9815516805660854,
      "grad_norm": 1.3558357954025269,
      "learning_rate": 1.0227848101265823e-05,
      "loss": 0.10895586013793945,
      "num_input_tokens_seen": 1827850,
      "step": 971,
      "train_runtime": 2900.5995,
      "train_tokens_per_second": 630.163
    },
    {
      "epoch": 0.9825625473843821,
      "grad_norm": 1.047476887702942,
      "learning_rate": 1.0217721518987342e-05,
      "loss": 0.08318047225475311,
      "num_input_tokens_seen": 1829772,
      "step": 972,
      "train_runtime": 2902.9263,
      "train_tokens_per_second": 630.32
    },
    {
      "epoch": 0.9835734142026787,
      "grad_norm": 1.2297887802124023,
      "learning_rate": 1.0207594936708863e-05,
      "loss": 0.09063711762428284,
      "num_input_tokens_seen": 1831816,
      "step": 973,
      "train_runtime": 2905.3397,
      "train_tokens_per_second": 630.5
    },
    {
      "epoch": 0.9845842810209755,
      "grad_norm": 1.2506780624389648,
      "learning_rate": 1.019746835443038e-05,
      "loss": 0.09520822763442993,
      "num_input_tokens_seen": 1833678,
      "step": 974,
      "train_runtime": 2907.6767,
      "train_tokens_per_second": 630.633
    },
    {
      "epoch": 0.9855951478392722,
      "grad_norm": 1.036617398262024,
      "learning_rate": 1.01873417721519e-05,
      "loss": 0.05149223655462265,
      "num_input_tokens_seen": 1835588,
      "step": 975,
      "train_runtime": 2910.004,
      "train_tokens_per_second": 630.785
    },
    {
      "epoch": 0.9866060146575689,
      "grad_norm": 1.249518632888794,
      "learning_rate": 1.017721518987342e-05,
      "loss": 0.08044559508562088,
      "num_input_tokens_seen": 1837268,
      "step": 976,
      "train_runtime": 2912.2941,
      "train_tokens_per_second": 630.866
    },
    {
      "epoch": 0.9876168814758656,
      "grad_norm": 1.2275365591049194,
      "learning_rate": 1.0167088607594938e-05,
      "loss": 0.09414568543434143,
      "num_input_tokens_seen": 1839268,
      "step": 977,
      "train_runtime": 2914.6048,
      "train_tokens_per_second": 631.052
    },
    {
      "epoch": 0.9886277482941622,
      "grad_norm": 1.383276104927063,
      "learning_rate": 1.0156962025316457e-05,
      "loss": 0.1156046986579895,
      "num_input_tokens_seen": 1841210,
      "step": 978,
      "train_runtime": 2916.9128,
      "train_tokens_per_second": 631.219
    },
    {
      "epoch": 0.989638615112459,
      "grad_norm": 1.388059139251709,
      "learning_rate": 1.0146835443037974e-05,
      "loss": 0.0882212296128273,
      "num_input_tokens_seen": 1842680,
      "step": 979,
      "train_runtime": 2919.0657,
      "train_tokens_per_second": 631.257
    },
    {
      "epoch": 0.9906494819307556,
      "grad_norm": 1.3873894214630127,
      "learning_rate": 1.0136708860759495e-05,
      "loss": 0.11233676970005035,
      "num_input_tokens_seen": 1844666,
      "step": 980,
      "train_runtime": 2921.3698,
      "train_tokens_per_second": 631.439
    },
    {
      "epoch": 0.9916603487490523,
      "grad_norm": 1.4980145692825317,
      "learning_rate": 1.0126582278481014e-05,
      "loss": 0.0733579695224762,
      "num_input_tokens_seen": 1846488,
      "step": 981,
      "train_runtime": 2923.6221,
      "train_tokens_per_second": 631.575
    },
    {
      "epoch": 0.992671215567349,
      "grad_norm": 1.169364333152771,
      "learning_rate": 1.0116455696202531e-05,
      "loss": 0.07496096938848495,
      "num_input_tokens_seen": 1847874,
      "step": 982,
      "train_runtime": 2925.7767,
      "train_tokens_per_second": 631.584
    },
    {
      "epoch": 0.9936820823856457,
      "grad_norm": 1.3312809467315674,
      "learning_rate": 1.0106329113924052e-05,
      "loss": 0.125822514295578,
      "num_input_tokens_seen": 1850558,
      "step": 983,
      "train_runtime": 2928.3045,
      "train_tokens_per_second": 631.955
    },
    {
      "epoch": 0.9946929492039424,
      "grad_norm": 1.3782777786254883,
      "learning_rate": 1.0096202531645571e-05,
      "loss": 0.10531510412693024,
      "num_input_tokens_seen": 1852386,
      "step": 984,
      "train_runtime": 2930.5511,
      "train_tokens_per_second": 632.095
    },
    {
      "epoch": 0.9957038160222391,
      "grad_norm": 1.1993474960327148,
      "learning_rate": 1.0086075949367089e-05,
      "loss": 0.07088446617126465,
      "num_input_tokens_seen": 1853996,
      "step": 985,
      "train_runtime": 2932.7736,
      "train_tokens_per_second": 632.165
    },
    {
      "epoch": 0.9967146828405358,
      "grad_norm": 1.3125388622283936,
      "learning_rate": 1.007594936708861e-05,
      "loss": 0.08627714216709137,
      "num_input_tokens_seen": 1855850,
      "step": 986,
      "train_runtime": 2935.0487,
      "train_tokens_per_second": 632.306
    },
    {
      "epoch": 0.9977255496588324,
      "grad_norm": 1.1183034181594849,
      "learning_rate": 1.0065822784810127e-05,
      "loss": 0.07033257186412811,
      "num_input_tokens_seen": 1857768,
      "step": 987,
      "train_runtime": 2937.326,
      "train_tokens_per_second": 632.469
    },
    {
      "epoch": 0.9987364164771292,
      "grad_norm": 1.2570730447769165,
      "learning_rate": 1.0055696202531646e-05,
      "loss": 0.08110842853784561,
      "num_input_tokens_seen": 1859650,
      "step": 988,
      "train_runtime": 2939.6413,
      "train_tokens_per_second": 632.611
    },
    {
      "epoch": 0.9997472832954258,
      "grad_norm": 1.5044852495193481,
      "learning_rate": 1.0045569620253167e-05,
      "loss": 0.1258259266614914,
      "num_input_tokens_seen": 1861366,
      "step": 989,
      "train_runtime": 2941.8745,
      "train_tokens_per_second": 632.714
    },
    {
      "epoch": 1.0,
      "grad_norm": 21.823984146118164,
      "learning_rate": 1.0035443037974684e-05,
      "loss": 0.24952997267246246,
      "num_input_tokens_seen": 1861552,
      "step": 990,
      "train_runtime": 2946.0328,
      "train_tokens_per_second": 631.884
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.3537242114543915,
      "eval_runtime": 54.2852,
      "eval_samples_per_second": 16.192,
      "eval_steps_per_second": 8.105,
      "num_input_tokens_seen": 1861552,
      "step": 990
    },
    {
      "epoch": 1.0010108668182967,
      "grad_norm": 1.090188980102539,
      "learning_rate": 1.0025316455696203e-05,
      "loss": 0.059413257986307144,
      "num_input_tokens_seen": 1863014,
      "step": 991,
      "train_runtime": 3015.8505,
      "train_tokens_per_second": 617.741
    },
    {
      "epoch": 1.0020217336365933,
      "grad_norm": 1.0046461820602417,
      "learning_rate": 1.0015189873417724e-05,
      "loss": 0.08590734004974365,
      "num_input_tokens_seen": 1865072,
      "step": 992,
      "train_runtime": 3018.1929,
      "train_tokens_per_second": 617.943
    },
    {
      "epoch": 1.0030326004548902,
      "grad_norm": 0.9283597469329834,
      "learning_rate": 1.0005063291139241e-05,
      "loss": 0.07354570925235748,
      "num_input_tokens_seen": 1867012,
      "step": 993,
      "train_runtime": 3020.4699,
      "train_tokens_per_second": 618.12
    },
    {
      "epoch": 1.0040434672731868,
      "grad_norm": 1.0630100965499878,
      "learning_rate": 9.99493670886076e-06,
      "loss": 0.08562401682138443,
      "num_input_tokens_seen": 1868564,
      "step": 994,
      "train_runtime": 3022.6476,
      "train_tokens_per_second": 618.188
    },
    {
      "epoch": 1.0050543340914835,
      "grad_norm": 0.7541443705558777,
      "learning_rate": 9.98481012658228e-06,
      "loss": 0.052490636706352234,
      "num_input_tokens_seen": 1870560,
      "step": 995,
      "train_runtime": 3024.9181,
      "train_tokens_per_second": 618.384
    },
    {
      "epoch": 1.0060652009097801,
      "grad_norm": 0.8982731699943542,
      "learning_rate": 9.974683544303799e-06,
      "loss": 0.0504504069685936,
      "num_input_tokens_seen": 1872450,
      "step": 996,
      "train_runtime": 3027.1886,
      "train_tokens_per_second": 618.544
    },
    {
      "epoch": 1.0070760677280768,
      "grad_norm": 1.615670919418335,
      "learning_rate": 9.964556962025318e-06,
      "loss": 0.07105681300163269,
      "num_input_tokens_seen": 1874370,
      "step": 997,
      "train_runtime": 3029.4799,
      "train_tokens_per_second": 618.71
    },
    {
      "epoch": 1.0080869345463734,
      "grad_norm": 1.0482178926467896,
      "learning_rate": 9.954430379746837e-06,
      "loss": 0.0805853009223938,
      "num_input_tokens_seen": 1876216,
      "step": 998,
      "train_runtime": 3031.7269,
      "train_tokens_per_second": 618.86
    },
    {
      "epoch": 1.0090978013646703,
      "grad_norm": 0.9585389494895935,
      "learning_rate": 9.944303797468356e-06,
      "loss": 0.047966934740543365,
      "num_input_tokens_seen": 1878082,
      "step": 999,
      "train_runtime": 3034.0109,
      "train_tokens_per_second": 619.01
    },
    {
      "epoch": 1.010108668182967,
      "grad_norm": 1.2671263217926025,
      "learning_rate": 9.934177215189875e-06,
      "loss": 0.07086250931024551,
      "num_input_tokens_seen": 1879650,
      "step": 1000,
      "train_runtime": 3036.1834,
      "train_tokens_per_second": 619.083
    },
    {
      "epoch": 1.0111195350012636,
      "grad_norm": 0.9175081849098206,
      "learning_rate": 9.924050632911392e-06,
      "loss": 0.055121179670095444,
      "num_input_tokens_seen": 1881368,
      "step": 1001,
      "train_runtime": 3038.6548,
      "train_tokens_per_second": 619.145
    },
    {
      "epoch": 1.0121304018195603,
      "grad_norm": 1.1538333892822266,
      "learning_rate": 9.913924050632913e-06,
      "loss": 0.051699936389923096,
      "num_input_tokens_seen": 1882806,
      "step": 1002,
      "train_runtime": 3040.8521,
      "train_tokens_per_second": 619.171
    },
    {
      "epoch": 1.013141268637857,
      "grad_norm": 0.9294945597648621,
      "learning_rate": 9.903797468354432e-06,
      "loss": 0.052237533032894135,
      "num_input_tokens_seen": 1884588,
      "step": 1003,
      "train_runtime": 3043.1036,
      "train_tokens_per_second": 619.298
    },
    {
      "epoch": 1.0141521354561536,
      "grad_norm": 1.0442825555801392,
      "learning_rate": 9.89367088607595e-06,
      "loss": 0.06165135279297829,
      "num_input_tokens_seen": 1886554,
      "step": 1004,
      "train_runtime": 3045.3521,
      "train_tokens_per_second": 619.486
    },
    {
      "epoch": 1.0151630022744504,
      "grad_norm": 1.1734623908996582,
      "learning_rate": 9.883544303797469e-06,
      "loss": 0.05909053236246109,
      "num_input_tokens_seen": 1888610,
      "step": 1005,
      "train_runtime": 3047.6581,
      "train_tokens_per_second": 619.692
    },
    {
      "epoch": 1.016173869092747,
      "grad_norm": 1.118070363998413,
      "learning_rate": 9.87341772151899e-06,
      "loss": 0.053689271211624146,
      "num_input_tokens_seen": 1890282,
      "step": 1006,
      "train_runtime": 3049.9039,
      "train_tokens_per_second": 619.784
    },
    {
      "epoch": 1.0171847359110437,
      "grad_norm": 0.7742297053337097,
      "learning_rate": 9.863291139240507e-06,
      "loss": 0.03591981530189514,
      "num_input_tokens_seen": 1891916,
      "step": 1007,
      "train_runtime": 3052.0998,
      "train_tokens_per_second": 619.874
    },
    {
      "epoch": 1.0181956027293404,
      "grad_norm": 1.0873570442199707,
      "learning_rate": 9.853164556962026e-06,
      "loss": 0.051595840603113174,
      "num_input_tokens_seen": 1893622,
      "step": 1008,
      "train_runtime": 3054.3191,
      "train_tokens_per_second": 619.982
    },
    {
      "epoch": 1.019206469547637,
      "grad_norm": 1.2785816192626953,
      "learning_rate": 9.843037974683545e-06,
      "loss": 0.06961002200841904,
      "num_input_tokens_seen": 1895592,
      "step": 1009,
      "train_runtime": 3056.6275,
      "train_tokens_per_second": 620.158
    },
    {
      "epoch": 1.0202173363659337,
      "grad_norm": 0.9817463755607605,
      "learning_rate": 9.832911392405064e-06,
      "loss": 0.050475168973207474,
      "num_input_tokens_seen": 1897352,
      "step": 1010,
      "train_runtime": 3058.8932,
      "train_tokens_per_second": 620.274
    },
    {
      "epoch": 1.0212282031842306,
      "grad_norm": 1.2408863306045532,
      "learning_rate": 9.822784810126583e-06,
      "loss": 0.062181588262319565,
      "num_input_tokens_seen": 1898760,
      "step": 1011,
      "train_runtime": 3061.0348,
      "train_tokens_per_second": 620.3
    },
    {
      "epoch": 1.0222390700025272,
      "grad_norm": 1.0834853649139404,
      "learning_rate": 9.812658227848102e-06,
      "loss": 0.053876448422670364,
      "num_input_tokens_seen": 1900986,
      "step": 1012,
      "train_runtime": 3063.3761,
      "train_tokens_per_second": 620.553
    },
    {
      "epoch": 1.0232499368208239,
      "grad_norm": 1.0059926509857178,
      "learning_rate": 9.802531645569621e-06,
      "loss": 0.045563697814941406,
      "num_input_tokens_seen": 1902980,
      "step": 1013,
      "train_runtime": 3065.6718,
      "train_tokens_per_second": 620.738
    },
    {
      "epoch": 1.0242608036391205,
      "grad_norm": 1.8022197484970093,
      "learning_rate": 9.79240506329114e-06,
      "loss": 0.06333720684051514,
      "num_input_tokens_seen": 1904582,
      "step": 1014,
      "train_runtime": 3067.8646,
      "train_tokens_per_second": 620.817
    },
    {
      "epoch": 1.0252716704574172,
      "grad_norm": 1.1936384439468384,
      "learning_rate": 9.78227848101266e-06,
      "loss": 0.06332344561815262,
      "num_input_tokens_seen": 1906454,
      "step": 1015,
      "train_runtime": 3070.1066,
      "train_tokens_per_second": 620.973
    },
    {
      "epoch": 1.0262825372757138,
      "grad_norm": 1.0198880434036255,
      "learning_rate": 9.772151898734179e-06,
      "loss": 0.05684672296047211,
      "num_input_tokens_seen": 1908566,
      "step": 1016,
      "train_runtime": 3072.4785,
      "train_tokens_per_second": 621.181
    },
    {
      "epoch": 1.0272934040940107,
      "grad_norm": 1.1272902488708496,
      "learning_rate": 9.762025316455696e-06,
      "loss": 0.04964352026581764,
      "num_input_tokens_seen": 1910098,
      "step": 1017,
      "train_runtime": 3074.6987,
      "train_tokens_per_second": 621.231
    },
    {
      "epoch": 1.0283042709123074,
      "grad_norm": 1.453578233718872,
      "learning_rate": 9.751898734177217e-06,
      "loss": 0.035638563334941864,
      "num_input_tokens_seen": 1911660,
      "step": 1018,
      "train_runtime": 3076.9556,
      "train_tokens_per_second": 621.283
    },
    {
      "epoch": 1.029315137730604,
      "grad_norm": 1.4090577363967896,
      "learning_rate": 9.741772151898736e-06,
      "loss": 0.08451429754495621,
      "num_input_tokens_seen": 1914106,
      "step": 1019,
      "train_runtime": 3079.3753,
      "train_tokens_per_second": 621.589
    },
    {
      "epoch": 1.0303260045489007,
      "grad_norm": 1.227582335472107,
      "learning_rate": 9.731645569620253e-06,
      "loss": 0.051470160484313965,
      "num_input_tokens_seen": 1916250,
      "step": 1020,
      "train_runtime": 3081.737,
      "train_tokens_per_second": 621.808
    },
    {
      "epoch": 1.0313368713671973,
      "grad_norm": 1.048599362373352,
      "learning_rate": 9.721518987341772e-06,
      "loss": 0.04310810565948486,
      "num_input_tokens_seen": 1918076,
      "step": 1021,
      "train_runtime": 3096.2147,
      "train_tokens_per_second": 619.491
    },
    {
      "epoch": 1.032347738185494,
      "grad_norm": 1.0663803815841675,
      "learning_rate": 9.711392405063293e-06,
      "loss": 0.05646198242902756,
      "num_input_tokens_seen": 1919904,
      "step": 1022,
      "train_runtime": 3098.5638,
      "train_tokens_per_second": 619.611
    },
    {
      "epoch": 1.0333586050037908,
      "grad_norm": 0.9684441685676575,
      "learning_rate": 9.70126582278481e-06,
      "loss": 0.04822005331516266,
      "num_input_tokens_seen": 1921690,
      "step": 1023,
      "train_runtime": 3100.8399,
      "train_tokens_per_second": 619.732
    },
    {
      "epoch": 1.0343694718220875,
      "grad_norm": 1.1114486455917358,
      "learning_rate": 9.69113924050633e-06,
      "loss": 0.05490822717547417,
      "num_input_tokens_seen": 1923734,
      "step": 1024,
      "train_runtime": 3103.1667,
      "train_tokens_per_second": 619.926
    },
    {
      "epoch": 1.0353803386403841,
      "grad_norm": 0.8609014749526978,
      "learning_rate": 9.681012658227849e-06,
      "loss": 0.03940562903881073,
      "num_input_tokens_seen": 1925552,
      "step": 1025,
      "train_runtime": 3105.4492,
      "train_tokens_per_second": 620.056
    },
    {
      "epoch": 1.0363912054586808,
      "grad_norm": 1.2412512302398682,
      "learning_rate": 9.670886075949368e-06,
      "loss": 0.07361410558223724,
      "num_input_tokens_seen": 1927862,
      "step": 1026,
      "train_runtime": 3107.826,
      "train_tokens_per_second": 620.325
    },
    {
      "epoch": 1.0374020722769774,
      "grad_norm": 1.5505046844482422,
      "learning_rate": 9.660759493670887e-06,
      "loss": 0.06292421370744705,
      "num_input_tokens_seen": 1930110,
      "step": 1027,
      "train_runtime": 3110.1941,
      "train_tokens_per_second": 620.575
    },
    {
      "epoch": 1.038412939095274,
      "grad_norm": 1.5445483922958374,
      "learning_rate": 9.650632911392406e-06,
      "loss": 0.06876043975353241,
      "num_input_tokens_seen": 1932032,
      "step": 1028,
      "train_runtime": 3112.4815,
      "train_tokens_per_second": 620.737
    },
    {
      "epoch": 1.039423805913571,
      "grad_norm": 1.0396747589111328,
      "learning_rate": 9.640506329113925e-06,
      "loss": 0.04588109254837036,
      "num_input_tokens_seen": 1934146,
      "step": 1029,
      "train_runtime": 3115.1357,
      "train_tokens_per_second": 620.887
    },
    {
      "epoch": 1.0404346727318676,
      "grad_norm": 1.227482557296753,
      "learning_rate": 9.630379746835444e-06,
      "loss": 0.0684301108121872,
      "num_input_tokens_seen": 1936018,
      "step": 1030,
      "train_runtime": 3117.4068,
      "train_tokens_per_second": 621.035
    },
    {
      "epoch": 1.0414455395501643,
      "grad_norm": 1.1692776679992676,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.06540321558713913,
      "num_input_tokens_seen": 1937840,
      "step": 1031,
      "train_runtime": 3119.7217,
      "train_tokens_per_second": 621.158
    },
    {
      "epoch": 1.042456406368461,
      "grad_norm": 1.1892913579940796,
      "learning_rate": 9.610126582278482e-06,
      "loss": 0.0633426383137703,
      "num_input_tokens_seen": 1939736,
      "step": 1032,
      "train_runtime": 3122.08,
      "train_tokens_per_second": 621.296
    },
    {
      "epoch": 1.0434672731867576,
      "grad_norm": 1.1392109394073486,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.06080387160181999,
      "num_input_tokens_seen": 1941232,
      "step": 1033,
      "train_runtime": 3124.3392,
      "train_tokens_per_second": 621.326
    },
    {
      "epoch": 1.0444781400050542,
      "grad_norm": 1.3295029401779175,
      "learning_rate": 9.58987341772152e-06,
      "loss": 0.06093157082796097,
      "num_input_tokens_seen": 1943502,
      "step": 1034,
      "train_runtime": 3126.7182,
      "train_tokens_per_second": 621.579
    },
    {
      "epoch": 1.045489006823351,
      "grad_norm": 1.7235727310180664,
      "learning_rate": 9.57974683544304e-06,
      "loss": 0.08213045448064804,
      "num_input_tokens_seen": 1945588,
      "step": 1035,
      "train_runtime": 3129.0319,
      "train_tokens_per_second": 621.786
    },
    {
      "epoch": 1.0464998736416478,
      "grad_norm": 1.2702877521514893,
      "learning_rate": 9.569620253164559e-06,
      "loss": 0.057667411863803864,
      "num_input_tokens_seen": 1947762,
      "step": 1036,
      "train_runtime": 3131.3751,
      "train_tokens_per_second": 622.015
    },
    {
      "epoch": 1.0475107404599444,
      "grad_norm": 1.0603086948394775,
      "learning_rate": 9.559493670886076e-06,
      "loss": 0.051095783710479736,
      "num_input_tokens_seen": 1949644,
      "step": 1037,
      "train_runtime": 3133.7008,
      "train_tokens_per_second": 622.154
    },
    {
      "epoch": 1.048521607278241,
      "grad_norm": 1.0244430303573608,
      "learning_rate": 9.549367088607597e-06,
      "loss": 0.04529440030455589,
      "num_input_tokens_seen": 1951800,
      "step": 1038,
      "train_runtime": 3136.1466,
      "train_tokens_per_second": 622.356
    },
    {
      "epoch": 1.0495324740965377,
      "grad_norm": 1.6149035692214966,
      "learning_rate": 9.539240506329114e-06,
      "loss": 0.06959658861160278,
      "num_input_tokens_seen": 1953690,
      "step": 1039,
      "train_runtime": 3138.4268,
      "train_tokens_per_second": 622.506
    },
    {
      "epoch": 1.0505433409148344,
      "grad_norm": 0.9820495247840881,
      "learning_rate": 9.529113924050633e-06,
      "loss": 0.05469644069671631,
      "num_input_tokens_seen": 1955684,
      "step": 1040,
      "train_runtime": 3140.7473,
      "train_tokens_per_second": 622.681
    },
    {
      "epoch": 1.0515542077331312,
      "grad_norm": 1.5499889850616455,
      "learning_rate": 9.518987341772152e-06,
      "loss": 0.07478737831115723,
      "num_input_tokens_seen": 1957194,
      "step": 1041,
      "train_runtime": 3142.9782,
      "train_tokens_per_second": 622.72
    },
    {
      "epoch": 1.052565074551428,
      "grad_norm": 0.9262644052505493,
      "learning_rate": 9.508860759493671e-06,
      "loss": 0.048888735473155975,
      "num_input_tokens_seen": 1959200,
      "step": 1042,
      "train_runtime": 3145.3251,
      "train_tokens_per_second": 622.893
    },
    {
      "epoch": 1.0535759413697245,
      "grad_norm": 1.365299105644226,
      "learning_rate": 9.49873417721519e-06,
      "loss": 0.0644601434469223,
      "num_input_tokens_seen": 1960834,
      "step": 1043,
      "train_runtime": 3147.5374,
      "train_tokens_per_second": 622.974
    },
    {
      "epoch": 1.0545868081880212,
      "grad_norm": 2.561173677444458,
      "learning_rate": 9.48860759493671e-06,
      "loss": 0.06372687220573425,
      "num_input_tokens_seen": 1962374,
      "step": 1044,
      "train_runtime": 3149.7262,
      "train_tokens_per_second": 623.03
    },
    {
      "epoch": 1.0555976750063178,
      "grad_norm": 1.5165947675704956,
      "learning_rate": 9.478481012658229e-06,
      "loss": 0.071042999625206,
      "num_input_tokens_seen": 1964462,
      "step": 1045,
      "train_runtime": 3152.0308,
      "train_tokens_per_second": 623.237
    },
    {
      "epoch": 1.0566085418246145,
      "grad_norm": 1.3632253408432007,
      "learning_rate": 9.468354430379748e-06,
      "loss": 0.07348132133483887,
      "num_input_tokens_seen": 1966108,
      "step": 1046,
      "train_runtime": 3154.2419,
      "train_tokens_per_second": 623.322
    },
    {
      "epoch": 1.0576194086429114,
      "grad_norm": 1.3681747913360596,
      "learning_rate": 9.458227848101267e-06,
      "loss": 0.06395052373409271,
      "num_input_tokens_seen": 1968492,
      "step": 1047,
      "train_runtime": 3156.7553,
      "train_tokens_per_second": 623.581
    },
    {
      "epoch": 1.058630275461208,
      "grad_norm": 1.3107277154922485,
      "learning_rate": 9.448101265822786e-06,
      "loss": 0.0616154819726944,
      "num_input_tokens_seen": 1970174,
      "step": 1048,
      "train_runtime": 3159.0565,
      "train_tokens_per_second": 623.659
    },
    {
      "epoch": 1.0596411422795047,
      "grad_norm": 1.3554524183273315,
      "learning_rate": 9.437974683544305e-06,
      "loss": 0.047980040311813354,
      "num_input_tokens_seen": 1972186,
      "step": 1049,
      "train_runtime": 3161.3657,
      "train_tokens_per_second": 623.84
    },
    {
      "epoch": 1.0606520090978013,
      "grad_norm": 1.1780120134353638,
      "learning_rate": 9.427848101265824e-06,
      "loss": 0.07722882181406021,
      "num_input_tokens_seen": 1974348,
      "step": 1050,
      "train_runtime": 3163.7132,
      "train_tokens_per_second": 624.06
    },
    {
      "epoch": 1.061662875916098,
      "grad_norm": 0.9879222512245178,
      "learning_rate": 9.417721518987343e-06,
      "loss": 0.02780820056796074,
      "num_input_tokens_seen": 1976044,
      "step": 1051,
      "train_runtime": 3178.1542,
      "train_tokens_per_second": 621.758
    },
    {
      "epoch": 1.0626737427343946,
      "grad_norm": 1.1026676893234253,
      "learning_rate": 9.407594936708862e-06,
      "loss": 0.049638256430625916,
      "num_input_tokens_seen": 1977710,
      "step": 1052,
      "train_runtime": 3180.4168,
      "train_tokens_per_second": 621.84
    },
    {
      "epoch": 1.0636846095526915,
      "grad_norm": 1.2987622022628784,
      "learning_rate": 9.39746835443038e-06,
      "loss": 0.07319340109825134,
      "num_input_tokens_seen": 1979460,
      "step": 1053,
      "train_runtime": 3182.7512,
      "train_tokens_per_second": 621.934
    },
    {
      "epoch": 1.0646954763709882,
      "grad_norm": 1.4163175821304321,
      "learning_rate": 9.3873417721519e-06,
      "loss": 0.054538704454898834,
      "num_input_tokens_seen": 1981128,
      "step": 1054,
      "train_runtime": 3184.9272,
      "train_tokens_per_second": 622.032
    },
    {
      "epoch": 1.0657063431892848,
      "grad_norm": 3.484403133392334,
      "learning_rate": 9.37721518987342e-06,
      "loss": 0.0736357644200325,
      "num_input_tokens_seen": 1983712,
      "step": 1055,
      "train_runtime": 3187.3684,
      "train_tokens_per_second": 622.367
    },
    {
      "epoch": 1.0667172100075815,
      "grad_norm": 1.5456719398498535,
      "learning_rate": 9.367088607594937e-06,
      "loss": 0.04635040834546089,
      "num_input_tokens_seen": 1985418,
      "step": 1056,
      "train_runtime": 3189.6463,
      "train_tokens_per_second": 622.457
    },
    {
      "epoch": 1.0677280768258781,
      "grad_norm": 1.1814062595367432,
      "learning_rate": 9.356962025316456e-06,
      "loss": 0.06225111335515976,
      "num_input_tokens_seen": 1987526,
      "step": 1057,
      "train_runtime": 3192.012,
      "train_tokens_per_second": 622.656
    },
    {
      "epoch": 1.068738943644175,
      "grad_norm": 1.2772281169891357,
      "learning_rate": 9.346835443037977e-06,
      "loss": 0.04973405972123146,
      "num_input_tokens_seen": 1989352,
      "step": 1058,
      "train_runtime": 3194.2846,
      "train_tokens_per_second": 622.785
    },
    {
      "epoch": 1.0697498104624716,
      "grad_norm": 1.4468482732772827,
      "learning_rate": 9.336708860759494e-06,
      "loss": 0.06707950681447983,
      "num_input_tokens_seen": 1991022,
      "step": 1059,
      "train_runtime": 3196.4927,
      "train_tokens_per_second": 622.877
    },
    {
      "epoch": 1.0707606772807683,
      "grad_norm": 1.2882323265075684,
      "learning_rate": 9.326582278481013e-06,
      "loss": 0.06885062903165817,
      "num_input_tokens_seen": 1992680,
      "step": 1060,
      "train_runtime": 3198.7043,
      "train_tokens_per_second": 622.965
    },
    {
      "epoch": 1.071771544099065,
      "grad_norm": 1.5417802333831787,
      "learning_rate": 9.316455696202532e-06,
      "loss": 0.06702878326177597,
      "num_input_tokens_seen": 1995068,
      "step": 1061,
      "train_runtime": 3201.3821,
      "train_tokens_per_second": 623.19
    },
    {
      "epoch": 1.0727824109173616,
      "grad_norm": 1.1460639238357544,
      "learning_rate": 9.306329113924051e-06,
      "loss": 0.05765167623758316,
      "num_input_tokens_seen": 1996926,
      "step": 1062,
      "train_runtime": 3203.7526,
      "train_tokens_per_second": 623.308
    },
    {
      "epoch": 1.0737932777356582,
      "grad_norm": 1.452125906944275,
      "learning_rate": 9.29620253164557e-06,
      "loss": 0.0469055250287056,
      "num_input_tokens_seen": 1998294,
      "step": 1063,
      "train_runtime": 3205.8998,
      "train_tokens_per_second": 623.318
    },
    {
      "epoch": 1.074804144553955,
      "grad_norm": 1.635292649269104,
      "learning_rate": 9.28607594936709e-06,
      "loss": 0.07289348542690277,
      "num_input_tokens_seen": 2000332,
      "step": 1064,
      "train_runtime": 3208.1716,
      "train_tokens_per_second": 623.512
    },
    {
      "epoch": 1.0758150113722518,
      "grad_norm": 1.4546730518341064,
      "learning_rate": 9.275949367088609e-06,
      "loss": 0.06234751641750336,
      "num_input_tokens_seen": 2002028,
      "step": 1065,
      "train_runtime": 3210.3387,
      "train_tokens_per_second": 623.619
    },
    {
      "epoch": 1.0768258781905484,
      "grad_norm": 1.0785565376281738,
      "learning_rate": 9.265822784810128e-06,
      "loss": 0.05137242004275322,
      "num_input_tokens_seen": 2004148,
      "step": 1066,
      "train_runtime": 3212.6724,
      "train_tokens_per_second": 623.826
    },
    {
      "epoch": 1.077836745008845,
      "grad_norm": 1.1812286376953125,
      "learning_rate": 9.255696202531647e-06,
      "loss": 0.05945739522576332,
      "num_input_tokens_seen": 2006290,
      "step": 1067,
      "train_runtime": 3214.9667,
      "train_tokens_per_second": 624.047
    },
    {
      "epoch": 1.0788476118271417,
      "grad_norm": 1.445425033569336,
      "learning_rate": 9.245569620253166e-06,
      "loss": 0.05825085937976837,
      "num_input_tokens_seen": 2008128,
      "step": 1068,
      "train_runtime": 3217.1989,
      "train_tokens_per_second": 624.185
    },
    {
      "epoch": 1.0798584786454384,
      "grad_norm": 1.2002581357955933,
      "learning_rate": 9.235443037974683e-06,
      "loss": 0.05513761565089226,
      "num_input_tokens_seen": 2010360,
      "step": 1069,
      "train_runtime": 3219.5556,
      "train_tokens_per_second": 624.422
    },
    {
      "epoch": 1.0808693454637353,
      "grad_norm": 1.1316932439804077,
      "learning_rate": 9.225316455696204e-06,
      "loss": 0.05086222290992737,
      "num_input_tokens_seen": 2012614,
      "step": 1070,
      "train_runtime": 3221.8849,
      "train_tokens_per_second": 624.67
    },
    {
      "epoch": 1.081880212282032,
      "grad_norm": 1.5588624477386475,
      "learning_rate": 9.215189873417723e-06,
      "loss": 0.05340712144970894,
      "num_input_tokens_seen": 2014350,
      "step": 1071,
      "train_runtime": 3224.1143,
      "train_tokens_per_second": 624.776
    },
    {
      "epoch": 1.0828910791003286,
      "grad_norm": 1.281995177268982,
      "learning_rate": 9.20506329113924e-06,
      "loss": 0.06859094649553299,
      "num_input_tokens_seen": 2016280,
      "step": 1072,
      "train_runtime": 3226.3736,
      "train_tokens_per_second": 624.937
    },
    {
      "epoch": 1.0839019459186252,
      "grad_norm": 1.3881559371948242,
      "learning_rate": 9.19493670886076e-06,
      "loss": 0.06530670821666718,
      "num_input_tokens_seen": 2018160,
      "step": 1073,
      "train_runtime": 3228.5993,
      "train_tokens_per_second": 625.088
    },
    {
      "epoch": 1.0849128127369219,
      "grad_norm": 1.2683074474334717,
      "learning_rate": 9.18481012658228e-06,
      "loss": 0.0578412264585495,
      "num_input_tokens_seen": 2020416,
      "step": 1074,
      "train_runtime": 3230.959,
      "train_tokens_per_second": 625.33
    },
    {
      "epoch": 1.0859236795552185,
      "grad_norm": 1.052309274673462,
      "learning_rate": 9.174683544303798e-06,
      "loss": 0.059152744710445404,
      "num_input_tokens_seen": 2022348,
      "step": 1075,
      "train_runtime": 3233.2263,
      "train_tokens_per_second": 625.489
    },
    {
      "epoch": 1.0869345463735154,
      "grad_norm": 1.2832680940628052,
      "learning_rate": 9.164556962025317e-06,
      "loss": 0.05791567265987396,
      "num_input_tokens_seen": 2024030,
      "step": 1076,
      "train_runtime": 3235.4212,
      "train_tokens_per_second": 625.585
    },
    {
      "epoch": 1.087945413191812,
      "grad_norm": 1.4182716608047485,
      "learning_rate": 9.154430379746836e-06,
      "loss": 0.06937595456838608,
      "num_input_tokens_seen": 2025526,
      "step": 1077,
      "train_runtime": 3237.5638,
      "train_tokens_per_second": 625.633
    },
    {
      "epoch": 1.0889562800101087,
      "grad_norm": 1.2201087474822998,
      "learning_rate": 9.144303797468355e-06,
      "loss": 0.05016665160655975,
      "num_input_tokens_seen": 2027310,
      "step": 1078,
      "train_runtime": 3239.8149,
      "train_tokens_per_second": 625.749
    },
    {
      "epoch": 1.0899671468284053,
      "grad_norm": 1.0213748216629028,
      "learning_rate": 9.134177215189874e-06,
      "loss": 0.045126307755708694,
      "num_input_tokens_seen": 2029130,
      "step": 1079,
      "train_runtime": 3242.0877,
      "train_tokens_per_second": 625.871
    },
    {
      "epoch": 1.090978013646702,
      "grad_norm": 1.107077956199646,
      "learning_rate": 9.124050632911393e-06,
      "loss": 0.05676788091659546,
      "num_input_tokens_seen": 2031392,
      "step": 1080,
      "train_runtime": 3244.453,
      "train_tokens_per_second": 626.112
    },
    {
      "epoch": 1.0919888804649986,
      "grad_norm": 1.2918614149093628,
      "learning_rate": 9.113924050632912e-06,
      "loss": 0.06851403415203094,
      "num_input_tokens_seen": 2033048,
      "step": 1081,
      "train_runtime": 3258.8928,
      "train_tokens_per_second": 623.846
    },
    {
      "epoch": 1.0929997472832955,
      "grad_norm": 0.9818199276924133,
      "learning_rate": 9.103797468354431e-06,
      "loss": 0.047627776861190796,
      "num_input_tokens_seen": 2034860,
      "step": 1082,
      "train_runtime": 3261.1933,
      "train_tokens_per_second": 623.962
    },
    {
      "epoch": 1.0940106141015922,
      "grad_norm": 1.2099056243896484,
      "learning_rate": 9.09367088607595e-06,
      "loss": 0.04850488528609276,
      "num_input_tokens_seen": 2036578,
      "step": 1083,
      "train_runtime": 3263.4922,
      "train_tokens_per_second": 624.049
    },
    {
      "epoch": 1.0950214809198888,
      "grad_norm": 1.2491371631622314,
      "learning_rate": 9.08354430379747e-06,
      "loss": 0.04796820878982544,
      "num_input_tokens_seen": 2038364,
      "step": 1084,
      "train_runtime": 3265.7635,
      "train_tokens_per_second": 624.162
    },
    {
      "epoch": 1.0960323477381855,
      "grad_norm": 0.9802494049072266,
      "learning_rate": 9.073417721518989e-06,
      "loss": 0.033174172043800354,
      "num_input_tokens_seen": 2040330,
      "step": 1085,
      "train_runtime": 3268.1811,
      "train_tokens_per_second": 624.301
    },
    {
      "epoch": 1.0970432145564821,
      "grad_norm": 1.0527691841125488,
      "learning_rate": 9.063291139240506e-06,
      "loss": 0.03538256138563156,
      "num_input_tokens_seen": 2042242,
      "step": 1086,
      "train_runtime": 3270.5916,
      "train_tokens_per_second": 624.426
    },
    {
      "epoch": 1.0980540813747788,
      "grad_norm": 1.4610166549682617,
      "learning_rate": 9.053164556962027e-06,
      "loss": 0.06688831746578217,
      "num_input_tokens_seen": 2044062,
      "step": 1087,
      "train_runtime": 3272.8711,
      "train_tokens_per_second": 624.547
    },
    {
      "epoch": 1.0990649481930757,
      "grad_norm": 1.2289828062057495,
      "learning_rate": 9.043037974683546e-06,
      "loss": 0.05300654470920563,
      "num_input_tokens_seen": 2045578,
      "step": 1088,
      "train_runtime": 3275.0406,
      "train_tokens_per_second": 624.596
    },
    {
      "epoch": 1.1000758150113723,
      "grad_norm": 1.2802950143814087,
      "learning_rate": 9.032911392405063e-06,
      "loss": 0.05777037516236305,
      "num_input_tokens_seen": 2047120,
      "step": 1089,
      "train_runtime": 3277.2231,
      "train_tokens_per_second": 624.651
    },
    {
      "epoch": 1.101086681829669,
      "grad_norm": 1.4316028356552124,
      "learning_rate": 9.022784810126582e-06,
      "loss": 0.06609408557415009,
      "num_input_tokens_seen": 2048692,
      "step": 1090,
      "train_runtime": 3279.3916,
      "train_tokens_per_second": 624.717
    },
    {
      "epoch": 1.1020975486479656,
      "grad_norm": 1.4526844024658203,
      "learning_rate": 9.012658227848103e-06,
      "loss": 0.0704093649983406,
      "num_input_tokens_seen": 2050836,
      "step": 1091,
      "train_runtime": 3281.6971,
      "train_tokens_per_second": 624.932
    },
    {
      "epoch": 1.1031084154662623,
      "grad_norm": 1.185599684715271,
      "learning_rate": 9.00253164556962e-06,
      "loss": 0.08056314289569855,
      "num_input_tokens_seen": 2052972,
      "step": 1092,
      "train_runtime": 3284.0631,
      "train_tokens_per_second": 625.132
    },
    {
      "epoch": 1.104119282284559,
      "grad_norm": 1.4482617378234863,
      "learning_rate": 8.99240506329114e-06,
      "loss": 0.052174363285303116,
      "num_input_tokens_seen": 2055178,
      "step": 1093,
      "train_runtime": 3286.4338,
      "train_tokens_per_second": 625.352
    },
    {
      "epoch": 1.1051301491028558,
      "grad_norm": 1.218012809753418,
      "learning_rate": 8.982278481012659e-06,
      "loss": 0.06104224920272827,
      "num_input_tokens_seen": 2056806,
      "step": 1094,
      "train_runtime": 3288.6456,
      "train_tokens_per_second": 625.426
    },
    {
      "epoch": 1.1061410159211524,
      "grad_norm": 1.0124777555465698,
      "learning_rate": 8.972151898734178e-06,
      "loss": 0.04813598096370697,
      "num_input_tokens_seen": 2059252,
      "step": 1095,
      "train_runtime": 3291.0474,
      "train_tokens_per_second": 625.713
    },
    {
      "epoch": 1.107151882739449,
      "grad_norm": 0.9334543943405151,
      "learning_rate": 8.962025316455697e-06,
      "loss": 0.03611904755234718,
      "num_input_tokens_seen": 2061036,
      "step": 1096,
      "train_runtime": 3293.2934,
      "train_tokens_per_second": 625.828
    },
    {
      "epoch": 1.1081627495577457,
      "grad_norm": 1.174657940864563,
      "learning_rate": 8.951898734177216e-06,
      "loss": 0.05326293408870697,
      "num_input_tokens_seen": 2063098,
      "step": 1097,
      "train_runtime": 3295.5584,
      "train_tokens_per_second": 626.024
    },
    {
      "epoch": 1.1091736163760424,
      "grad_norm": 1.4373377561569214,
      "learning_rate": 8.941772151898735e-06,
      "loss": 0.07180237025022507,
      "num_input_tokens_seen": 2065018,
      "step": 1098,
      "train_runtime": 3297.8642,
      "train_tokens_per_second": 626.168
    },
    {
      "epoch": 1.110184483194339,
      "grad_norm": 1.0424410104751587,
      "learning_rate": 8.931645569620254e-06,
      "loss": 0.0501631423830986,
      "num_input_tokens_seen": 2067042,
      "step": 1099,
      "train_runtime": 3300.125,
      "train_tokens_per_second": 626.353
    },
    {
      "epoch": 1.111195350012636,
      "grad_norm": 1.1881895065307617,
      "learning_rate": 8.921518987341773e-06,
      "loss": 0.05463118851184845,
      "num_input_tokens_seen": 2068762,
      "step": 1100,
      "train_runtime": 3302.376,
      "train_tokens_per_second": 626.447
    },
    {
      "epoch": 1.1122062168309326,
      "grad_norm": 1.6544052362442017,
      "learning_rate": 8.911392405063292e-06,
      "loss": 0.06034505367279053,
      "num_input_tokens_seen": 2070476,
      "step": 1101,
      "train_runtime": 3304.6072,
      "train_tokens_per_second": 626.542
    },
    {
      "epoch": 1.1132170836492292,
      "grad_norm": 1.0974977016448975,
      "learning_rate": 8.90126582278481e-06,
      "loss": 0.05488116666674614,
      "num_input_tokens_seen": 2072284,
      "step": 1102,
      "train_runtime": 3306.8385,
      "train_tokens_per_second": 626.666
    },
    {
      "epoch": 1.1142279504675259,
      "grad_norm": 1.380908727645874,
      "learning_rate": 8.89113924050633e-06,
      "loss": 0.053319301456213,
      "num_input_tokens_seen": 2074296,
      "step": 1103,
      "train_runtime": 3309.183,
      "train_tokens_per_second": 626.83
    },
    {
      "epoch": 1.1152388172858225,
      "grad_norm": 1.1827045679092407,
      "learning_rate": 8.88101265822785e-06,
      "loss": 0.045568160712718964,
      "num_input_tokens_seen": 2076596,
      "step": 1104,
      "train_runtime": 3311.5583,
      "train_tokens_per_second": 627.075
    },
    {
      "epoch": 1.1162496841041194,
      "grad_norm": 1.226758599281311,
      "learning_rate": 8.870886075949367e-06,
      "loss": 0.0543665811419487,
      "num_input_tokens_seen": 2078276,
      "step": 1105,
      "train_runtime": 3313.7593,
      "train_tokens_per_second": 627.166
    },
    {
      "epoch": 1.117260550922416,
      "grad_norm": 1.2752267122268677,
      "learning_rate": 8.860759493670886e-06,
      "loss": 0.06291696429252625,
      "num_input_tokens_seen": 2080060,
      "step": 1106,
      "train_runtime": 3315.9981,
      "train_tokens_per_second": 627.28
    },
    {
      "epoch": 1.1182714177407127,
      "grad_norm": 1.2787002325057983,
      "learning_rate": 8.850632911392407e-06,
      "loss": 0.0689842626452446,
      "num_input_tokens_seen": 2082446,
      "step": 1107,
      "train_runtime": 3318.3661,
      "train_tokens_per_second": 627.552
    },
    {
      "epoch": 1.1192822845590094,
      "grad_norm": 1.2914907932281494,
      "learning_rate": 8.840506329113924e-06,
      "loss": 0.052648428827524185,
      "num_input_tokens_seen": 2083832,
      "step": 1108,
      "train_runtime": 3320.5176,
      "train_tokens_per_second": 627.562
    },
    {
      "epoch": 1.120293151377306,
      "grad_norm": 1.1470319032669067,
      "learning_rate": 8.830379746835443e-06,
      "loss": 0.0601373016834259,
      "num_input_tokens_seen": 2086060,
      "step": 1109,
      "train_runtime": 3322.8417,
      "train_tokens_per_second": 627.794
    },
    {
      "epoch": 1.1213040181956027,
      "grad_norm": 1.1858102083206177,
      "learning_rate": 8.820253164556962e-06,
      "loss": 0.03755956143140793,
      "num_input_tokens_seen": 2087998,
      "step": 1110,
      "train_runtime": 3325.106,
      "train_tokens_per_second": 627.949
    },
    {
      "epoch": 1.1223148850138993,
      "grad_norm": 1.2062602043151855,
      "learning_rate": 8.810126582278481e-06,
      "loss": 0.05221632495522499,
      "num_input_tokens_seen": 2090364,
      "step": 1111,
      "train_runtime": 3341.8286,
      "train_tokens_per_second": 625.515
    },
    {
      "epoch": 1.1233257518321962,
      "grad_norm": 0.9900615811347961,
      "learning_rate": 8.8e-06,
      "loss": 0.039271071553230286,
      "num_input_tokens_seen": 2092166,
      "step": 1112,
      "train_runtime": 3344.0715,
      "train_tokens_per_second": 625.634
    },
    {
      "epoch": 1.1243366186504928,
      "grad_norm": 1.0736415386199951,
      "learning_rate": 8.78987341772152e-06,
      "loss": 0.048174936324357986,
      "num_input_tokens_seen": 2094464,
      "step": 1113,
      "train_runtime": 3346.4141,
      "train_tokens_per_second": 625.883
    },
    {
      "epoch": 1.1253474854687895,
      "grad_norm": 1.6410328149795532,
      "learning_rate": 8.779746835443039e-06,
      "loss": 0.11351293325424194,
      "num_input_tokens_seen": 2096704,
      "step": 1114,
      "train_runtime": 3348.8682,
      "train_tokens_per_second": 626.093
    },
    {
      "epoch": 1.1263583522870861,
      "grad_norm": 1.3577684164047241,
      "learning_rate": 8.769620253164558e-06,
      "loss": 0.06730435788631439,
      "num_input_tokens_seen": 2098626,
      "step": 1115,
      "train_runtime": 3351.1424,
      "train_tokens_per_second": 626.242
    },
    {
      "epoch": 1.1273692191053828,
      "grad_norm": 1.2522841691970825,
      "learning_rate": 8.759493670886077e-06,
      "loss": 0.06695160269737244,
      "num_input_tokens_seen": 2100258,
      "step": 1116,
      "train_runtime": 3353.3985,
      "train_tokens_per_second": 626.307
    },
    {
      "epoch": 1.1283800859236797,
      "grad_norm": 1.22892427444458,
      "learning_rate": 8.749367088607596e-06,
      "loss": 0.05180026963353157,
      "num_input_tokens_seen": 2102152,
      "step": 1117,
      "train_runtime": 3355.6891,
      "train_tokens_per_second": 626.444
    },
    {
      "epoch": 1.1293909527419763,
      "grad_norm": 1.385102391242981,
      "learning_rate": 8.739240506329115e-06,
      "loss": 0.08410146832466125,
      "num_input_tokens_seen": 2104124,
      "step": 1118,
      "train_runtime": 3358.0609,
      "train_tokens_per_second": 626.589
    },
    {
      "epoch": 1.130401819560273,
      "grad_norm": 1.1967195272445679,
      "learning_rate": 8.729113924050634e-06,
      "loss": 0.0709337592124939,
      "num_input_tokens_seen": 2106328,
      "step": 1119,
      "train_runtime": 3360.4864,
      "train_tokens_per_second": 626.793
    },
    {
      "epoch": 1.1314126863785696,
      "grad_norm": 1.071784496307373,
      "learning_rate": 8.718987341772153e-06,
      "loss": 0.05565584823489189,
      "num_input_tokens_seen": 2108402,
      "step": 1120,
      "train_runtime": 3362.8,
      "train_tokens_per_second": 626.978
    },
    {
      "epoch": 1.1324235531968663,
      "grad_norm": 1.5142130851745605,
      "learning_rate": 8.708860759493672e-06,
      "loss": 0.05373769253492355,
      "num_input_tokens_seen": 2110132,
      "step": 1121,
      "train_runtime": 3365.1115,
      "train_tokens_per_second": 627.062
    },
    {
      "epoch": 1.133434420015163,
      "grad_norm": 0.9879475235939026,
      "learning_rate": 8.69873417721519e-06,
      "loss": 0.051886241883039474,
      "num_input_tokens_seen": 2112090,
      "step": 1122,
      "train_runtime": 3367.423,
      "train_tokens_per_second": 627.213
    },
    {
      "epoch": 1.1344452868334596,
      "grad_norm": 1.239451289176941,
      "learning_rate": 8.68860759493671e-06,
      "loss": 0.05598483979701996,
      "num_input_tokens_seen": 2113744,
      "step": 1123,
      "train_runtime": 3369.6472,
      "train_tokens_per_second": 627.289
    },
    {
      "epoch": 1.1354561536517565,
      "grad_norm": 1.5516914129257202,
      "learning_rate": 8.67848101265823e-06,
      "loss": 0.07529913634061813,
      "num_input_tokens_seen": 2115786,
      "step": 1124,
      "train_runtime": 3372.0645,
      "train_tokens_per_second": 627.445
    },
    {
      "epoch": 1.136467020470053,
      "grad_norm": 1.0989845991134644,
      "learning_rate": 8.668354430379747e-06,
      "loss": 0.04982699081301689,
      "num_input_tokens_seen": 2117714,
      "step": 1125,
      "train_runtime": 3374.3774,
      "train_tokens_per_second": 627.587
    },
    {
      "epoch": 1.1374778872883498,
      "grad_norm": 1.6945449113845825,
      "learning_rate": 8.658227848101266e-06,
      "loss": 0.07149287313222885,
      "num_input_tokens_seen": 2119428,
      "step": 1126,
      "train_runtime": 3376.5955,
      "train_tokens_per_second": 627.682
    },
    {
      "epoch": 1.1384887541066464,
      "grad_norm": 1.6831201314926147,
      "learning_rate": 8.648101265822785e-06,
      "loss": 0.07772374153137207,
      "num_input_tokens_seen": 2121072,
      "step": 1127,
      "train_runtime": 3378.842,
      "train_tokens_per_second": 627.751
    },
    {
      "epoch": 1.139499620924943,
      "grad_norm": 1.1811333894729614,
      "learning_rate": 8.637974683544304e-06,
      "loss": 0.0715109333395958,
      "num_input_tokens_seen": 2123588,
      "step": 1128,
      "train_runtime": 3381.2699,
      "train_tokens_per_second": 628.045
    },
    {
      "epoch": 1.14051048774324,
      "grad_norm": 0.9881759881973267,
      "learning_rate": 8.627848101265823e-06,
      "loss": 0.04984269291162491,
      "num_input_tokens_seen": 2125496,
      "step": 1129,
      "train_runtime": 3383.5601,
      "train_tokens_per_second": 628.183
    },
    {
      "epoch": 1.1415213545615366,
      "grad_norm": 1.1169991493225098,
      "learning_rate": 8.617721518987342e-06,
      "loss": 0.047656506299972534,
      "num_input_tokens_seen": 2127294,
      "step": 1130,
      "train_runtime": 3385.8155,
      "train_tokens_per_second": 628.296
    },
    {
      "epoch": 1.1425322213798332,
      "grad_norm": 1.1157288551330566,
      "learning_rate": 8.607594936708861e-06,
      "loss": 0.0789991095662117,
      "num_input_tokens_seen": 2130054,
      "step": 1131,
      "train_runtime": 3388.3242,
      "train_tokens_per_second": 628.645
    },
    {
      "epoch": 1.1435430881981299,
      "grad_norm": 1.2771586179733276,
      "learning_rate": 8.59746835443038e-06,
      "loss": 0.07300856709480286,
      "num_input_tokens_seen": 2131806,
      "step": 1132,
      "train_runtime": 3390.5449,
      "train_tokens_per_second": 628.75
    },
    {
      "epoch": 1.1445539550164265,
      "grad_norm": 1.3346760272979736,
      "learning_rate": 8.5873417721519e-06,
      "loss": 0.052616190165281296,
      "num_input_tokens_seen": 2133332,
      "step": 1133,
      "train_runtime": 3392.7349,
      "train_tokens_per_second": 628.794
    },
    {
      "epoch": 1.1455648218347232,
      "grad_norm": 1.368011713027954,
      "learning_rate": 8.577215189873419e-06,
      "loss": 0.06489899754524231,
      "num_input_tokens_seen": 2134852,
      "step": 1134,
      "train_runtime": 3394.9329,
      "train_tokens_per_second": 628.835
    },
    {
      "epoch": 1.1465756886530198,
      "grad_norm": 1.4414085149765015,
      "learning_rate": 8.567088607594938e-06,
      "loss": 0.0653754249215126,
      "num_input_tokens_seen": 2136526,
      "step": 1135,
      "train_runtime": 3397.1582,
      "train_tokens_per_second": 628.916
    },
    {
      "epoch": 1.1475865554713167,
      "grad_norm": 1.2470828294754028,
      "learning_rate": 8.556962025316457e-06,
      "loss": 0.06085464730858803,
      "num_input_tokens_seen": 2138368,
      "step": 1136,
      "train_runtime": 3399.4507,
      "train_tokens_per_second": 629.033
    },
    {
      "epoch": 1.1485974222896134,
      "grad_norm": 1.3528227806091309,
      "learning_rate": 8.546835443037976e-06,
      "loss": 0.07301013171672821,
      "num_input_tokens_seen": 2140978,
      "step": 1137,
      "train_runtime": 3401.93,
      "train_tokens_per_second": 629.342
    },
    {
      "epoch": 1.14960828910791,
      "grad_norm": 1.1700470447540283,
      "learning_rate": 8.536708860759493e-06,
      "loss": 0.053934965282678604,
      "num_input_tokens_seen": 2143190,
      "step": 1138,
      "train_runtime": 3404.3181,
      "train_tokens_per_second": 629.55
    },
    {
      "epoch": 1.1506191559262067,
      "grad_norm": 1.2100038528442383,
      "learning_rate": 8.526582278481014e-06,
      "loss": 0.041298866271972656,
      "num_input_tokens_seen": 2144560,
      "step": 1139,
      "train_runtime": 3406.4386,
      "train_tokens_per_second": 629.561
    },
    {
      "epoch": 1.1516300227445033,
      "grad_norm": 1.1575100421905518,
      "learning_rate": 8.516455696202533e-06,
      "loss": 0.04897582158446312,
      "num_input_tokens_seen": 2146132,
      "step": 1140,
      "train_runtime": 3408.6236,
      "train_tokens_per_second": 629.618
    },
    {
      "epoch": 1.1526408895628002,
      "grad_norm": 1.4517966508865356,
      "learning_rate": 8.50632911392405e-06,
      "loss": 0.0664825588464737,
      "num_input_tokens_seen": 2148060,
      "step": 1141,
      "train_runtime": 3424.7222,
      "train_tokens_per_second": 627.222
    },
    {
      "epoch": 1.1536517563810968,
      "grad_norm": 1.500447392463684,
      "learning_rate": 8.49620253164557e-06,
      "loss": 0.07931496202945709,
      "num_input_tokens_seen": 2149560,
      "step": 1142,
      "train_runtime": 3426.8771,
      "train_tokens_per_second": 627.265
    },
    {
      "epoch": 1.1546626231993935,
      "grad_norm": 1.2216112613677979,
      "learning_rate": 8.48607594936709e-06,
      "loss": 0.05845990031957626,
      "num_input_tokens_seen": 2151656,
      "step": 1143,
      "train_runtime": 3429.222,
      "train_tokens_per_second": 627.447
    },
    {
      "epoch": 1.1556734900176902,
      "grad_norm": 1.1007955074310303,
      "learning_rate": 8.475949367088608e-06,
      "loss": 0.04873022064566612,
      "num_input_tokens_seen": 2153778,
      "step": 1144,
      "train_runtime": 3431.5552,
      "train_tokens_per_second": 627.639
    },
    {
      "epoch": 1.1566843568359868,
      "grad_norm": 1.0702317953109741,
      "learning_rate": 8.465822784810127e-06,
      "loss": 0.04530014470219612,
      "num_input_tokens_seen": 2156104,
      "step": 1145,
      "train_runtime": 3433.9013,
      "train_tokens_per_second": 627.888
    },
    {
      "epoch": 1.1576952236542835,
      "grad_norm": 1.0356625318527222,
      "learning_rate": 8.455696202531646e-06,
      "loss": 0.048252686858177185,
      "num_input_tokens_seen": 2158052,
      "step": 1146,
      "train_runtime": 3436.1734,
      "train_tokens_per_second": 628.039
    },
    {
      "epoch": 1.1587060904725803,
      "grad_norm": 1.275313377380371,
      "learning_rate": 8.445569620253165e-06,
      "loss": 0.06277644634246826,
      "num_input_tokens_seen": 2159532,
      "step": 1147,
      "train_runtime": 3438.3302,
      "train_tokens_per_second": 628.076
    },
    {
      "epoch": 1.159716957290877,
      "grad_norm": 1.675642967224121,
      "learning_rate": 8.435443037974684e-06,
      "loss": 0.08211557567119598,
      "num_input_tokens_seen": 2161148,
      "step": 1148,
      "train_runtime": 3440.5232,
      "train_tokens_per_second": 628.145
    },
    {
      "epoch": 1.1607278241091736,
      "grad_norm": 1.333276629447937,
      "learning_rate": 8.425316455696203e-06,
      "loss": 0.05804230272769928,
      "num_input_tokens_seen": 2163152,
      "step": 1149,
      "train_runtime": 3442.8351,
      "train_tokens_per_second": 628.305
    },
    {
      "epoch": 1.1617386909274703,
      "grad_norm": 1.1461790800094604,
      "learning_rate": 8.415189873417722e-06,
      "loss": 0.04971778392791748,
      "num_input_tokens_seen": 2165346,
      "step": 1150,
      "train_runtime": 3445.1913,
      "train_tokens_per_second": 628.513
    },
    {
      "epoch": 1.162749557745767,
      "grad_norm": 1.3544596433639526,
      "learning_rate": 8.405063291139241e-06,
      "loss": 0.06081068515777588,
      "num_input_tokens_seen": 2167748,
      "step": 1151,
      "train_runtime": 3447.6256,
      "train_tokens_per_second": 628.765
    },
    {
      "epoch": 1.1637604245640636,
      "grad_norm": 1.0783390998840332,
      "learning_rate": 8.39493670886076e-06,
      "loss": 0.0675191655755043,
      "num_input_tokens_seen": 2169900,
      "step": 1152,
      "train_runtime": 3450.0377,
      "train_tokens_per_second": 628.95
    },
    {
      "epoch": 1.1647712913823605,
      "grad_norm": 1.427892804145813,
      "learning_rate": 8.38481012658228e-06,
      "loss": 0.06211494281888008,
      "num_input_tokens_seen": 2171650,
      "step": 1153,
      "train_runtime": 3452.3415,
      "train_tokens_per_second": 629.037
    },
    {
      "epoch": 1.1657821582006571,
      "grad_norm": 1.1272474527359009,
      "learning_rate": 8.374683544303797e-06,
      "loss": 0.04372329264879227,
      "num_input_tokens_seen": 2173872,
      "step": 1154,
      "train_runtime": 3454.71,
      "train_tokens_per_second": 629.249
    },
    {
      "epoch": 1.1667930250189538,
      "grad_norm": 1.1589261293411255,
      "learning_rate": 8.364556962025318e-06,
      "loss": 0.036889009177684784,
      "num_input_tokens_seen": 2175528,
      "step": 1155,
      "train_runtime": 3456.9587,
      "train_tokens_per_second": 629.318
    },
    {
      "epoch": 1.1678038918372504,
      "grad_norm": 1.3673213720321655,
      "learning_rate": 8.354430379746837e-06,
      "loss": 0.05929093807935715,
      "num_input_tokens_seen": 2177430,
      "step": 1156,
      "train_runtime": 3459.2146,
      "train_tokens_per_second": 629.458
    },
    {
      "epoch": 1.168814758655547,
      "grad_norm": 1.4972892999649048,
      "learning_rate": 8.344303797468354e-06,
      "loss": 0.052913933992385864,
      "num_input_tokens_seen": 2178862,
      "step": 1157,
      "train_runtime": 3461.3917,
      "train_tokens_per_second": 629.476
    },
    {
      "epoch": 1.1698256254738437,
      "grad_norm": 1.1993151903152466,
      "learning_rate": 8.334177215189873e-06,
      "loss": 0.05444449186325073,
      "num_input_tokens_seen": 2180700,
      "step": 1158,
      "train_runtime": 3463.6655,
      "train_tokens_per_second": 629.593
    },
    {
      "epoch": 1.1708364922921406,
      "grad_norm": 1.5625636577606201,
      "learning_rate": 8.324050632911394e-06,
      "loss": 0.07694568485021591,
      "num_input_tokens_seen": 2182790,
      "step": 1159,
      "train_runtime": 3465.9798,
      "train_tokens_per_second": 629.776
    },
    {
      "epoch": 1.1718473591104372,
      "grad_norm": 1.707200050354004,
      "learning_rate": 8.313924050632911e-06,
      "loss": 0.10209743678569794,
      "num_input_tokens_seen": 2184616,
      "step": 1160,
      "train_runtime": 3468.2923,
      "train_tokens_per_second": 629.882
    },
    {
      "epoch": 1.172858225928734,
      "grad_norm": 1.5268076658248901,
      "learning_rate": 8.30379746835443e-06,
      "loss": 0.06384491175413132,
      "num_input_tokens_seen": 2186642,
      "step": 1161,
      "train_runtime": 3470.6357,
      "train_tokens_per_second": 630.041
    },
    {
      "epoch": 1.1738690927470306,
      "grad_norm": 1.423162579536438,
      "learning_rate": 8.29367088607595e-06,
      "loss": 0.0545058473944664,
      "num_input_tokens_seen": 2188852,
      "step": 1162,
      "train_runtime": 3473.0832,
      "train_tokens_per_second": 630.233
    },
    {
      "epoch": 1.1748799595653272,
      "grad_norm": 1.1817657947540283,
      "learning_rate": 8.283544303797469e-06,
      "loss": 0.06416331231594086,
      "num_input_tokens_seen": 2190730,
      "step": 1163,
      "train_runtime": 3475.4334,
      "train_tokens_per_second": 630.347
    },
    {
      "epoch": 1.175890826383624,
      "grad_norm": 1.054011583328247,
      "learning_rate": 8.273417721518988e-06,
      "loss": 0.03909891098737717,
      "num_input_tokens_seen": 2192420,
      "step": 1164,
      "train_runtime": 3477.6456,
      "train_tokens_per_second": 630.432
    },
    {
      "epoch": 1.1769016932019207,
      "grad_norm": 1.0762697458267212,
      "learning_rate": 8.263291139240507e-06,
      "loss": 0.03878183290362358,
      "num_input_tokens_seen": 2194292,
      "step": 1165,
      "train_runtime": 3479.9587,
      "train_tokens_per_second": 630.551
    },
    {
      "epoch": 1.1779125600202174,
      "grad_norm": 0.9733167290687561,
      "learning_rate": 8.253164556962026e-06,
      "loss": 0.03576451539993286,
      "num_input_tokens_seen": 2196356,
      "step": 1166,
      "train_runtime": 3482.3145,
      "train_tokens_per_second": 630.717
    },
    {
      "epoch": 1.178923426838514,
      "grad_norm": 1.3922288417816162,
      "learning_rate": 8.243037974683545e-06,
      "loss": 0.07775457203388214,
      "num_input_tokens_seen": 2198536,
      "step": 1167,
      "train_runtime": 3484.6529,
      "train_tokens_per_second": 630.92
    },
    {
      "epoch": 1.1799342936568107,
      "grad_norm": 1.4985376596450806,
      "learning_rate": 8.232911392405064e-06,
      "loss": 0.06136911362409592,
      "num_input_tokens_seen": 2200490,
      "step": 1168,
      "train_runtime": 3486.9543,
      "train_tokens_per_second": 631.064
    },
    {
      "epoch": 1.1809451604751073,
      "grad_norm": 1.588498592376709,
      "learning_rate": 8.222784810126583e-06,
      "loss": 0.05329131707549095,
      "num_input_tokens_seen": 2202402,
      "step": 1169,
      "train_runtime": 3489.2492,
      "train_tokens_per_second": 631.197
    },
    {
      "epoch": 1.181956027293404,
      "grad_norm": 1.248744010925293,
      "learning_rate": 8.212658227848102e-06,
      "loss": 0.043204858899116516,
      "num_input_tokens_seen": 2204174,
      "step": 1170,
      "train_runtime": 3491.4847,
      "train_tokens_per_second": 631.3
    },
    {
      "epoch": 1.1829668941117009,
      "grad_norm": 1.2362407445907593,
      "learning_rate": 8.202531645569621e-06,
      "loss": 0.055201996117830276,
      "num_input_tokens_seen": 2206006,
      "step": 1171,
      "train_runtime": 3508.0583,
      "train_tokens_per_second": 628.84
    },
    {
      "epoch": 1.1839777609299975,
      "grad_norm": 1.1224781274795532,
      "learning_rate": 8.19240506329114e-06,
      "loss": 0.05462568625807762,
      "num_input_tokens_seen": 2207678,
      "step": 1172,
      "train_runtime": 3510.2786,
      "train_tokens_per_second": 628.918
    },
    {
      "epoch": 1.1849886277482942,
      "grad_norm": 1.4427181482315063,
      "learning_rate": 8.18227848101266e-06,
      "loss": 0.055276695638895035,
      "num_input_tokens_seen": 2209704,
      "step": 1173,
      "train_runtime": 3512.5815,
      "train_tokens_per_second": 629.083
    },
    {
      "epoch": 1.1859994945665908,
      "grad_norm": 1.1946271657943726,
      "learning_rate": 8.172151898734177e-06,
      "loss": 0.05866575613617897,
      "num_input_tokens_seen": 2211704,
      "step": 1174,
      "train_runtime": 3514.9645,
      "train_tokens_per_second": 629.225
    },
    {
      "epoch": 1.1870103613848875,
      "grad_norm": 1.156442642211914,
      "learning_rate": 8.162025316455698e-06,
      "loss": 0.059589408338069916,
      "num_input_tokens_seen": 2213624,
      "step": 1175,
      "train_runtime": 3517.2364,
      "train_tokens_per_second": 629.365
    },
    {
      "epoch": 1.1880212282031843,
      "grad_norm": 1.233392596244812,
      "learning_rate": 8.151898734177217e-06,
      "loss": 0.06653599441051483,
      "num_input_tokens_seen": 2215790,
      "step": 1176,
      "train_runtime": 3519.5687,
      "train_tokens_per_second": 629.563
    },
    {
      "epoch": 1.189032095021481,
      "grad_norm": 1.1985082626342773,
      "learning_rate": 8.141772151898734e-06,
      "loss": 0.06252250075340271,
      "num_input_tokens_seen": 2217462,
      "step": 1177,
      "train_runtime": 3521.7911,
      "train_tokens_per_second": 629.64
    },
    {
      "epoch": 1.1900429618397776,
      "grad_norm": 1.437553882598877,
      "learning_rate": 8.131645569620253e-06,
      "loss": 0.07226981222629547,
      "num_input_tokens_seen": 2219292,
      "step": 1178,
      "train_runtime": 3524.0288,
      "train_tokens_per_second": 629.76
    },
    {
      "epoch": 1.1910538286580743,
      "grad_norm": 1.4563100337982178,
      "learning_rate": 8.121518987341774e-06,
      "loss": 0.0649099051952362,
      "num_input_tokens_seen": 2221190,
      "step": 1179,
      "train_runtime": 3526.3067,
      "train_tokens_per_second": 629.891
    },
    {
      "epoch": 1.192064695476371,
      "grad_norm": 1.53696608543396,
      "learning_rate": 8.111392405063291e-06,
      "loss": 0.08135532587766647,
      "num_input_tokens_seen": 2223196,
      "step": 1180,
      "train_runtime": 3528.6214,
      "train_tokens_per_second": 630.047
    },
    {
      "epoch": 1.1930755622946676,
      "grad_norm": 1.2576117515563965,
      "learning_rate": 8.10126582278481e-06,
      "loss": 0.055172935128211975,
      "num_input_tokens_seen": 2224970,
      "step": 1181,
      "train_runtime": 3530.8656,
      "train_tokens_per_second": 630.149
    },
    {
      "epoch": 1.1940864291129643,
      "grad_norm": 1.5804694890975952,
      "learning_rate": 8.09113924050633e-06,
      "loss": 0.07707778364419937,
      "num_input_tokens_seen": 2227122,
      "step": 1182,
      "train_runtime": 3533.176,
      "train_tokens_per_second": 630.346
    },
    {
      "epoch": 1.1950972959312611,
      "grad_norm": 1.2614355087280273,
      "learning_rate": 8.081012658227849e-06,
      "loss": 0.042505331337451935,
      "num_input_tokens_seen": 2229118,
      "step": 1183,
      "train_runtime": 3535.4597,
      "train_tokens_per_second": 630.503
    },
    {
      "epoch": 1.1961081627495578,
      "grad_norm": 1.6972736120224,
      "learning_rate": 8.070886075949368e-06,
      "loss": 0.05081099644303322,
      "num_input_tokens_seen": 2230492,
      "step": 1184,
      "train_runtime": 3537.5771,
      "train_tokens_per_second": 630.514
    },
    {
      "epoch": 1.1971190295678544,
      "grad_norm": 1.3364206552505493,
      "learning_rate": 8.060759493670887e-06,
      "loss": 0.05734006687998772,
      "num_input_tokens_seen": 2232598,
      "step": 1185,
      "train_runtime": 3539.9292,
      "train_tokens_per_second": 630.69
    },
    {
      "epoch": 1.198129896386151,
      "grad_norm": 1.1175353527069092,
      "learning_rate": 8.050632911392406e-06,
      "loss": 0.042109165340662,
      "num_input_tokens_seen": 2234364,
      "step": 1186,
      "train_runtime": 3542.1742,
      "train_tokens_per_second": 630.789
    },
    {
      "epoch": 1.1991407632044477,
      "grad_norm": 1.4004695415496826,
      "learning_rate": 8.040506329113925e-06,
      "loss": 0.07079403102397919,
      "num_input_tokens_seen": 2236454,
      "step": 1187,
      "train_runtime": 3544.5006,
      "train_tokens_per_second": 630.964
    },
    {
      "epoch": 1.2001516300227446,
      "grad_norm": 1.8970848321914673,
      "learning_rate": 8.030379746835444e-06,
      "loss": 0.07711949944496155,
      "num_input_tokens_seen": 2237934,
      "step": 1188,
      "train_runtime": 3546.6795,
      "train_tokens_per_second": 630.994
    },
    {
      "epoch": 1.2001516300227446,
      "eval_loss": 0.3687424063682556,
      "eval_runtime": 54.3625,
      "eval_samples_per_second": 16.169,
      "eval_steps_per_second": 8.094,
      "num_input_tokens_seen": 2237934,
      "step": 1188
    },
    {
      "epoch": 1.2011624968410413,
      "grad_norm": 1.3333828449249268,
      "learning_rate": 8.020253164556963e-06,
      "loss": 0.05751827731728554,
      "num_input_tokens_seen": 2239668,
      "step": 1189,
      "train_runtime": 3603.3172,
      "train_tokens_per_second": 621.557
    },
    {
      "epoch": 1.202173363659338,
      "grad_norm": 0.967258632183075,
      "learning_rate": 8.01012658227848e-06,
      "loss": 0.04930635541677475,
      "num_input_tokens_seen": 2241998,
      "step": 1190,
      "train_runtime": 3605.7206,
      "train_tokens_per_second": 621.789
    },
    {
      "epoch": 1.2031842304776346,
      "grad_norm": 1.7901451587677002,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.06181413680315018,
      "num_input_tokens_seen": 2243886,
      "step": 1191,
      "train_runtime": 3607.9935,
      "train_tokens_per_second": 621.921
    },
    {
      "epoch": 1.2041950972959312,
      "grad_norm": 1.2481358051300049,
      "learning_rate": 7.98987341772152e-06,
      "loss": 0.048134684562683105,
      "num_input_tokens_seen": 2245630,
      "step": 1192,
      "train_runtime": 3610.2343,
      "train_tokens_per_second": 622.018
    },
    {
      "epoch": 1.2052059641142279,
      "grad_norm": 1.5724238157272339,
      "learning_rate": 7.979746835443038e-06,
      "loss": 0.07093324512243271,
      "num_input_tokens_seen": 2247608,
      "step": 1193,
      "train_runtime": 3612.546,
      "train_tokens_per_second": 622.167
    },
    {
      "epoch": 1.2062168309325245,
      "grad_norm": 1.4698814153671265,
      "learning_rate": 7.969620253164557e-06,
      "loss": 0.04218070209026337,
      "num_input_tokens_seen": 2249956,
      "step": 1194,
      "train_runtime": 3614.9292,
      "train_tokens_per_second": 622.407
    },
    {
      "epoch": 1.2072276977508214,
      "grad_norm": 1.8323177099227905,
      "learning_rate": 7.959493670886078e-06,
      "loss": 0.0682552233338356,
      "num_input_tokens_seen": 2251722,
      "step": 1195,
      "train_runtime": 3617.1781,
      "train_tokens_per_second": 622.508
    },
    {
      "epoch": 1.208238564569118,
      "grad_norm": 1.7569893598556519,
      "learning_rate": 7.949367088607595e-06,
      "loss": 0.07162746787071228,
      "num_input_tokens_seen": 2253246,
      "step": 1196,
      "train_runtime": 3619.3612,
      "train_tokens_per_second": 622.554
    },
    {
      "epoch": 1.2092494313874147,
      "grad_norm": 1.1859911680221558,
      "learning_rate": 7.939240506329114e-06,
      "loss": 0.04805188626050949,
      "num_input_tokens_seen": 2254870,
      "step": 1197,
      "train_runtime": 3621.5788,
      "train_tokens_per_second": 622.621
    },
    {
      "epoch": 1.2102602982057113,
      "grad_norm": 1.1672230958938599,
      "learning_rate": 7.929113924050633e-06,
      "loss": 0.047470320016145706,
      "num_input_tokens_seen": 2256480,
      "step": 1198,
      "train_runtime": 3623.8601,
      "train_tokens_per_second": 622.673
    },
    {
      "epoch": 1.211271165024008,
      "grad_norm": 1.3061758279800415,
      "learning_rate": 7.918987341772152e-06,
      "loss": 0.05248216167092323,
      "num_input_tokens_seen": 2257978,
      "step": 1199,
      "train_runtime": 3626.0217,
      "train_tokens_per_second": 622.715
    },
    {
      "epoch": 1.2122820318423049,
      "grad_norm": 1.6017221212387085,
      "learning_rate": 7.908860759493671e-06,
      "loss": 0.072176493704319,
      "num_input_tokens_seen": 2259402,
      "step": 1200,
      "train_runtime": 3628.1755,
      "train_tokens_per_second": 622.738
    },
    {
      "epoch": 1.2132928986606015,
      "grad_norm": 1.2543563842773438,
      "learning_rate": 7.89873417721519e-06,
      "loss": 0.050776977092027664,
      "num_input_tokens_seen": 2260964,
      "step": 1201,
      "train_runtime": 3644.6274,
      "train_tokens_per_second": 620.355
    },
    {
      "epoch": 1.2143037654788982,
      "grad_norm": 0.9192790389060974,
      "learning_rate": 7.88860759493671e-06,
      "loss": 0.03936336934566498,
      "num_input_tokens_seen": 2262478,
      "step": 1202,
      "train_runtime": 3646.8141,
      "train_tokens_per_second": 620.399
    },
    {
      "epoch": 1.2153146322971948,
      "grad_norm": 1.234483242034912,
      "learning_rate": 7.878481012658229e-06,
      "loss": 0.0455319881439209,
      "num_input_tokens_seen": 2263914,
      "step": 1203,
      "train_runtime": 3649.1016,
      "train_tokens_per_second": 620.403
    },
    {
      "epoch": 1.2163254991154915,
      "grad_norm": 1.1697463989257812,
      "learning_rate": 7.868354430379748e-06,
      "loss": 0.055918656289577484,
      "num_input_tokens_seen": 2266010,
      "step": 1204,
      "train_runtime": 3651.4805,
      "train_tokens_per_second": 620.573
    },
    {
      "epoch": 1.2173363659337881,
      "grad_norm": 1.7618815898895264,
      "learning_rate": 7.858227848101267e-06,
      "loss": 0.06636051088571548,
      "num_input_tokens_seen": 2267862,
      "step": 1205,
      "train_runtime": 3653.7851,
      "train_tokens_per_second": 620.688
    },
    {
      "epoch": 1.218347232752085,
      "grad_norm": 1.242294192314148,
      "learning_rate": 7.848101265822786e-06,
      "loss": 0.04812423139810562,
      "num_input_tokens_seen": 2269564,
      "step": 1206,
      "train_runtime": 3656.0364,
      "train_tokens_per_second": 620.772
    },
    {
      "epoch": 1.2193580995703817,
      "grad_norm": 1.1910755634307861,
      "learning_rate": 7.837974683544305e-06,
      "loss": 0.05991936847567558,
      "num_input_tokens_seen": 2272080,
      "step": 1207,
      "train_runtime": 3658.4764,
      "train_tokens_per_second": 621.045
    },
    {
      "epoch": 1.2203689663886783,
      "grad_norm": 1.2928847074508667,
      "learning_rate": 7.827848101265824e-06,
      "loss": 0.052168942987918854,
      "num_input_tokens_seen": 2273334,
      "step": 1208,
      "train_runtime": 3660.5719,
      "train_tokens_per_second": 621.032
    },
    {
      "epoch": 1.221379833206975,
      "grad_norm": 1.2783948183059692,
      "learning_rate": 7.817721518987343e-06,
      "loss": 0.06510590016841888,
      "num_input_tokens_seen": 2275454,
      "step": 1209,
      "train_runtime": 3662.8793,
      "train_tokens_per_second": 621.22
    },
    {
      "epoch": 1.2223907000252716,
      "grad_norm": 1.7674273252487183,
      "learning_rate": 7.80759493670886e-06,
      "loss": 0.04185160994529724,
      "num_input_tokens_seen": 2277052,
      "step": 1210,
      "train_runtime": 3665.126,
      "train_tokens_per_second": 621.275
    },
    {
      "epoch": 1.2234015668435683,
      "grad_norm": 1.0900136232376099,
      "learning_rate": 7.797468354430381e-06,
      "loss": 0.035728149116039276,
      "num_input_tokens_seen": 2278706,
      "step": 1211,
      "train_runtime": 3667.343,
      "train_tokens_per_second": 621.351
    },
    {
      "epoch": 1.2244124336618651,
      "grad_norm": 1.7899435758590698,
      "learning_rate": 7.7873417721519e-06,
      "loss": 0.06428918242454529,
      "num_input_tokens_seen": 2280384,
      "step": 1212,
      "train_runtime": 3669.57,
      "train_tokens_per_second": 621.431
    },
    {
      "epoch": 1.2254233004801618,
      "grad_norm": 1.4243944883346558,
      "learning_rate": 7.777215189873418e-06,
      "loss": 0.06730536371469498,
      "num_input_tokens_seen": 2282836,
      "step": 1213,
      "train_runtime": 3671.9922,
      "train_tokens_per_second": 621.689
    },
    {
      "epoch": 1.2264341672984584,
      "grad_norm": 1.6462485790252686,
      "learning_rate": 7.767088607594937e-06,
      "loss": 0.057023413479328156,
      "num_input_tokens_seen": 2284726,
      "step": 1214,
      "train_runtime": 3674.2725,
      "train_tokens_per_second": 621.817
    },
    {
      "epoch": 1.227445034116755,
      "grad_norm": 1.2234854698181152,
      "learning_rate": 7.756962025316456e-06,
      "loss": 0.06149659305810928,
      "num_input_tokens_seen": 2287752,
      "step": 1215,
      "train_runtime": 3676.8565,
      "train_tokens_per_second": 622.203
    },
    {
      "epoch": 1.2284559009350517,
      "grad_norm": 1.159865379333496,
      "learning_rate": 7.746835443037975e-06,
      "loss": 0.04254642128944397,
      "num_input_tokens_seen": 2289266,
      "step": 1216,
      "train_runtime": 3679.0534,
      "train_tokens_per_second": 622.243
    },
    {
      "epoch": 1.2294667677533484,
      "grad_norm": 1.3408385515213013,
      "learning_rate": 7.736708860759494e-06,
      "loss": 0.053428713232278824,
      "num_input_tokens_seen": 2291274,
      "step": 1217,
      "train_runtime": 3681.3653,
      "train_tokens_per_second": 622.398
    },
    {
      "epoch": 1.2304776345716453,
      "grad_norm": 1.2482548952102661,
      "learning_rate": 7.726582278481013e-06,
      "loss": 0.05629862844944,
      "num_input_tokens_seen": 2293442,
      "step": 1218,
      "train_runtime": 3683.6741,
      "train_tokens_per_second": 622.596
    },
    {
      "epoch": 1.231488501389942,
      "grad_norm": 1.045045018196106,
      "learning_rate": 7.716455696202532e-06,
      "loss": 0.05139722675085068,
      "num_input_tokens_seen": 2295438,
      "step": 1219,
      "train_runtime": 3685.9778,
      "train_tokens_per_second": 622.749
    },
    {
      "epoch": 1.2324993682082386,
      "grad_norm": 1.2410844564437866,
      "learning_rate": 7.706329113924051e-06,
      "loss": 0.04247979819774628,
      "num_input_tokens_seen": 2297814,
      "step": 1220,
      "train_runtime": 3688.4051,
      "train_tokens_per_second": 622.983
    },
    {
      "epoch": 1.2335102350265352,
      "grad_norm": 1.216529130935669,
      "learning_rate": 7.69620253164557e-06,
      "loss": 0.049616873264312744,
      "num_input_tokens_seen": 2299686,
      "step": 1221,
      "train_runtime": 3690.6467,
      "train_tokens_per_second": 623.112
    },
    {
      "epoch": 1.2345211018448319,
      "grad_norm": 1.1823710203170776,
      "learning_rate": 7.68607594936709e-06,
      "loss": 0.04045352339744568,
      "num_input_tokens_seen": 2301578,
      "step": 1222,
      "train_runtime": 3692.8889,
      "train_tokens_per_second": 623.246
    },
    {
      "epoch": 1.2355319686631288,
      "grad_norm": 1.2898205518722534,
      "learning_rate": 7.675949367088609e-06,
      "loss": 0.05988980084657669,
      "num_input_tokens_seen": 2303610,
      "step": 1223,
      "train_runtime": 3695.1624,
      "train_tokens_per_second": 623.412
    },
    {
      "epoch": 1.2365428354814254,
      "grad_norm": 1.1859538555145264,
      "learning_rate": 7.665822784810128e-06,
      "loss": 0.048559948801994324,
      "num_input_tokens_seen": 2305512,
      "step": 1224,
      "train_runtime": 3697.4557,
      "train_tokens_per_second": 623.54
    },
    {
      "epoch": 1.237553702299722,
      "grad_norm": 1.8632739782333374,
      "learning_rate": 7.655696202531647e-06,
      "loss": 0.05580410733819008,
      "num_input_tokens_seen": 2307674,
      "step": 1225,
      "train_runtime": 3699.8035,
      "train_tokens_per_second": 623.729
    },
    {
      "epoch": 1.2385645691180187,
      "grad_norm": 1.4416550397872925,
      "learning_rate": 7.645569620253164e-06,
      "loss": 0.0591302216053009,
      "num_input_tokens_seen": 2309670,
      "step": 1226,
      "train_runtime": 3702.0836,
      "train_tokens_per_second": 623.884
    },
    {
      "epoch": 1.2395754359363154,
      "grad_norm": 1.3483821153640747,
      "learning_rate": 7.635443037974685e-06,
      "loss": 0.044712476432323456,
      "num_input_tokens_seen": 2311322,
      "step": 1227,
      "train_runtime": 3704.3366,
      "train_tokens_per_second": 623.95
    },
    {
      "epoch": 1.240586302754612,
      "grad_norm": 1.6321603059768677,
      "learning_rate": 7.625316455696203e-06,
      "loss": 0.04973188787698746,
      "num_input_tokens_seen": 2312726,
      "step": 1228,
      "train_runtime": 3706.5158,
      "train_tokens_per_second": 623.962
    },
    {
      "epoch": 1.2415971695729087,
      "grad_norm": 1.1471374034881592,
      "learning_rate": 7.615189873417722e-06,
      "loss": 0.05518278107047081,
      "num_input_tokens_seen": 2314644,
      "step": 1229,
      "train_runtime": 3708.8157,
      "train_tokens_per_second": 624.092
    },
    {
      "epoch": 1.2426080363912055,
      "grad_norm": 1.3072782754898071,
      "learning_rate": 7.6050632911392405e-06,
      "loss": 0.0519881397485733,
      "num_input_tokens_seen": 2316536,
      "step": 1230,
      "train_runtime": 3711.091,
      "train_tokens_per_second": 624.22
    },
    {
      "epoch": 1.2436189032095022,
      "grad_norm": 1.395263910293579,
      "learning_rate": 7.5949367088607605e-06,
      "loss": 0.06062118709087372,
      "num_input_tokens_seen": 2318024,
      "step": 1231,
      "train_runtime": 3726.8768,
      "train_tokens_per_second": 621.975
    },
    {
      "epoch": 1.2446297700277988,
      "grad_norm": 1.3331067562103271,
      "learning_rate": 7.5848101265822796e-06,
      "loss": 0.07556059211492538,
      "num_input_tokens_seen": 2320428,
      "step": 1232,
      "train_runtime": 3729.3148,
      "train_tokens_per_second": 622.213
    },
    {
      "epoch": 1.2456406368460955,
      "grad_norm": 1.5281811952590942,
      "learning_rate": 7.574683544303798e-06,
      "loss": 0.06573790311813354,
      "num_input_tokens_seen": 2322182,
      "step": 1233,
      "train_runtime": 3731.5742,
      "train_tokens_per_second": 622.306
    },
    {
      "epoch": 1.2466515036643921,
      "grad_norm": 1.394310474395752,
      "learning_rate": 7.564556962025317e-06,
      "loss": 0.06115046888589859,
      "num_input_tokens_seen": 2323616,
      "step": 1234,
      "train_runtime": 3733.765,
      "train_tokens_per_second": 622.325
    },
    {
      "epoch": 1.247662370482689,
      "grad_norm": 1.5008997917175293,
      "learning_rate": 7.554430379746837e-06,
      "loss": 0.07988383620977402,
      "num_input_tokens_seen": 2325408,
      "step": 1235,
      "train_runtime": 3736.0465,
      "train_tokens_per_second": 622.425
    },
    {
      "epoch": 1.2486732373009857,
      "grad_norm": 0.9623056054115295,
      "learning_rate": 7.544303797468355e-06,
      "loss": 0.045632608234882355,
      "num_input_tokens_seen": 2327858,
      "step": 1236,
      "train_runtime": 3738.5179,
      "train_tokens_per_second": 622.669
    },
    {
      "epoch": 1.2496841041192823,
      "grad_norm": 1.1915478706359863,
      "learning_rate": 7.534177215189874e-06,
      "loss": 0.043249502778053284,
      "num_input_tokens_seen": 2329798,
      "step": 1237,
      "train_runtime": 3740.7932,
      "train_tokens_per_second": 622.809
    },
    {
      "epoch": 1.250694970937579,
      "grad_norm": 1.0730327367782593,
      "learning_rate": 7.524050632911392e-06,
      "loss": 0.05751641094684601,
      "num_input_tokens_seen": 2332188,
      "step": 1238,
      "train_runtime": 3743.2875,
      "train_tokens_per_second": 623.032
    },
    {
      "epoch": 1.2517058377558756,
      "grad_norm": 1.4013127088546753,
      "learning_rate": 7.513924050632912e-06,
      "loss": 0.0634988471865654,
      "num_input_tokens_seen": 2334310,
      "step": 1239,
      "train_runtime": 3745.642,
      "train_tokens_per_second": 623.207
    },
    {
      "epoch": 1.2527167045741723,
      "grad_norm": 1.3232216835021973,
      "learning_rate": 7.503797468354431e-06,
      "loss": 0.0602821409702301,
      "num_input_tokens_seen": 2336354,
      "step": 1240,
      "train_runtime": 3747.9665,
      "train_tokens_per_second": 623.366
    },
    {
      "epoch": 1.253727571392469,
      "grad_norm": 1.1508218050003052,
      "learning_rate": 7.49367088607595e-06,
      "loss": 0.04428628832101822,
      "num_input_tokens_seen": 2338472,
      "step": 1241,
      "train_runtime": 3750.3185,
      "train_tokens_per_second": 623.54
    },
    {
      "epoch": 1.2547384382107658,
      "grad_norm": 1.5455573797225952,
      "learning_rate": 7.483544303797469e-06,
      "loss": 0.06668779253959656,
      "num_input_tokens_seen": 2340730,
      "step": 1242,
      "train_runtime": 3752.9886,
      "train_tokens_per_second": 623.698
    },
    {
      "epoch": 1.2557493050290625,
      "grad_norm": 1.5112093687057495,
      "learning_rate": 7.473417721518989e-06,
      "loss": 0.05614790692925453,
      "num_input_tokens_seen": 2342948,
      "step": 1243,
      "train_runtime": 3755.3258,
      "train_tokens_per_second": 623.9
    },
    {
      "epoch": 1.256760171847359,
      "grad_norm": 1.386748194694519,
      "learning_rate": 7.463291139240507e-06,
      "loss": 0.05175211653113365,
      "num_input_tokens_seen": 2345132,
      "step": 1244,
      "train_runtime": 3757.7003,
      "train_tokens_per_second": 624.087
    },
    {
      "epoch": 1.2577710386656558,
      "grad_norm": 1.5524296760559082,
      "learning_rate": 7.453164556962026e-06,
      "loss": 0.06800000369548798,
      "num_input_tokens_seen": 2346772,
      "step": 1245,
      "train_runtime": 3759.9756,
      "train_tokens_per_second": 624.146
    },
    {
      "epoch": 1.2587819054839524,
      "grad_norm": 1.4179092645645142,
      "learning_rate": 7.443037974683544e-06,
      "loss": 0.06086236983537674,
      "num_input_tokens_seen": 2348770,
      "step": 1246,
      "train_runtime": 3762.3294,
      "train_tokens_per_second": 624.286
    },
    {
      "epoch": 1.2597927723022493,
      "grad_norm": 1.2474757432937622,
      "learning_rate": 7.432911392405064e-06,
      "loss": 0.06393437832593918,
      "num_input_tokens_seen": 2350794,
      "step": 1247,
      "train_runtime": 3764.666,
      "train_tokens_per_second": 624.436
    },
    {
      "epoch": 1.260803639120546,
      "grad_norm": 1.182120680809021,
      "learning_rate": 7.422784810126583e-06,
      "loss": 0.054492250084877014,
      "num_input_tokens_seen": 2352492,
      "step": 1248,
      "train_runtime": 3766.8863,
      "train_tokens_per_second": 624.519
    },
    {
      "epoch": 1.2618145059388426,
      "grad_norm": 1.5892659425735474,
      "learning_rate": 7.4126582278481014e-06,
      "loss": 0.08521469682455063,
      "num_input_tokens_seen": 2354176,
      "step": 1249,
      "train_runtime": 3769.1181,
      "train_tokens_per_second": 624.596
    },
    {
      "epoch": 1.2628253727571392,
      "grad_norm": 1.1960848569869995,
      "learning_rate": 7.4025316455696205e-06,
      "loss": 0.055863481014966965,
      "num_input_tokens_seen": 2356116,
      "step": 1250,
      "train_runtime": 3771.467,
      "train_tokens_per_second": 624.721
    },
    {
      "epoch": 1.263836239575436,
      "grad_norm": 2.125699520111084,
      "learning_rate": 7.3924050632911405e-06,
      "loss": 0.07721327245235443,
      "num_input_tokens_seen": 2357662,
      "step": 1251,
      "train_runtime": 3773.7573,
      "train_tokens_per_second": 624.752
    },
    {
      "epoch": 1.2648471063937325,
      "grad_norm": 1.4590431451797485,
      "learning_rate": 7.382278481012659e-06,
      "loss": 0.052303265780210495,
      "num_input_tokens_seen": 2359266,
      "step": 1252,
      "train_runtime": 3775.9615,
      "train_tokens_per_second": 624.812
    },
    {
      "epoch": 1.2658579732120292,
      "grad_norm": 1.2073882818222046,
      "learning_rate": 7.372151898734178e-06,
      "loss": 0.04308575391769409,
      "num_input_tokens_seen": 2361054,
      "step": 1253,
      "train_runtime": 3778.2318,
      "train_tokens_per_second": 624.91
    },
    {
      "epoch": 1.266868840030326,
      "grad_norm": 1.4119540452957153,
      "learning_rate": 7.362025316455697e-06,
      "loss": 0.05672318488359451,
      "num_input_tokens_seen": 2362734,
      "step": 1254,
      "train_runtime": 3780.4321,
      "train_tokens_per_second": 624.99
    },
    {
      "epoch": 1.2678797068486227,
      "grad_norm": 1.2578742504119873,
      "learning_rate": 7.351898734177216e-06,
      "loss": 0.06953655183315277,
      "num_input_tokens_seen": 2364500,
      "step": 1255,
      "train_runtime": 3782.6941,
      "train_tokens_per_second": 625.084
    },
    {
      "epoch": 1.2688905736669194,
      "grad_norm": 1.3448150157928467,
      "learning_rate": 7.341772151898735e-06,
      "loss": 0.04783962666988373,
      "num_input_tokens_seen": 2366698,
      "step": 1256,
      "train_runtime": 3785.0436,
      "train_tokens_per_second": 625.276
    },
    {
      "epoch": 1.269901440485216,
      "grad_norm": 1.3297990560531616,
      "learning_rate": 7.331645569620254e-06,
      "loss": 0.050486110150814056,
      "num_input_tokens_seen": 2368722,
      "step": 1257,
      "train_runtime": 3787.3476,
      "train_tokens_per_second": 625.43
    },
    {
      "epoch": 1.270912307303513,
      "grad_norm": 1.0803709030151367,
      "learning_rate": 7.321518987341772e-06,
      "loss": 0.05573941767215729,
      "num_input_tokens_seen": 2370806,
      "step": 1258,
      "train_runtime": 3789.6801,
      "train_tokens_per_second": 625.595
    },
    {
      "epoch": 1.2719231741218096,
      "grad_norm": 1.1125187873840332,
      "learning_rate": 7.311392405063292e-06,
      "loss": 0.04718473553657532,
      "num_input_tokens_seen": 2372522,
      "step": 1259,
      "train_runtime": 3791.9679,
      "train_tokens_per_second": 625.67
    },
    {
      "epoch": 1.2729340409401062,
      "grad_norm": 1.0505123138427734,
      "learning_rate": 7.301265822784811e-06,
      "loss": 0.04578212648630142,
      "num_input_tokens_seen": 2374258,
      "step": 1260,
      "train_runtime": 3794.2543,
      "train_tokens_per_second": 625.751
    },
    {
      "epoch": 1.2739449077584029,
      "grad_norm": 1.0820823907852173,
      "learning_rate": 7.29113924050633e-06,
      "loss": 0.042990993708372116,
      "num_input_tokens_seen": 2375890,
      "step": 1261,
      "train_runtime": 3809.8469,
      "train_tokens_per_second": 623.618
    },
    {
      "epoch": 1.2749557745766995,
      "grad_norm": 0.9400637745857239,
      "learning_rate": 7.281012658227849e-06,
      "loss": 0.03949103131890297,
      "num_input_tokens_seen": 2377974,
      "step": 1262,
      "train_runtime": 3812.2122,
      "train_tokens_per_second": 623.778
    },
    {
      "epoch": 1.2759666413949962,
      "grad_norm": 1.3851995468139648,
      "learning_rate": 7.270886075949367e-06,
      "loss": 0.05658121407032013,
      "num_input_tokens_seen": 2379908,
      "step": 1263,
      "train_runtime": 3814.5151,
      "train_tokens_per_second": 623.908
    },
    {
      "epoch": 1.2769775082132928,
      "grad_norm": 1.4746298789978027,
      "learning_rate": 7.260759493670887e-06,
      "loss": 0.07203026115894318,
      "num_input_tokens_seen": 2382120,
      "step": 1264,
      "train_runtime": 3816.8503,
      "train_tokens_per_second": 624.106
    },
    {
      "epoch": 1.2779883750315895,
      "grad_norm": 1.1128922700881958,
      "learning_rate": 7.250632911392406e-06,
      "loss": 0.039608296006917953,
      "num_input_tokens_seen": 2383738,
      "step": 1265,
      "train_runtime": 3819.0324,
      "train_tokens_per_second": 624.173
    },
    {
      "epoch": 1.2789992418498863,
      "grad_norm": 1.2160228490829468,
      "learning_rate": 7.240506329113924e-06,
      "loss": 0.05472289398312569,
      "num_input_tokens_seen": 2385874,
      "step": 1266,
      "train_runtime": 3821.3407,
      "train_tokens_per_second": 624.355
    },
    {
      "epoch": 1.280010108668183,
      "grad_norm": 1.1629916429519653,
      "learning_rate": 7.230379746835443e-06,
      "loss": 0.06323608756065369,
      "num_input_tokens_seen": 2388636,
      "step": 1267,
      "train_runtime": 3823.8359,
      "train_tokens_per_second": 624.67
    },
    {
      "epoch": 1.2810209754864796,
      "grad_norm": 1.362239122390747,
      "learning_rate": 7.220253164556963e-06,
      "loss": 0.05546117573976517,
      "num_input_tokens_seen": 2390426,
      "step": 1268,
      "train_runtime": 3826.0566,
      "train_tokens_per_second": 624.775
    },
    {
      "epoch": 1.2820318423047763,
      "grad_norm": 1.3258403539657593,
      "learning_rate": 7.2101265822784814e-06,
      "loss": 0.06692615896463394,
      "num_input_tokens_seen": 2392536,
      "step": 1269,
      "train_runtime": 3828.3655,
      "train_tokens_per_second": 624.95
    },
    {
      "epoch": 1.2830427091230732,
      "grad_norm": 1.2993758916854858,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 0.047084324061870575,
      "num_input_tokens_seen": 2394276,
      "step": 1270,
      "train_runtime": 3830.6242,
      "train_tokens_per_second": 625.035
    },
    {
      "epoch": 1.2840535759413698,
      "grad_norm": 1.388413667678833,
      "learning_rate": 7.189873417721519e-06,
      "loss": 0.059688519686460495,
      "num_input_tokens_seen": 2396094,
      "step": 1271,
      "train_runtime": 3833.0621,
      "train_tokens_per_second": 625.112
    },
    {
      "epoch": 1.2850644427596665,
      "grad_norm": 1.6889876127243042,
      "learning_rate": 7.179746835443039e-06,
      "loss": 0.08853891491889954,
      "num_input_tokens_seen": 2398042,
      "step": 1272,
      "train_runtime": 3835.3231,
      "train_tokens_per_second": 625.252
    },
    {
      "epoch": 1.2860753095779631,
      "grad_norm": 1.4108036756515503,
      "learning_rate": 7.169620253164558e-06,
      "loss": 0.06351727992296219,
      "num_input_tokens_seen": 2400290,
      "step": 1273,
      "train_runtime": 3837.6515,
      "train_tokens_per_second": 625.458
    },
    {
      "epoch": 1.2870861763962598,
      "grad_norm": 1.4529653787612915,
      "learning_rate": 7.159493670886076e-06,
      "loss": 0.046431802213191986,
      "num_input_tokens_seen": 2402024,
      "step": 1274,
      "train_runtime": 3839.8616,
      "train_tokens_per_second": 625.55
    },
    {
      "epoch": 1.2880970432145564,
      "grad_norm": 1.927977442741394,
      "learning_rate": 7.149367088607595e-06,
      "loss": 0.06092975288629532,
      "num_input_tokens_seen": 2403846,
      "step": 1275,
      "train_runtime": 3842.1551,
      "train_tokens_per_second": 625.65
    },
    {
      "epoch": 1.289107910032853,
      "grad_norm": 1.2304484844207764,
      "learning_rate": 7.139240506329115e-06,
      "loss": 0.054616767913103104,
      "num_input_tokens_seen": 2405930,
      "step": 1276,
      "train_runtime": 3844.4937,
      "train_tokens_per_second": 625.812
    },
    {
      "epoch": 1.2901187768511497,
      "grad_norm": 1.2034251689910889,
      "learning_rate": 7.129113924050633e-06,
      "loss": 0.05521988123655319,
      "num_input_tokens_seen": 2407904,
      "step": 1277,
      "train_runtime": 3846.7853,
      "train_tokens_per_second": 625.952
    },
    {
      "epoch": 1.2911296436694466,
      "grad_norm": 1.896065592765808,
      "learning_rate": 7.118987341772152e-06,
      "loss": 0.06472885608673096,
      "num_input_tokens_seen": 2410116,
      "step": 1278,
      "train_runtime": 3849.1352,
      "train_tokens_per_second": 626.145
    },
    {
      "epoch": 1.2921405104877433,
      "grad_norm": 1.20551598072052,
      "learning_rate": 7.108860759493671e-06,
      "loss": 0.06310249865055084,
      "num_input_tokens_seen": 2412480,
      "step": 1279,
      "train_runtime": 3851.5265,
      "train_tokens_per_second": 626.37
    },
    {
      "epoch": 1.29315137730604,
      "grad_norm": 1.4300223588943481,
      "learning_rate": 7.0987341772151905e-06,
      "loss": 0.0528576634824276,
      "num_input_tokens_seen": 2413934,
      "step": 1280,
      "train_runtime": 3853.6476,
      "train_tokens_per_second": 626.402
    },
    {
      "epoch": 1.2941622441243366,
      "grad_norm": 1.3833800554275513,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.07344702631235123,
      "num_input_tokens_seen": 2415538,
      "step": 1281,
      "train_runtime": 3855.8363,
      "train_tokens_per_second": 626.463
    },
    {
      "epoch": 1.2951731109426334,
      "grad_norm": 1.0929604768753052,
      "learning_rate": 7.078481012658228e-06,
      "loss": 0.04777875915169716,
      "num_input_tokens_seen": 2418354,
      "step": 1282,
      "train_runtime": 3858.3512,
      "train_tokens_per_second": 626.784
    },
    {
      "epoch": 1.29618397776093,
      "grad_norm": 1.6052258014678955,
      "learning_rate": 7.068354430379747e-06,
      "loss": 0.052375856786966324,
      "num_input_tokens_seen": 2420410,
      "step": 1283,
      "train_runtime": 3860.6545,
      "train_tokens_per_second": 626.943
    },
    {
      "epoch": 1.2971948445792267,
      "grad_norm": 1.1358217000961304,
      "learning_rate": 7.058227848101267e-06,
      "loss": 0.03975092992186546,
      "num_input_tokens_seen": 2422074,
      "step": 1284,
      "train_runtime": 3862.8286,
      "train_tokens_per_second": 627.021
    },
    {
      "epoch": 1.2982057113975234,
      "grad_norm": 1.4586998224258423,
      "learning_rate": 7.048101265822785e-06,
      "loss": 0.05373745039105415,
      "num_input_tokens_seen": 2423840,
      "step": 1285,
      "train_runtime": 3865.0327,
      "train_tokens_per_second": 627.12
    },
    {
      "epoch": 1.29921657821582,
      "grad_norm": 1.156754493713379,
      "learning_rate": 7.037974683544304e-06,
      "loss": 0.05291145294904709,
      "num_input_tokens_seen": 2425850,
      "step": 1286,
      "train_runtime": 3867.3022,
      "train_tokens_per_second": 627.272
    },
    {
      "epoch": 1.3002274450341167,
      "grad_norm": 1.2575299739837646,
      "learning_rate": 7.027848101265823e-06,
      "loss": 0.0599808506667614,
      "num_input_tokens_seen": 2428110,
      "step": 1287,
      "train_runtime": 3869.6466,
      "train_tokens_per_second": 627.476
    },
    {
      "epoch": 1.3012383118524133,
      "grad_norm": 1.223726749420166,
      "learning_rate": 7.017721518987342e-06,
      "loss": 0.05270208418369293,
      "num_input_tokens_seen": 2430236,
      "step": 1288,
      "train_runtime": 3871.9793,
      "train_tokens_per_second": 627.647
    },
    {
      "epoch": 1.3022491786707102,
      "grad_norm": 1.3750054836273193,
      "learning_rate": 7.0075949367088614e-06,
      "loss": 0.0491962805390358,
      "num_input_tokens_seen": 2432132,
      "step": 1289,
      "train_runtime": 3874.2428,
      "train_tokens_per_second": 627.77
    },
    {
      "epoch": 1.3032600454890069,
      "grad_norm": 1.603071689605713,
      "learning_rate": 6.99746835443038e-06,
      "loss": 0.08003854006528854,
      "num_input_tokens_seen": 2433912,
      "step": 1290,
      "train_runtime": 3876.4749,
      "train_tokens_per_second": 627.867
    },
    {
      "epoch": 1.3042709123073035,
      "grad_norm": 1.4792207479476929,
      "learning_rate": 6.987341772151899e-06,
      "loss": 0.06192871928215027,
      "num_input_tokens_seen": 2435450,
      "step": 1291,
      "train_runtime": 3892.4066,
      "train_tokens_per_second": 625.693
    },
    {
      "epoch": 1.3052817791256002,
      "grad_norm": 1.1881080865859985,
      "learning_rate": 6.977215189873419e-06,
      "loss": 0.048305366188287735,
      "num_input_tokens_seen": 2437190,
      "step": 1292,
      "train_runtime": 3894.6507,
      "train_tokens_per_second": 625.779
    },
    {
      "epoch": 1.3062926459438968,
      "grad_norm": 1.3181772232055664,
      "learning_rate": 6.967088607594937e-06,
      "loss": 0.054380863904953,
      "num_input_tokens_seen": 2439320,
      "step": 1293,
      "train_runtime": 3897.0466,
      "train_tokens_per_second": 625.941
    },
    {
      "epoch": 1.3073035127621937,
      "grad_norm": 1.3481340408325195,
      "learning_rate": 6.956962025316456e-06,
      "loss": 0.05058111995458603,
      "num_input_tokens_seen": 2440856,
      "step": 1294,
      "train_runtime": 3899.2289,
      "train_tokens_per_second": 625.984
    },
    {
      "epoch": 1.3083143795804903,
      "grad_norm": 1.3278868198394775,
      "learning_rate": 6.946835443037975e-06,
      "loss": 0.04658079892396927,
      "num_input_tokens_seen": 2442592,
      "step": 1295,
      "train_runtime": 3901.4947,
      "train_tokens_per_second": 626.066
    },
    {
      "epoch": 1.309325246398787,
      "grad_norm": 1.3212907314300537,
      "learning_rate": 6.936708860759494e-06,
      "loss": 0.07202155143022537,
      "num_input_tokens_seen": 2444864,
      "step": 1296,
      "train_runtime": 3903.8581,
      "train_tokens_per_second": 626.269
    },
    {
      "epoch": 1.3103361132170837,
      "grad_norm": 1.452693223953247,
      "learning_rate": 6.926582278481013e-06,
      "loss": 0.058132391422986984,
      "num_input_tokens_seen": 2446738,
      "step": 1297,
      "train_runtime": 3906.182,
      "train_tokens_per_second": 626.376
    },
    {
      "epoch": 1.3113469800353803,
      "grad_norm": 1.4813017845153809,
      "learning_rate": 6.916455696202532e-06,
      "loss": 0.0758880078792572,
      "num_input_tokens_seen": 2448408,
      "step": 1298,
      "train_runtime": 3908.4216,
      "train_tokens_per_second": 626.444
    },
    {
      "epoch": 1.312357846853677,
      "grad_norm": 1.4571219682693481,
      "learning_rate": 6.906329113924051e-06,
      "loss": 0.07111041247844696,
      "num_input_tokens_seen": 2450152,
      "step": 1299,
      "train_runtime": 3910.6613,
      "train_tokens_per_second": 626.531
    },
    {
      "epoch": 1.3133687136719736,
      "grad_norm": 1.1863071918487549,
      "learning_rate": 6.8962025316455705e-06,
      "loss": 0.060917239636182785,
      "num_input_tokens_seen": 2451912,
      "step": 1300,
      "train_runtime": 3913.021,
      "train_tokens_per_second": 626.603
    },
    {
      "epoch": 1.3143795804902705,
      "grad_norm": 1.6423072814941406,
      "learning_rate": 6.88607594936709e-06,
      "loss": 0.04655277729034424,
      "num_input_tokens_seen": 2453816,
      "step": 1301,
      "train_runtime": 3915.3644,
      "train_tokens_per_second": 626.715
    },
    {
      "epoch": 1.3153904473085671,
      "grad_norm": 1.3400390148162842,
      "learning_rate": 6.875949367088608e-06,
      "loss": 0.04788367822766304,
      "num_input_tokens_seen": 2455718,
      "step": 1302,
      "train_runtime": 3917.7202,
      "train_tokens_per_second": 626.823
    },
    {
      "epoch": 1.3164013141268638,
      "grad_norm": 1.463574767112732,
      "learning_rate": 6.865822784810127e-06,
      "loss": 0.06563840061426163,
      "num_input_tokens_seen": 2457762,
      "step": 1303,
      "train_runtime": 3920.0283,
      "train_tokens_per_second": 626.976
    },
    {
      "epoch": 1.3174121809451604,
      "grad_norm": 1.3056226968765259,
      "learning_rate": 6.855696202531647e-06,
      "loss": 0.061565324664115906,
      "num_input_tokens_seen": 2460088,
      "step": 1304,
      "train_runtime": 3922.4999,
      "train_tokens_per_second": 627.173
    },
    {
      "epoch": 1.318423047763457,
      "grad_norm": 1.284570574760437,
      "learning_rate": 6.845569620253165e-06,
      "loss": 0.056237488985061646,
      "num_input_tokens_seen": 2461942,
      "step": 1305,
      "train_runtime": 3924.7495,
      "train_tokens_per_second": 627.286
    },
    {
      "epoch": 1.319433914581754,
      "grad_norm": 1.21128511428833,
      "learning_rate": 6.835443037974684e-06,
      "loss": 0.043198779225349426,
      "num_input_tokens_seen": 2463904,
      "step": 1306,
      "train_runtime": 3927.052,
      "train_tokens_per_second": 627.418
    },
    {
      "epoch": 1.3204447814000506,
      "grad_norm": 1.1530758142471313,
      "learning_rate": 6.825316455696202e-06,
      "loss": 0.0534062460064888,
      "num_input_tokens_seen": 2465722,
      "step": 1307,
      "train_runtime": 3929.3542,
      "train_tokens_per_second": 627.513
    },
    {
      "epoch": 1.3214556482183473,
      "grad_norm": 1.3490420579910278,
      "learning_rate": 6.815189873417722e-06,
      "loss": 0.05943992733955383,
      "num_input_tokens_seen": 2467266,
      "step": 1308,
      "train_runtime": 3931.5583,
      "train_tokens_per_second": 627.554
    },
    {
      "epoch": 1.322466515036644,
      "grad_norm": 1.3117525577545166,
      "learning_rate": 6.805063291139241e-06,
      "loss": 0.05493208020925522,
      "num_input_tokens_seen": 2469430,
      "step": 1309,
      "train_runtime": 3933.9128,
      "train_tokens_per_second": 627.729
    },
    {
      "epoch": 1.3234773818549406,
      "grad_norm": 1.1769880056381226,
      "learning_rate": 6.79493670886076e-06,
      "loss": 0.05271150916814804,
      "num_input_tokens_seen": 2471652,
      "step": 1310,
      "train_runtime": 3936.2597,
      "train_tokens_per_second": 627.919
    },
    {
      "epoch": 1.3244882486732372,
      "grad_norm": 1.6591688394546509,
      "learning_rate": 6.784810126582279e-06,
      "loss": 0.08554491400718689,
      "num_input_tokens_seen": 2473280,
      "step": 1311,
      "train_runtime": 3938.4836,
      "train_tokens_per_second": 627.978
    },
    {
      "epoch": 1.3254991154915339,
      "grad_norm": 1.4286710023880005,
      "learning_rate": 6.774683544303799e-06,
      "loss": 0.06489214301109314,
      "num_input_tokens_seen": 2475394,
      "step": 1312,
      "train_runtime": 3940.7957,
      "train_tokens_per_second": 628.146
    },
    {
      "epoch": 1.3265099823098307,
      "grad_norm": 1.3228434324264526,
      "learning_rate": 6.764556962025317e-06,
      "loss": 0.0692477747797966,
      "num_input_tokens_seen": 2477104,
      "step": 1313,
      "train_runtime": 3943.0216,
      "train_tokens_per_second": 628.225
    },
    {
      "epoch": 1.3275208491281274,
      "grad_norm": 1.3991831541061401,
      "learning_rate": 6.754430379746836e-06,
      "loss": 0.06959948688745499,
      "num_input_tokens_seen": 2479042,
      "step": 1314,
      "train_runtime": 3945.3329,
      "train_tokens_per_second": 628.348
    },
    {
      "epoch": 1.328531715946424,
      "grad_norm": 1.1324790716171265,
      "learning_rate": 6.744303797468354e-06,
      "loss": 0.03849688172340393,
      "num_input_tokens_seen": 2480676,
      "step": 1315,
      "train_runtime": 3947.5461,
      "train_tokens_per_second": 628.41
    },
    {
      "epoch": 1.3295425827647207,
      "grad_norm": 1.4252482652664185,
      "learning_rate": 6.734177215189874e-06,
      "loss": 0.06371836364269257,
      "num_input_tokens_seen": 2482858,
      "step": 1316,
      "train_runtime": 3949.9076,
      "train_tokens_per_second": 628.586
    },
    {
      "epoch": 1.3305534495830174,
      "grad_norm": 1.5386812686920166,
      "learning_rate": 6.724050632911393e-06,
      "loss": 0.07826663553714752,
      "num_input_tokens_seen": 2484364,
      "step": 1317,
      "train_runtime": 3952.1292,
      "train_tokens_per_second": 628.614
    },
    {
      "epoch": 1.3315643164013142,
      "grad_norm": 1.658719778060913,
      "learning_rate": 6.7139240506329115e-06,
      "loss": 0.05043855309486389,
      "num_input_tokens_seen": 2486306,
      "step": 1318,
      "train_runtime": 3954.4094,
      "train_tokens_per_second": 628.743
    },
    {
      "epoch": 1.3325751832196109,
      "grad_norm": 1.2035260200500488,
      "learning_rate": 6.7037974683544306e-06,
      "loss": 0.06669348478317261,
      "num_input_tokens_seen": 2488390,
      "step": 1319,
      "train_runtime": 3956.7139,
      "train_tokens_per_second": 628.903
    },
    {
      "epoch": 1.3335860500379075,
      "grad_norm": 1.6911429166793823,
      "learning_rate": 6.6936708860759505e-06,
      "loss": 0.051741354167461395,
      "num_input_tokens_seen": 2490366,
      "step": 1320,
      "train_runtime": 3959.009,
      "train_tokens_per_second": 629.038
    },
    {
      "epoch": 1.3345969168562042,
      "grad_norm": 1.2079441547393799,
      "learning_rate": 6.683544303797469e-06,
      "loss": 0.054227374494075775,
      "num_input_tokens_seen": 2492082,
      "step": 1321,
      "train_runtime": 3975.5875,
      "train_tokens_per_second": 626.846
    },
    {
      "epoch": 1.3356077836745008,
      "grad_norm": 1.141798496246338,
      "learning_rate": 6.673417721518988e-06,
      "loss": 0.0495566725730896,
      "num_input_tokens_seen": 2493558,
      "step": 1322,
      "train_runtime": 3977.7635,
      "train_tokens_per_second": 626.874
    },
    {
      "epoch": 1.3366186504927975,
      "grad_norm": 1.162388801574707,
      "learning_rate": 6.663291139240506e-06,
      "loss": 0.04409773647785187,
      "num_input_tokens_seen": 2494988,
      "step": 1323,
      "train_runtime": 3979.9302,
      "train_tokens_per_second": 626.892
    },
    {
      "epoch": 1.3376295173110941,
      "grad_norm": 1.5943406820297241,
      "learning_rate": 6.653164556962026e-06,
      "loss": 0.07473370432853699,
      "num_input_tokens_seen": 2496558,
      "step": 1324,
      "train_runtime": 3982.1704,
      "train_tokens_per_second": 626.934
    },
    {
      "epoch": 1.338640384129391,
      "grad_norm": 1.2546615600585938,
      "learning_rate": 6.643037974683545e-06,
      "loss": 0.05474703758955002,
      "num_input_tokens_seen": 2498468,
      "step": 1325,
      "train_runtime": 3984.4065,
      "train_tokens_per_second": 627.062
    },
    {
      "epoch": 1.3396512509476877,
      "grad_norm": 1.1279000043869019,
      "learning_rate": 6.632911392405063e-06,
      "loss": 0.038108162581920624,
      "num_input_tokens_seen": 2500298,
      "step": 1326,
      "train_runtime": 3986.642,
      "train_tokens_per_second": 627.169
    },
    {
      "epoch": 1.3406621177659843,
      "grad_norm": 1.7238247394561768,
      "learning_rate": 6.622784810126582e-06,
      "loss": 0.0777852013707161,
      "num_input_tokens_seen": 2502174,
      "step": 1327,
      "train_runtime": 3988.9541,
      "train_tokens_per_second": 627.276
    },
    {
      "epoch": 1.341672984584281,
      "grad_norm": 1.118719220161438,
      "learning_rate": 6.612658227848102e-06,
      "loss": 0.05201104283332825,
      "num_input_tokens_seen": 2503952,
      "step": 1328,
      "train_runtime": 3991.1962,
      "train_tokens_per_second": 627.369
    },
    {
      "epoch": 1.3426838514025778,
      "grad_norm": 1.1212520599365234,
      "learning_rate": 6.6025316455696206e-06,
      "loss": 0.05457029491662979,
      "num_input_tokens_seen": 2505714,
      "step": 1329,
      "train_runtime": 3993.4535,
      "train_tokens_per_second": 627.455
    },
    {
      "epoch": 1.3436947182208745,
      "grad_norm": 1.209311842918396,
      "learning_rate": 6.59240506329114e-06,
      "loss": 0.07266074419021606,
      "num_input_tokens_seen": 2507846,
      "step": 1330,
      "train_runtime": 3995.7723,
      "train_tokens_per_second": 627.625
    },
    {
      "epoch": 1.3447055850391711,
      "grad_norm": 1.234279751777649,
      "learning_rate": 6.582278481012659e-06,
      "loss": 0.03919084370136261,
      "num_input_tokens_seen": 2510048,
      "step": 1331,
      "train_runtime": 3998.1095,
      "train_tokens_per_second": 627.809
    },
    {
      "epoch": 1.3457164518574678,
      "grad_norm": 1.3071277141571045,
      "learning_rate": 6.572151898734178e-06,
      "loss": 0.05816945433616638,
      "num_input_tokens_seen": 2511812,
      "step": 1332,
      "train_runtime": 4000.3514,
      "train_tokens_per_second": 627.898
    },
    {
      "epoch": 1.3467273186757645,
      "grad_norm": 2.054145097732544,
      "learning_rate": 6.562025316455697e-06,
      "loss": 0.08112684637308121,
      "num_input_tokens_seen": 2514084,
      "step": 1333,
      "train_runtime": 4002.7095,
      "train_tokens_per_second": 628.096
    },
    {
      "epoch": 1.347738185494061,
      "grad_norm": 1.3808009624481201,
      "learning_rate": 6.551898734177215e-06,
      "loss": 0.045785874128341675,
      "num_input_tokens_seen": 2515838,
      "step": 1334,
      "train_runtime": 4004.9185,
      "train_tokens_per_second": 628.187
    },
    {
      "epoch": 1.3487490523123578,
      "grad_norm": 1.2221318483352661,
      "learning_rate": 6.541772151898734e-06,
      "loss": 0.054682791233062744,
      "num_input_tokens_seen": 2517778,
      "step": 1335,
      "train_runtime": 4007.2695,
      "train_tokens_per_second": 628.303
    },
    {
      "epoch": 1.3497599191306544,
      "grad_norm": 1.4122490882873535,
      "learning_rate": 6.531645569620254e-06,
      "loss": 0.05897856131196022,
      "num_input_tokens_seen": 2519282,
      "step": 1336,
      "train_runtime": 4009.468,
      "train_tokens_per_second": 628.333
    },
    {
      "epoch": 1.3507707859489513,
      "grad_norm": 1.4037399291992188,
      "learning_rate": 6.521518987341772e-06,
      "loss": 0.06538009643554688,
      "num_input_tokens_seen": 2520964,
      "step": 1337,
      "train_runtime": 4011.7061,
      "train_tokens_per_second": 628.402
    },
    {
      "epoch": 1.351781652767248,
      "grad_norm": 1.2378387451171875,
      "learning_rate": 6.5113924050632915e-06,
      "loss": 0.06422580778598785,
      "num_input_tokens_seen": 2523034,
      "step": 1338,
      "train_runtime": 4014.0088,
      "train_tokens_per_second": 628.557
    },
    {
      "epoch": 1.3527925195855446,
      "grad_norm": 1.3696224689483643,
      "learning_rate": 6.5012658227848106e-06,
      "loss": 0.054821230471134186,
      "num_input_tokens_seen": 2525260,
      "step": 1339,
      "train_runtime": 4016.3466,
      "train_tokens_per_second": 628.746
    },
    {
      "epoch": 1.3538033864038412,
      "grad_norm": 1.2748407125473022,
      "learning_rate": 6.49113924050633e-06,
      "loss": 0.0619157999753952,
      "num_input_tokens_seen": 2527338,
      "step": 1340,
      "train_runtime": 4018.6636,
      "train_tokens_per_second": 628.9
    },
    {
      "epoch": 1.354814253222138,
      "grad_norm": 1.5537949800491333,
      "learning_rate": 6.481012658227849e-06,
      "loss": 0.06975711137056351,
      "num_input_tokens_seen": 2529494,
      "step": 1341,
      "train_runtime": 4020.973,
      "train_tokens_per_second": 629.075
    },
    {
      "epoch": 1.3558251200404348,
      "grad_norm": 1.4514718055725098,
      "learning_rate": 6.470886075949368e-06,
      "loss": 0.08354055136442184,
      "num_input_tokens_seen": 2531304,
      "step": 1342,
      "train_runtime": 4023.2288,
      "train_tokens_per_second": 629.172
    },
    {
      "epoch": 1.3568359868587314,
      "grad_norm": 1.5649539232254028,
      "learning_rate": 6.460759493670886e-06,
      "loss": 0.07228446006774902,
      "num_input_tokens_seen": 2533062,
      "step": 1343,
      "train_runtime": 4025.476,
      "train_tokens_per_second": 629.258
    },
    {
      "epoch": 1.357846853677028,
      "grad_norm": 1.298068642616272,
      "learning_rate": 6.450632911392406e-06,
      "loss": 0.06058548018336296,
      "num_input_tokens_seen": 2535360,
      "step": 1344,
      "train_runtime": 4027.896,
      "train_tokens_per_second": 629.45
    },
    {
      "epoch": 1.3588577204953247,
      "grad_norm": 1.2628767490386963,
      "learning_rate": 6.440506329113925e-06,
      "loss": 0.059151772409677505,
      "num_input_tokens_seen": 2537128,
      "step": 1345,
      "train_runtime": 4030.13,
      "train_tokens_per_second": 629.54
    },
    {
      "epoch": 1.3598685873136214,
      "grad_norm": 1.291844129562378,
      "learning_rate": 6.430379746835443e-06,
      "loss": 0.0626007616519928,
      "num_input_tokens_seen": 2539182,
      "step": 1346,
      "train_runtime": 4032.4581,
      "train_tokens_per_second": 629.686
    },
    {
      "epoch": 1.360879454131918,
      "grad_norm": 2.242800235748291,
      "learning_rate": 6.420253164556962e-06,
      "loss": 0.0355905145406723,
      "num_input_tokens_seen": 2541200,
      "step": 1347,
      "train_runtime": 4034.7924,
      "train_tokens_per_second": 629.822
    },
    {
      "epoch": 1.3618903209502147,
      "grad_norm": 1.0931439399719238,
      "learning_rate": 6.410126582278482e-06,
      "loss": 0.04841747507452965,
      "num_input_tokens_seen": 2543378,
      "step": 1348,
      "train_runtime": 4037.1568,
      "train_tokens_per_second": 629.992
    },
    {
      "epoch": 1.3629011877685115,
      "grad_norm": 1.017020344734192,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.04899681359529495,
      "num_input_tokens_seen": 2545350,
      "step": 1349,
      "train_runtime": 4039.4283,
      "train_tokens_per_second": 630.126
    },
    {
      "epoch": 1.3639120545868082,
      "grad_norm": 1.3168435096740723,
      "learning_rate": 6.38987341772152e-06,
      "loss": 0.061506081372499466,
      "num_input_tokens_seen": 2547356,
      "step": 1350,
      "train_runtime": 4041.7682,
      "train_tokens_per_second": 630.258
    },
    {
      "epoch": 1.3649229214051048,
      "grad_norm": 1.3419066667556763,
      "learning_rate": 6.379746835443038e-06,
      "loss": 0.051432445645332336,
      "num_input_tokens_seen": 2549126,
      "step": 1351,
      "train_runtime": 4057.4305,
      "train_tokens_per_second": 628.261
    },
    {
      "epoch": 1.3659337882234015,
      "grad_norm": 1.3823827505111694,
      "learning_rate": 6.369620253164558e-06,
      "loss": 0.059081874787807465,
      "num_input_tokens_seen": 2551066,
      "step": 1352,
      "train_runtime": 4059.7077,
      "train_tokens_per_second": 628.387
    },
    {
      "epoch": 1.3669446550416984,
      "grad_norm": 1.4057139158248901,
      "learning_rate": 6.359493670886077e-06,
      "loss": 0.07083790749311447,
      "num_input_tokens_seen": 2553038,
      "step": 1353,
      "train_runtime": 4061.9986,
      "train_tokens_per_second": 628.518
    },
    {
      "epoch": 1.367955521859995,
      "grad_norm": 1.2331106662750244,
      "learning_rate": 6.349367088607595e-06,
      "loss": 0.049345988780260086,
      "num_input_tokens_seen": 2555022,
      "step": 1354,
      "train_runtime": 4064.2766,
      "train_tokens_per_second": 628.654
    },
    {
      "epoch": 1.3689663886782917,
      "grad_norm": 1.599181890487671,
      "learning_rate": 6.339240506329114e-06,
      "loss": 0.03669581562280655,
      "num_input_tokens_seen": 2556426,
      "step": 1355,
      "train_runtime": 4066.4235,
      "train_tokens_per_second": 628.667
    },
    {
      "epoch": 1.3699772554965883,
      "grad_norm": 1.5763306617736816,
      "learning_rate": 6.329113924050634e-06,
      "loss": 0.0630222037434578,
      "num_input_tokens_seen": 2558122,
      "step": 1356,
      "train_runtime": 4068.6434,
      "train_tokens_per_second": 628.741
    },
    {
      "epoch": 1.370988122314885,
      "grad_norm": 1.3165096044540405,
      "learning_rate": 6.318987341772152e-06,
      "loss": 0.060910359025001526,
      "num_input_tokens_seen": 2560308,
      "step": 1357,
      "train_runtime": 4070.9815,
      "train_tokens_per_second": 628.917
    },
    {
      "epoch": 1.3719989891331816,
      "grad_norm": 1.2796040773391724,
      "learning_rate": 6.3088607594936715e-06,
      "loss": 0.05310271307826042,
      "num_input_tokens_seen": 2562338,
      "step": 1358,
      "train_runtime": 4073.2616,
      "train_tokens_per_second": 629.063
    },
    {
      "epoch": 1.3730098559514783,
      "grad_norm": 1.2358952760696411,
      "learning_rate": 6.29873417721519e-06,
      "loss": 0.05557725951075554,
      "num_input_tokens_seen": 2563842,
      "step": 1359,
      "train_runtime": 4075.4347,
      "train_tokens_per_second": 629.097
    },
    {
      "epoch": 1.3740207227697752,
      "grad_norm": 1.4226715564727783,
      "learning_rate": 6.28860759493671e-06,
      "loss": 0.054653145372867584,
      "num_input_tokens_seen": 2565870,
      "step": 1360,
      "train_runtime": 4077.7367,
      "train_tokens_per_second": 629.239
    },
    {
      "epoch": 1.3750315895880718,
      "grad_norm": 1.3460056781768799,
      "learning_rate": 6.278481012658229e-06,
      "loss": 0.05934952199459076,
      "num_input_tokens_seen": 2567614,
      "step": 1361,
      "train_runtime": 4080.234,
      "train_tokens_per_second": 629.281
    },
    {
      "epoch": 1.3760424564063685,
      "grad_norm": 1.0250606536865234,
      "learning_rate": 6.268354430379747e-06,
      "loss": 0.05108030512928963,
      "num_input_tokens_seen": 2569486,
      "step": 1362,
      "train_runtime": 4082.5838,
      "train_tokens_per_second": 629.377
    },
    {
      "epoch": 1.3770533232246651,
      "grad_norm": 1.0781219005584717,
      "learning_rate": 6.258227848101266e-06,
      "loss": 0.04799749329686165,
      "num_input_tokens_seen": 2571478,
      "step": 1363,
      "train_runtime": 4084.972,
      "train_tokens_per_second": 629.497
    },
    {
      "epoch": 1.3780641900429618,
      "grad_norm": 1.516428828239441,
      "learning_rate": 6.248101265822786e-06,
      "loss": 0.062270086258649826,
      "num_input_tokens_seen": 2573162,
      "step": 1364,
      "train_runtime": 4087.2647,
      "train_tokens_per_second": 629.556
    },
    {
      "epoch": 1.3790750568612586,
      "grad_norm": 1.3159164190292358,
      "learning_rate": 6.237974683544304e-06,
      "loss": 0.045468129217624664,
      "num_input_tokens_seen": 2574666,
      "step": 1365,
      "train_runtime": 4089.4563,
      "train_tokens_per_second": 629.586
    },
    {
      "epoch": 1.3800859236795553,
      "grad_norm": 1.253880500793457,
      "learning_rate": 6.227848101265823e-06,
      "loss": 0.04943940415978432,
      "num_input_tokens_seen": 2576094,
      "step": 1366,
      "train_runtime": 4091.6038,
      "train_tokens_per_second": 629.605
    },
    {
      "epoch": 1.381096790497852,
      "grad_norm": 1.144112467765808,
      "learning_rate": 6.2177215189873415e-06,
      "loss": 0.04010799527168274,
      "num_input_tokens_seen": 2577802,
      "step": 1367,
      "train_runtime": 4093.9055,
      "train_tokens_per_second": 629.668
    },
    {
      "epoch": 1.3821076573161486,
      "grad_norm": 1.289851427078247,
      "learning_rate": 6.2075949367088615e-06,
      "loss": 0.05895276367664337,
      "num_input_tokens_seen": 2580078,
      "step": 1368,
      "train_runtime": 4096.2773,
      "train_tokens_per_second": 629.859
    },
    {
      "epoch": 1.3831185241344452,
      "grad_norm": 1.3746508359909058,
      "learning_rate": 6.1974683544303805e-06,
      "loss": 0.05186750739812851,
      "num_input_tokens_seen": 2581822,
      "step": 1369,
      "train_runtime": 4098.5361,
      "train_tokens_per_second": 629.938
    },
    {
      "epoch": 1.384129390952742,
      "grad_norm": 1.1321715116500854,
      "learning_rate": 6.187341772151899e-06,
      "loss": 0.05690278112888336,
      "num_input_tokens_seen": 2584068,
      "step": 1370,
      "train_runtime": 4100.8891,
      "train_tokens_per_second": 630.124
    },
    {
      "epoch": 1.3851402577710386,
      "grad_norm": 1.1682864427566528,
      "learning_rate": 6.177215189873418e-06,
      "loss": 0.035503584891557693,
      "num_input_tokens_seen": 2585860,
      "step": 1371,
      "train_runtime": 4103.1333,
      "train_tokens_per_second": 630.216
    },
    {
      "epoch": 1.3861511245893354,
      "grad_norm": 1.2780135869979858,
      "learning_rate": 6.167088607594938e-06,
      "loss": 0.05588618293404579,
      "num_input_tokens_seen": 2587634,
      "step": 1372,
      "train_runtime": 4105.369,
      "train_tokens_per_second": 630.305
    },
    {
      "epoch": 1.387161991407632,
      "grad_norm": 1.4143054485321045,
      "learning_rate": 6.156962025316456e-06,
      "loss": 0.05068785697221756,
      "num_input_tokens_seen": 2589498,
      "step": 1373,
      "train_runtime": 4107.6688,
      "train_tokens_per_second": 630.406
    },
    {
      "epoch": 1.3881728582259287,
      "grad_norm": 2.3067827224731445,
      "learning_rate": 6.146835443037975e-06,
      "loss": 0.04583851993083954,
      "num_input_tokens_seen": 2591676,
      "step": 1374,
      "train_runtime": 4110.008,
      "train_tokens_per_second": 630.577
    },
    {
      "epoch": 1.3891837250442254,
      "grad_norm": 1.3914011716842651,
      "learning_rate": 6.136708860759494e-06,
      "loss": 0.04922180250287056,
      "num_input_tokens_seen": 2593544,
      "step": 1375,
      "train_runtime": 4112.3136,
      "train_tokens_per_second": 630.678
    },
    {
      "epoch": 1.390194591862522,
      "grad_norm": 1.1687066555023193,
      "learning_rate": 6.126582278481013e-06,
      "loss": 0.05256985127925873,
      "num_input_tokens_seen": 2595692,
      "step": 1376,
      "train_runtime": 4114.6516,
      "train_tokens_per_second": 630.841
    },
    {
      "epoch": 1.391205458680819,
      "grad_norm": 1.7809689044952393,
      "learning_rate": 6.116455696202532e-06,
      "loss": 0.088633693754673,
      "num_input_tokens_seen": 2597534,
      "step": 1377,
      "train_runtime": 4116.8917,
      "train_tokens_per_second": 630.945
    },
    {
      "epoch": 1.3922163254991156,
      "grad_norm": 1.2149196863174438,
      "learning_rate": 6.106329113924051e-06,
      "loss": 0.07367751747369766,
      "num_input_tokens_seen": 2600018,
      "step": 1378,
      "train_runtime": 4119.299,
      "train_tokens_per_second": 631.18
    },
    {
      "epoch": 1.3932271923174122,
      "grad_norm": 1.497098684310913,
      "learning_rate": 6.09620253164557e-06,
      "loss": 0.05990324541926384,
      "num_input_tokens_seen": 2602052,
      "step": 1379,
      "train_runtime": 4121.5898,
      "train_tokens_per_second": 631.322
    },
    {
      "epoch": 1.3942380591357089,
      "grad_norm": 1.3180423974990845,
      "learning_rate": 6.08607594936709e-06,
      "loss": 0.05272394418716431,
      "num_input_tokens_seen": 2603758,
      "step": 1380,
      "train_runtime": 4123.8066,
      "train_tokens_per_second": 631.397
    },
    {
      "epoch": 1.3952489259540055,
      "grad_norm": 1.1364562511444092,
      "learning_rate": 6.075949367088608e-06,
      "loss": 0.04444532468914986,
      "num_input_tokens_seen": 2605644,
      "step": 1381,
      "train_runtime": 4139.3944,
      "train_tokens_per_second": 629.475
    },
    {
      "epoch": 1.3962597927723022,
      "grad_norm": 1.0885640382766724,
      "learning_rate": 6.065822784810127e-06,
      "loss": 0.04684608429670334,
      "num_input_tokens_seen": 2607354,
      "step": 1382,
      "train_runtime": 4141.7807,
      "train_tokens_per_second": 629.525
    },
    {
      "epoch": 1.3972706595905988,
      "grad_norm": 1.0900933742523193,
      "learning_rate": 6.055696202531646e-06,
      "loss": 0.04039217159152031,
      "num_input_tokens_seen": 2609280,
      "step": 1383,
      "train_runtime": 4144.264,
      "train_tokens_per_second": 629.612
    },
    {
      "epoch": 1.3982815264088957,
      "grad_norm": 1.595590591430664,
      "learning_rate": 6.045569620253165e-06,
      "loss": 0.09087800979614258,
      "num_input_tokens_seen": 2611402,
      "step": 1384,
      "train_runtime": 4146.6957,
      "train_tokens_per_second": 629.755
    },
    {
      "epoch": 1.3992923932271923,
      "grad_norm": 1.808232069015503,
      "learning_rate": 6.035443037974684e-06,
      "loss": 0.06717859208583832,
      "num_input_tokens_seen": 2612776,
      "step": 1385,
      "train_runtime": 4148.8773,
      "train_tokens_per_second": 629.755
    },
    {
      "epoch": 1.400303260045489,
      "grad_norm": 1.3342633247375488,
      "learning_rate": 6.025316455696203e-06,
      "loss": 0.06523144990205765,
      "num_input_tokens_seen": 2614562,
      "step": 1386,
      "train_runtime": 4151.161,
      "train_tokens_per_second": 629.839
    },
    {
      "epoch": 1.400303260045489,
      "eval_loss": 0.3663868308067322,
      "eval_runtime": 54.4304,
      "eval_samples_per_second": 16.149,
      "eval_steps_per_second": 8.084,
      "num_input_tokens_seen": 2614562,
      "step": 1386
    },
    {
      "epoch": 1.4013141268637856,
      "grad_norm": 1.2589857578277588,
      "learning_rate": 6.0151898734177215e-06,
      "loss": 0.051366157829761505,
      "num_input_tokens_seen": 2616790,
      "step": 1387,
      "train_runtime": 4207.9601,
      "train_tokens_per_second": 621.867
    },
    {
      "epoch": 1.4023249936820825,
      "grad_norm": 1.1806261539459229,
      "learning_rate": 6.0050632911392415e-06,
      "loss": 0.047723814845085144,
      "num_input_tokens_seen": 2618498,
      "step": 1388,
      "train_runtime": 4210.2811,
      "train_tokens_per_second": 621.929
    },
    {
      "epoch": 1.4033358605003792,
      "grad_norm": 1.5206774473190308,
      "learning_rate": 5.9949367088607605e-06,
      "loss": 0.05788637325167656,
      "num_input_tokens_seen": 2620730,
      "step": 1389,
      "train_runtime": 4212.6223,
      "train_tokens_per_second": 622.114
    },
    {
      "epoch": 1.4043467273186758,
      "grad_norm": 1.2093030214309692,
      "learning_rate": 5.984810126582279e-06,
      "loss": 0.051944561302661896,
      "num_input_tokens_seen": 2622210,
      "step": 1390,
      "train_runtime": 4214.809,
      "train_tokens_per_second": 622.142
    },
    {
      "epoch": 1.4053575941369725,
      "grad_norm": 1.2598261833190918,
      "learning_rate": 5.974683544303798e-06,
      "loss": 0.060215115547180176,
      "num_input_tokens_seen": 2623708,
      "step": 1391,
      "train_runtime": 4216.9732,
      "train_tokens_per_second": 622.178
    },
    {
      "epoch": 1.4063684609552691,
      "grad_norm": 1.3054869174957275,
      "learning_rate": 5.964556962025318e-06,
      "loss": 0.07101056724786758,
      "num_input_tokens_seen": 2625760,
      "step": 1392,
      "train_runtime": 4219.3319,
      "train_tokens_per_second": 622.317
    },
    {
      "epoch": 1.4073793277735658,
      "grad_norm": 1.3434244394302368,
      "learning_rate": 5.954430379746836e-06,
      "loss": 0.06410141289234161,
      "num_input_tokens_seen": 2627750,
      "step": 1393,
      "train_runtime": 4221.6614,
      "train_tokens_per_second": 622.445
    },
    {
      "epoch": 1.4083901945918624,
      "grad_norm": 1.4165593385696411,
      "learning_rate": 5.944303797468355e-06,
      "loss": 0.062341753393411636,
      "num_input_tokens_seen": 2630024,
      "step": 1394,
      "train_runtime": 4224.0667,
      "train_tokens_per_second": 622.628
    },
    {
      "epoch": 1.409401061410159,
      "grad_norm": 1.2924467325210571,
      "learning_rate": 5.934177215189873e-06,
      "loss": 0.044377949088811874,
      "num_input_tokens_seen": 2631576,
      "step": 1395,
      "train_runtime": 4226.2458,
      "train_tokens_per_second": 622.675
    },
    {
      "epoch": 1.410411928228456,
      "grad_norm": 1.0085480213165283,
      "learning_rate": 5.924050632911393e-06,
      "loss": 0.046639084815979004,
      "num_input_tokens_seen": 2633758,
      "step": 1396,
      "train_runtime": 4228.5679,
      "train_tokens_per_second": 622.849
    },
    {
      "epoch": 1.4114227950467526,
      "grad_norm": 1.485747218132019,
      "learning_rate": 5.913924050632912e-06,
      "loss": 0.057200685143470764,
      "num_input_tokens_seen": 2635928,
      "step": 1397,
      "train_runtime": 4230.8845,
      "train_tokens_per_second": 623.021
    },
    {
      "epoch": 1.4124336618650493,
      "grad_norm": 1.4903545379638672,
      "learning_rate": 5.903797468354431e-06,
      "loss": 0.05418243259191513,
      "num_input_tokens_seen": 2637284,
      "step": 1398,
      "train_runtime": 4233.0715,
      "train_tokens_per_second": 623.019
    },
    {
      "epoch": 1.413444528683346,
      "grad_norm": 1.4975517988204956,
      "learning_rate": 5.89367088607595e-06,
      "loss": 0.06205286830663681,
      "num_input_tokens_seen": 2639002,
      "step": 1399,
      "train_runtime": 4235.2916,
      "train_tokens_per_second": 623.098
    },
    {
      "epoch": 1.4144553955016428,
      "grad_norm": 1.6095198392868042,
      "learning_rate": 5.88354430379747e-06,
      "loss": 0.053863804787397385,
      "num_input_tokens_seen": 2640572,
      "step": 1400,
      "train_runtime": 4237.4897,
      "train_tokens_per_second": 623.145
    },
    {
      "epoch": 1.4154662623199394,
      "grad_norm": 1.2906925678253174,
      "learning_rate": 5.873417721518988e-06,
      "loss": 0.05912800133228302,
      "num_input_tokens_seen": 2642784,
      "step": 1401,
      "train_runtime": 4239.8582,
      "train_tokens_per_second": 623.319
    },
    {
      "epoch": 1.416477129138236,
      "grad_norm": 0.9742889404296875,
      "learning_rate": 5.863291139240507e-06,
      "loss": 0.04222407191991806,
      "num_input_tokens_seen": 2645018,
      "step": 1402,
      "train_runtime": 4242.2005,
      "train_tokens_per_second": 623.501
    },
    {
      "epoch": 1.4174879959565327,
      "grad_norm": 1.3450779914855957,
      "learning_rate": 5.853164556962025e-06,
      "loss": 0.04991640895605087,
      "num_input_tokens_seen": 2646966,
      "step": 1403,
      "train_runtime": 4244.4943,
      "train_tokens_per_second": 623.623
    },
    {
      "epoch": 1.4184988627748294,
      "grad_norm": 1.3153343200683594,
      "learning_rate": 5.843037974683545e-06,
      "loss": 0.06341435760259628,
      "num_input_tokens_seen": 2649610,
      "step": 1404,
      "train_runtime": 4246.9311,
      "train_tokens_per_second": 623.888
    },
    {
      "epoch": 1.419509729593126,
      "grad_norm": 1.331592321395874,
      "learning_rate": 5.832911392405064e-06,
      "loss": 0.05741046741604805,
      "num_input_tokens_seen": 2651580,
      "step": 1405,
      "train_runtime": 4249.2265,
      "train_tokens_per_second": 624.015
    },
    {
      "epoch": 1.4205205964114227,
      "grad_norm": 1.0401424169540405,
      "learning_rate": 5.8227848101265824e-06,
      "loss": 0.0531093031167984,
      "num_input_tokens_seen": 2653682,
      "step": 1406,
      "train_runtime": 4251.5097,
      "train_tokens_per_second": 624.174
    },
    {
      "epoch": 1.4215314632297193,
      "grad_norm": 1.3022911548614502,
      "learning_rate": 5.8126582278481015e-06,
      "loss": 0.0452243909239769,
      "num_input_tokens_seen": 2655742,
      "step": 1407,
      "train_runtime": 4253.8024,
      "train_tokens_per_second": 624.322
    },
    {
      "epoch": 1.4225423300480162,
      "grad_norm": 1.4204517602920532,
      "learning_rate": 5.8025316455696214e-06,
      "loss": 0.06786120682954788,
      "num_input_tokens_seen": 2658112,
      "step": 1408,
      "train_runtime": 4256.172,
      "train_tokens_per_second": 624.531
    },
    {
      "epoch": 1.4235531968663129,
      "grad_norm": 1.1559796333312988,
      "learning_rate": 5.79240506329114e-06,
      "loss": 0.03615058958530426,
      "num_input_tokens_seen": 2659640,
      "step": 1409,
      "train_runtime": 4258.3596,
      "train_tokens_per_second": 624.569
    },
    {
      "epoch": 1.4245640636846095,
      "grad_norm": 1.2559725046157837,
      "learning_rate": 5.782278481012659e-06,
      "loss": 0.05145192891359329,
      "num_input_tokens_seen": 2661468,
      "step": 1410,
      "train_runtime": 4260.6147,
      "train_tokens_per_second": 624.668
    },
    {
      "epoch": 1.4255749305029062,
      "grad_norm": 1.135015606880188,
      "learning_rate": 5.772151898734177e-06,
      "loss": 0.04511488974094391,
      "num_input_tokens_seen": 2663680,
      "step": 1411,
      "train_runtime": 4277.3982,
      "train_tokens_per_second": 622.734
    },
    {
      "epoch": 1.426585797321203,
      "grad_norm": 1.108528733253479,
      "learning_rate": 5.762025316455697e-06,
      "loss": 0.058695998042821884,
      "num_input_tokens_seen": 2666480,
      "step": 1412,
      "train_runtime": 4279.9157,
      "train_tokens_per_second": 623.022
    },
    {
      "epoch": 1.4275966641394997,
      "grad_norm": 1.3734612464904785,
      "learning_rate": 5.751898734177216e-06,
      "loss": 0.06597540527582169,
      "num_input_tokens_seen": 2668490,
      "step": 1413,
      "train_runtime": 4282.2171,
      "train_tokens_per_second": 623.156
    },
    {
      "epoch": 1.4286075309577964,
      "grad_norm": 1.483980655670166,
      "learning_rate": 5.741772151898734e-06,
      "loss": 0.06790570914745331,
      "num_input_tokens_seen": 2670236,
      "step": 1414,
      "train_runtime": 4284.5602,
      "train_tokens_per_second": 623.223
    },
    {
      "epoch": 1.429618397776093,
      "grad_norm": 1.3360049724578857,
      "learning_rate": 5.731645569620253e-06,
      "loss": 0.0684797614812851,
      "num_input_tokens_seen": 2672264,
      "step": 1415,
      "train_runtime": 4286.8856,
      "train_tokens_per_second": 623.358
    },
    {
      "epoch": 1.4306292645943897,
      "grad_norm": 1.18574857711792,
      "learning_rate": 5.721518987341773e-06,
      "loss": 0.05380445718765259,
      "num_input_tokens_seen": 2674302,
      "step": 1416,
      "train_runtime": 4289.1527,
      "train_tokens_per_second": 623.504
    },
    {
      "epoch": 1.4316401314126863,
      "grad_norm": 1.2353146076202393,
      "learning_rate": 5.7113924050632915e-06,
      "loss": 0.04403803125023842,
      "num_input_tokens_seen": 2676036,
      "step": 1417,
      "train_runtime": 4291.3891,
      "train_tokens_per_second": 623.583
    },
    {
      "epoch": 1.432650998230983,
      "grad_norm": 1.565489649772644,
      "learning_rate": 5.701265822784811e-06,
      "loss": 0.08054578304290771,
      "num_input_tokens_seen": 2678014,
      "step": 1418,
      "train_runtime": 4293.6992,
      "train_tokens_per_second": 623.708
    },
    {
      "epoch": 1.4336618650492798,
      "grad_norm": 1.164918065071106,
      "learning_rate": 5.69113924050633e-06,
      "loss": 0.04634508490562439,
      "num_input_tokens_seen": 2679970,
      "step": 1419,
      "train_runtime": 4295.9793,
      "train_tokens_per_second": 623.832
    },
    {
      "epoch": 1.4346727318675765,
      "grad_norm": 1.5861961841583252,
      "learning_rate": 5.681012658227849e-06,
      "loss": 0.07068551331758499,
      "num_input_tokens_seen": 2681874,
      "step": 1420,
      "train_runtime": 4298.263,
      "train_tokens_per_second": 623.944
    },
    {
      "epoch": 1.4356835986858731,
      "grad_norm": 1.173180341720581,
      "learning_rate": 5.670886075949368e-06,
      "loss": 0.04166347160935402,
      "num_input_tokens_seen": 2683802,
      "step": 1421,
      "train_runtime": 4300.5509,
      "train_tokens_per_second": 624.06
    },
    {
      "epoch": 1.4366944655041698,
      "grad_norm": 1.6341310739517212,
      "learning_rate": 5.660759493670886e-06,
      "loss": 0.06834426522254944,
      "num_input_tokens_seen": 2685204,
      "step": 1422,
      "train_runtime": 4302.6621,
      "train_tokens_per_second": 624.08
    },
    {
      "epoch": 1.4377053323224664,
      "grad_norm": 1.2905670404434204,
      "learning_rate": 5.650632911392405e-06,
      "loss": 0.0567033626139164,
      "num_input_tokens_seen": 2687734,
      "step": 1423,
      "train_runtime": 4305.0868,
      "train_tokens_per_second": 624.316
    },
    {
      "epoch": 1.4387161991407633,
      "grad_norm": 1.6841981410980225,
      "learning_rate": 5.640506329113925e-06,
      "loss": 0.06087081879377365,
      "num_input_tokens_seen": 2689396,
      "step": 1424,
      "train_runtime": 4307.329,
      "train_tokens_per_second": 624.377
    },
    {
      "epoch": 1.43972706595906,
      "grad_norm": 1.744522213935852,
      "learning_rate": 5.630379746835443e-06,
      "loss": 0.07477660477161407,
      "num_input_tokens_seen": 2691060,
      "step": 1425,
      "train_runtime": 4309.5386,
      "train_tokens_per_second": 624.443
    },
    {
      "epoch": 1.4407379327773566,
      "grad_norm": 1.0791798830032349,
      "learning_rate": 5.6202531645569624e-06,
      "loss": 0.03928723931312561,
      "num_input_tokens_seen": 2692586,
      "step": 1426,
      "train_runtime": 4311.6925,
      "train_tokens_per_second": 624.485
    },
    {
      "epoch": 1.4417487995956533,
      "grad_norm": 1.3450736999511719,
      "learning_rate": 5.6101265822784815e-06,
      "loss": 0.06375723332166672,
      "num_input_tokens_seen": 2694948,
      "step": 1427,
      "train_runtime": 4314.0879,
      "train_tokens_per_second": 624.685
    },
    {
      "epoch": 1.44275966641395,
      "grad_norm": 2.001465320587158,
      "learning_rate": 5.600000000000001e-06,
      "loss": 0.05643053352832794,
      "num_input_tokens_seen": 2697050,
      "step": 1428,
      "train_runtime": 4316.3871,
      "train_tokens_per_second": 624.84
    },
    {
      "epoch": 1.4437705332322466,
      "grad_norm": 1.0974299907684326,
      "learning_rate": 5.58987341772152e-06,
      "loss": 0.05764308571815491,
      "num_input_tokens_seen": 2699474,
      "step": 1429,
      "train_runtime": 4318.7847,
      "train_tokens_per_second": 625.054
    },
    {
      "epoch": 1.4447814000505432,
      "grad_norm": 1.6183689832687378,
      "learning_rate": 5.579746835443039e-06,
      "loss": 0.07020246237516403,
      "num_input_tokens_seen": 2701090,
      "step": 1430,
      "train_runtime": 4320.9917,
      "train_tokens_per_second": 625.109
    },
    {
      "epoch": 1.44579226686884,
      "grad_norm": 1.2373324632644653,
      "learning_rate": 5.569620253164557e-06,
      "loss": 0.0557381808757782,
      "num_input_tokens_seen": 2702828,
      "step": 1431,
      "train_runtime": 4323.2133,
      "train_tokens_per_second": 625.19
    },
    {
      "epoch": 1.4468031336871368,
      "grad_norm": 1.3397997617721558,
      "learning_rate": 5.559493670886077e-06,
      "loss": 0.05817887932062149,
      "num_input_tokens_seen": 2704850,
      "step": 1432,
      "train_runtime": 4325.4847,
      "train_tokens_per_second": 625.329
    },
    {
      "epoch": 1.4478140005054334,
      "grad_norm": 1.0370874404907227,
      "learning_rate": 5.549367088607596e-06,
      "loss": 0.04069850966334343,
      "num_input_tokens_seen": 2706554,
      "step": 1433,
      "train_runtime": 4327.7331,
      "train_tokens_per_second": 625.398
    },
    {
      "epoch": 1.44882486732373,
      "grad_norm": 2.3947393894195557,
      "learning_rate": 5.539240506329114e-06,
      "loss": 0.06741803884506226,
      "num_input_tokens_seen": 2708422,
      "step": 1434,
      "train_runtime": 4330.0563,
      "train_tokens_per_second": 625.493
    },
    {
      "epoch": 1.4498357341420267,
      "grad_norm": 2.450782060623169,
      "learning_rate": 5.529113924050633e-06,
      "loss": 0.0634034126996994,
      "num_input_tokens_seen": 2710138,
      "step": 1435,
      "train_runtime": 4332.33,
      "train_tokens_per_second": 625.561
    },
    {
      "epoch": 1.4508466009603236,
      "grad_norm": 1.5516140460968018,
      "learning_rate": 5.518987341772153e-06,
      "loss": 0.08794175833463669,
      "num_input_tokens_seen": 2712362,
      "step": 1436,
      "train_runtime": 4334.739,
      "train_tokens_per_second": 625.727
    },
    {
      "epoch": 1.4518574677786202,
      "grad_norm": 1.4325538873672485,
      "learning_rate": 5.5088607594936715e-06,
      "loss": 0.06939689069986343,
      "num_input_tokens_seen": 2714324,
      "step": 1437,
      "train_runtime": 4337.0449,
      "train_tokens_per_second": 625.846
    },
    {
      "epoch": 1.4528683345969169,
      "grad_norm": 1.338346004486084,
      "learning_rate": 5.498734177215191e-06,
      "loss": 0.03815968334674835,
      "num_input_tokens_seen": 2716354,
      "step": 1438,
      "train_runtime": 4339.3454,
      "train_tokens_per_second": 625.982
    },
    {
      "epoch": 1.4538792014152135,
      "grad_norm": 1.0405141115188599,
      "learning_rate": 5.488607594936709e-06,
      "loss": 0.035293880850076675,
      "num_input_tokens_seen": 2718476,
      "step": 1439,
      "train_runtime": 4341.6901,
      "train_tokens_per_second": 626.133
    },
    {
      "epoch": 1.4548900682335102,
      "grad_norm": 2.9229345321655273,
      "learning_rate": 5.478481012658229e-06,
      "loss": 0.05270947888493538,
      "num_input_tokens_seen": 2720324,
      "step": 1440,
      "train_runtime": 4343.9417,
      "train_tokens_per_second": 626.234
    },
    {
      "epoch": 1.4559009350518068,
      "grad_norm": 1.3125211000442505,
      "learning_rate": 5.468354430379748e-06,
      "loss": 0.047921039164066315,
      "num_input_tokens_seen": 2721974,
      "step": 1441,
      "train_runtime": 4359.5344,
      "train_tokens_per_second": 624.373
    },
    {
      "epoch": 1.4569118018701035,
      "grad_norm": 1.2566816806793213,
      "learning_rate": 5.458227848101266e-06,
      "loss": 0.0636405348777771,
      "num_input_tokens_seen": 2724214,
      "step": 1442,
      "train_runtime": 4361.9161,
      "train_tokens_per_second": 624.545
    },
    {
      "epoch": 1.4579226686884004,
      "grad_norm": 2.127671480178833,
      "learning_rate": 5.448101265822785e-06,
      "loss": 0.09057963639497757,
      "num_input_tokens_seen": 2725932,
      "step": 1443,
      "train_runtime": 4364.43,
      "train_tokens_per_second": 624.579
    },
    {
      "epoch": 1.458933535506697,
      "grad_norm": 2.1411712169647217,
      "learning_rate": 5.437974683544303e-06,
      "loss": 0.060190021991729736,
      "num_input_tokens_seen": 2727498,
      "step": 1444,
      "train_runtime": 4366.7173,
      "train_tokens_per_second": 624.611
    },
    {
      "epoch": 1.4599444023249937,
      "grad_norm": 1.2093969583511353,
      "learning_rate": 5.427848101265823e-06,
      "loss": 0.042883116751909256,
      "num_input_tokens_seen": 2729276,
      "step": 1445,
      "train_runtime": 4368.9357,
      "train_tokens_per_second": 624.7
    },
    {
      "epoch": 1.4609552691432903,
      "grad_norm": 1.0948315858840942,
      "learning_rate": 5.417721518987342e-06,
      "loss": 0.04013436287641525,
      "num_input_tokens_seen": 2731094,
      "step": 1446,
      "train_runtime": 4371.2213,
      "train_tokens_per_second": 624.79
    },
    {
      "epoch": 1.4619661359615872,
      "grad_norm": 1.258665680885315,
      "learning_rate": 5.407594936708861e-06,
      "loss": 0.04527812823653221,
      "num_input_tokens_seen": 2732632,
      "step": 1447,
      "train_runtime": 4373.3988,
      "train_tokens_per_second": 624.83
    },
    {
      "epoch": 1.4629770027798839,
      "grad_norm": 1.2352830171585083,
      "learning_rate": 5.39746835443038e-06,
      "loss": 0.06785471737384796,
      "num_input_tokens_seen": 2734660,
      "step": 1448,
      "train_runtime": 4375.7124,
      "train_tokens_per_second": 624.963
    },
    {
      "epoch": 1.4639878695981805,
      "grad_norm": 1.411289930343628,
      "learning_rate": 5.3873417721519e-06,
      "loss": 0.06744305789470673,
      "num_input_tokens_seen": 2736088,
      "step": 1449,
      "train_runtime": 4377.8714,
      "train_tokens_per_second": 624.981
    },
    {
      "epoch": 1.4649987364164772,
      "grad_norm": 1.1486389636993408,
      "learning_rate": 5.377215189873418e-06,
      "loss": 0.041869692504405975,
      "num_input_tokens_seen": 2737516,
      "step": 1450,
      "train_runtime": 4380.0118,
      "train_tokens_per_second": 625.002
    },
    {
      "epoch": 1.4660096032347738,
      "grad_norm": 1.1115456819534302,
      "learning_rate": 5.367088607594937e-06,
      "loss": 0.05015439912676811,
      "num_input_tokens_seen": 2740122,
      "step": 1451,
      "train_runtime": 4382.459,
      "train_tokens_per_second": 625.248
    },
    {
      "epoch": 1.4670204700530705,
      "grad_norm": 1.5198140144348145,
      "learning_rate": 5.356962025316455e-06,
      "loss": 0.05162966996431351,
      "num_input_tokens_seen": 2741914,
      "step": 1452,
      "train_runtime": 4384.7514,
      "train_tokens_per_second": 625.329
    },
    {
      "epoch": 1.468031336871367,
      "grad_norm": 1.2787257432937622,
      "learning_rate": 5.346835443037975e-06,
      "loss": 0.04590470343828201,
      "num_input_tokens_seen": 2743786,
      "step": 1453,
      "train_runtime": 4387.0441,
      "train_tokens_per_second": 625.429
    },
    {
      "epoch": 1.4690422036896638,
      "grad_norm": 1.1875731945037842,
      "learning_rate": 5.336708860759494e-06,
      "loss": 0.048388317227363586,
      "num_input_tokens_seen": 2745376,
      "step": 1454,
      "train_runtime": 4389.209,
      "train_tokens_per_second": 625.483
    },
    {
      "epoch": 1.4700530705079606,
      "grad_norm": 1.0641950368881226,
      "learning_rate": 5.3265822784810125e-06,
      "loss": 0.04221426323056221,
      "num_input_tokens_seen": 2747474,
      "step": 1455,
      "train_runtime": 4391.5383,
      "train_tokens_per_second": 625.629
    },
    {
      "epoch": 1.4710639373262573,
      "grad_norm": 1.1338188648223877,
      "learning_rate": 5.3164556962025316e-06,
      "loss": 0.04780226945877075,
      "num_input_tokens_seen": 2749496,
      "step": 1456,
      "train_runtime": 4393.8789,
      "train_tokens_per_second": 625.756
    },
    {
      "epoch": 1.472074804144554,
      "grad_norm": 1.335322380065918,
      "learning_rate": 5.3063291139240515e-06,
      "loss": 0.06389928609132767,
      "num_input_tokens_seen": 2751386,
      "step": 1457,
      "train_runtime": 4396.1554,
      "train_tokens_per_second": 625.862
    },
    {
      "epoch": 1.4730856709628506,
      "grad_norm": 1.1962878704071045,
      "learning_rate": 5.29620253164557e-06,
      "loss": 0.05268601328134537,
      "num_input_tokens_seen": 2753164,
      "step": 1458,
      "train_runtime": 4398.3912,
      "train_tokens_per_second": 625.948
    },
    {
      "epoch": 1.4740965377811475,
      "grad_norm": 1.2526472806930542,
      "learning_rate": 5.286075949367089e-06,
      "loss": 0.050169482827186584,
      "num_input_tokens_seen": 2755032,
      "step": 1459,
      "train_runtime": 4400.6965,
      "train_tokens_per_second": 626.045
    },
    {
      "epoch": 1.4751074045994441,
      "grad_norm": 1.687121033668518,
      "learning_rate": 5.275949367088608e-06,
      "loss": 0.046259500086307526,
      "num_input_tokens_seen": 2756680,
      "step": 1460,
      "train_runtime": 4402.9523,
      "train_tokens_per_second": 626.098
    },
    {
      "epoch": 1.4761182714177408,
      "grad_norm": 1.152158260345459,
      "learning_rate": 5.265822784810127e-06,
      "loss": 0.04726831614971161,
      "num_input_tokens_seen": 2758496,
      "step": 1461,
      "train_runtime": 4405.1903,
      "train_tokens_per_second": 626.192
    },
    {
      "epoch": 1.4771291382360374,
      "grad_norm": 1.302581787109375,
      "learning_rate": 5.255696202531646e-06,
      "loss": 0.05704841390252113,
      "num_input_tokens_seen": 2760264,
      "step": 1462,
      "train_runtime": 4407.4014,
      "train_tokens_per_second": 626.279
    },
    {
      "epoch": 1.478140005054334,
      "grad_norm": 1.2298142910003662,
      "learning_rate": 5.245569620253165e-06,
      "loss": 0.053375955671072006,
      "num_input_tokens_seen": 2762474,
      "step": 1463,
      "train_runtime": 4409.7349,
      "train_tokens_per_second": 626.449
    },
    {
      "epoch": 1.4791508718726307,
      "grad_norm": 1.3026584386825562,
      "learning_rate": 5.235443037974683e-06,
      "loss": 0.07974346727132797,
      "num_input_tokens_seen": 2764950,
      "step": 1464,
      "train_runtime": 4412.1653,
      "train_tokens_per_second": 626.665
    },
    {
      "epoch": 1.4801617386909274,
      "grad_norm": 1.2211552858352661,
      "learning_rate": 5.225316455696203e-06,
      "loss": 0.04752020537853241,
      "num_input_tokens_seen": 2766558,
      "step": 1465,
      "train_runtime": 4414.3802,
      "train_tokens_per_second": 626.715
    },
    {
      "epoch": 1.481172605509224,
      "grad_norm": 1.1044403314590454,
      "learning_rate": 5.2151898734177216e-06,
      "loss": 0.04026167094707489,
      "num_input_tokens_seen": 2768498,
      "step": 1466,
      "train_runtime": 4416.6897,
      "train_tokens_per_second": 626.826
    },
    {
      "epoch": 1.482183472327521,
      "grad_norm": 1.61728835105896,
      "learning_rate": 5.205063291139241e-06,
      "loss": 0.06809349358081818,
      "num_input_tokens_seen": 2770106,
      "step": 1467,
      "train_runtime": 4418.9009,
      "train_tokens_per_second": 626.877
    },
    {
      "epoch": 1.4831943391458176,
      "grad_norm": 1.3205126523971558,
      "learning_rate": 5.19493670886076e-06,
      "loss": 0.0746985524892807,
      "num_input_tokens_seen": 2771954,
      "step": 1468,
      "train_runtime": 4421.1396,
      "train_tokens_per_second": 626.977
    },
    {
      "epoch": 1.4842052059641142,
      "grad_norm": 1.513152003288269,
      "learning_rate": 5.184810126582279e-06,
      "loss": 0.04724295809864998,
      "num_input_tokens_seen": 2773760,
      "step": 1469,
      "train_runtime": 4423.3732,
      "train_tokens_per_second": 627.069
    },
    {
      "epoch": 1.4852160727824109,
      "grad_norm": 1.2082949876785278,
      "learning_rate": 5.174683544303798e-06,
      "loss": 0.06078195571899414,
      "num_input_tokens_seen": 2776132,
      "step": 1470,
      "train_runtime": 4425.7458,
      "train_tokens_per_second": 627.269
    },
    {
      "epoch": 1.4862269396007077,
      "grad_norm": 1.462839961051941,
      "learning_rate": 5.164556962025317e-06,
      "loss": 0.04663144052028656,
      "num_input_tokens_seen": 2777870,
      "step": 1471,
      "train_runtime": 4442.6069,
      "train_tokens_per_second": 625.279
    },
    {
      "epoch": 1.4872378064190044,
      "grad_norm": 1.2602648735046387,
      "learning_rate": 5.154430379746835e-06,
      "loss": 0.05627944692969322,
      "num_input_tokens_seen": 2779626,
      "step": 1472,
      "train_runtime": 4444.8361,
      "train_tokens_per_second": 625.361
    },
    {
      "epoch": 1.488248673237301,
      "grad_norm": 1.3976972103118896,
      "learning_rate": 5.144303797468355e-06,
      "loss": 0.057730481028556824,
      "num_input_tokens_seen": 2780798,
      "step": 1473,
      "train_runtime": 4446.9424,
      "train_tokens_per_second": 625.328
    },
    {
      "epoch": 1.4892595400555977,
      "grad_norm": 1.3106977939605713,
      "learning_rate": 5.134177215189874e-06,
      "loss": 0.0638170912861824,
      "num_input_tokens_seen": 2782748,
      "step": 1474,
      "train_runtime": 4449.1785,
      "train_tokens_per_second": 625.452
    },
    {
      "epoch": 1.4902704068738943,
      "grad_norm": 1.3717737197875977,
      "learning_rate": 5.1240506329113925e-06,
      "loss": 0.041115276515483856,
      "num_input_tokens_seen": 2784756,
      "step": 1475,
      "train_runtime": 4451.4857,
      "train_tokens_per_second": 625.579
    },
    {
      "epoch": 1.491281273692191,
      "grad_norm": 1.4171078205108643,
      "learning_rate": 5.1139240506329116e-06,
      "loss": 0.04597721993923187,
      "num_input_tokens_seen": 2786760,
      "step": 1476,
      "train_runtime": 4453.792,
      "train_tokens_per_second": 625.705
    },
    {
      "epoch": 1.4922921405104876,
      "grad_norm": 1.3031914234161377,
      "learning_rate": 5.1037974683544315e-06,
      "loss": 0.06663566827774048,
      "num_input_tokens_seen": 2788472,
      "step": 1477,
      "train_runtime": 4456.0151,
      "train_tokens_per_second": 625.777
    },
    {
      "epoch": 1.4933030073287845,
      "grad_norm": 1.0535391569137573,
      "learning_rate": 5.09367088607595e-06,
      "loss": 0.0422329418361187,
      "num_input_tokens_seen": 2790172,
      "step": 1478,
      "train_runtime": 4458.2598,
      "train_tokens_per_second": 625.843
    },
    {
      "epoch": 1.4943138741470812,
      "grad_norm": 1.1845626831054688,
      "learning_rate": 5.083544303797469e-06,
      "loss": 0.04153173416852951,
      "num_input_tokens_seen": 2791846,
      "step": 1479,
      "train_runtime": 4460.509,
      "train_tokens_per_second": 625.903
    },
    {
      "epoch": 1.4953247409653778,
      "grad_norm": 1.359662652015686,
      "learning_rate": 5.073417721518987e-06,
      "loss": 0.06811964511871338,
      "num_input_tokens_seen": 2793818,
      "step": 1480,
      "train_runtime": 4462.8299,
      "train_tokens_per_second": 626.019
    },
    {
      "epoch": 1.4963356077836745,
      "grad_norm": 1.278808355331421,
      "learning_rate": 5.063291139240507e-06,
      "loss": 0.051876235753297806,
      "num_input_tokens_seen": 2795626,
      "step": 1481,
      "train_runtime": 4465.0788,
      "train_tokens_per_second": 626.109
    },
    {
      "epoch": 1.4973464746019711,
      "grad_norm": 1.8733328580856323,
      "learning_rate": 5.053164556962026e-06,
      "loss": 0.0724148079752922,
      "num_input_tokens_seen": 2797286,
      "step": 1482,
      "train_runtime": 4467.2961,
      "train_tokens_per_second": 626.17
    },
    {
      "epoch": 1.498357341420268,
      "grad_norm": 1.9308044910430908,
      "learning_rate": 5.043037974683544e-06,
      "loss": 0.06310687959194183,
      "num_input_tokens_seen": 2798946,
      "step": 1483,
      "train_runtime": 4469.5211,
      "train_tokens_per_second": 626.23
    },
    {
      "epoch": 1.4993682082385646,
      "grad_norm": 1.4102615118026733,
      "learning_rate": 5.032911392405063e-06,
      "loss": 0.06878439337015152,
      "num_input_tokens_seen": 2800864,
      "step": 1484,
      "train_runtime": 4471.8112,
      "train_tokens_per_second": 626.338
    },
    {
      "epoch": 1.5003790750568613,
      "grad_norm": 1.3868674039840698,
      "learning_rate": 5.022784810126583e-06,
      "loss": 0.04760458692908287,
      "num_input_tokens_seen": 2802846,
      "step": 1485,
      "train_runtime": 4474.1019,
      "train_tokens_per_second": 626.46
    },
    {
      "epoch": 1.501389941875158,
      "grad_norm": 1.689648151397705,
      "learning_rate": 5.0126582278481016e-06,
      "loss": 0.08007767796516418,
      "num_input_tokens_seen": 2804606,
      "step": 1486,
      "train_runtime": 4476.3512,
      "train_tokens_per_second": 626.538
    },
    {
      "epoch": 1.5024008086934546,
      "grad_norm": 1.2327882051467896,
      "learning_rate": 5.002531645569621e-06,
      "loss": 0.047864578664302826,
      "num_input_tokens_seen": 2806254,
      "step": 1487,
      "train_runtime": 4478.5873,
      "train_tokens_per_second": 626.594
    },
    {
      "epoch": 1.5034116755117513,
      "grad_norm": 1.3961400985717773,
      "learning_rate": 4.99240506329114e-06,
      "loss": 0.057346735149621964,
      "num_input_tokens_seen": 2807812,
      "step": 1488,
      "train_runtime": 4480.7738,
      "train_tokens_per_second": 626.636
    },
    {
      "epoch": 1.504422542330048,
      "grad_norm": 1.029375433921814,
      "learning_rate": 4.982278481012659e-06,
      "loss": 0.04208916798233986,
      "num_input_tokens_seen": 2809866,
      "step": 1489,
      "train_runtime": 4483.0693,
      "train_tokens_per_second": 626.773
    },
    {
      "epoch": 1.5054334091483446,
      "grad_norm": 1.3852033615112305,
      "learning_rate": 4.972151898734178e-06,
      "loss": 0.05466679483652115,
      "num_input_tokens_seen": 2811744,
      "step": 1490,
      "train_runtime": 4485.3238,
      "train_tokens_per_second": 626.876
    },
    {
      "epoch": 1.5064442759666414,
      "grad_norm": 1.0315070152282715,
      "learning_rate": 4.962025316455696e-06,
      "loss": 0.05923735350370407,
      "num_input_tokens_seen": 2814384,
      "step": 1491,
      "train_runtime": 4487.7906,
      "train_tokens_per_second": 627.12
    },
    {
      "epoch": 1.507455142784938,
      "grad_norm": 1.108384132385254,
      "learning_rate": 4.951898734177216e-06,
      "loss": 0.05330324172973633,
      "num_input_tokens_seen": 2816184,
      "step": 1492,
      "train_runtime": 4490.0461,
      "train_tokens_per_second": 627.206
    },
    {
      "epoch": 1.5084660096032347,
      "grad_norm": 1.3703676462173462,
      "learning_rate": 4.941772151898734e-06,
      "loss": 0.05342656746506691,
      "num_input_tokens_seen": 2818180,
      "step": 1493,
      "train_runtime": 4492.3853,
      "train_tokens_per_second": 627.324
    },
    {
      "epoch": 1.5094768764215316,
      "grad_norm": 1.3506438732147217,
      "learning_rate": 4.931645569620253e-06,
      "loss": 0.05577048659324646,
      "num_input_tokens_seen": 2820102,
      "step": 1494,
      "train_runtime": 4494.6641,
      "train_tokens_per_second": 627.433
    },
    {
      "epoch": 1.5104877432398283,
      "grad_norm": 1.4416794776916504,
      "learning_rate": 4.9215189873417725e-06,
      "loss": 0.04947788268327713,
      "num_input_tokens_seen": 2821580,
      "step": 1495,
      "train_runtime": 4496.8705,
      "train_tokens_per_second": 627.454
    },
    {
      "epoch": 1.511498610058125,
      "grad_norm": 1.0802147388458252,
      "learning_rate": 4.9113924050632915e-06,
      "loss": 0.042999714612960815,
      "num_input_tokens_seen": 2824202,
      "step": 1496,
      "train_runtime": 4499.3361,
      "train_tokens_per_second": 627.693
    },
    {
      "epoch": 1.5125094768764216,
      "grad_norm": 1.066115379333496,
      "learning_rate": 4.901265822784811e-06,
      "loss": 0.05374641716480255,
      "num_input_tokens_seen": 2826432,
      "step": 1497,
      "train_runtime": 4501.7328,
      "train_tokens_per_second": 627.854
    },
    {
      "epoch": 1.5135203436947182,
      "grad_norm": 1.0133928060531616,
      "learning_rate": 4.89113924050633e-06,
      "loss": 0.04801242798566818,
      "num_input_tokens_seen": 2828690,
      "step": 1498,
      "train_runtime": 4504.1005,
      "train_tokens_per_second": 628.026
    },
    {
      "epoch": 1.5145312105130149,
      "grad_norm": 1.4185975790023804,
      "learning_rate": 4.881012658227848e-06,
      "loss": 0.053610898554325104,
      "num_input_tokens_seen": 2830248,
      "step": 1499,
      "train_runtime": 4506.3101,
      "train_tokens_per_second": 628.063
    },
    {
      "epoch": 1.5155420773313115,
      "grad_norm": 1.0259369611740112,
      "learning_rate": 4.870886075949368e-06,
      "loss": 0.04819566756486893,
      "num_input_tokens_seen": 2832522,
      "step": 1500,
      "train_runtime": 4508.6847,
      "train_tokens_per_second": 628.237
    },
    {
      "epoch": 1.5165529441496082,
      "grad_norm": 1.22133469581604,
      "learning_rate": 4.860759493670886e-06,
      "loss": 0.05674818158149719,
      "num_input_tokens_seen": 2834288,
      "step": 1501,
      "train_runtime": 4524.4207,
      "train_tokens_per_second": 626.442
    },
    {
      "epoch": 1.5175638109679048,
      "grad_norm": 1.2766376733779907,
      "learning_rate": 4.850632911392405e-06,
      "loss": 0.05139131098985672,
      "num_input_tokens_seen": 2836114,
      "step": 1502,
      "train_runtime": 4526.8123,
      "train_tokens_per_second": 626.515
    },
    {
      "epoch": 1.5185746777862017,
      "grad_norm": 1.3432129621505737,
      "learning_rate": 4.840506329113924e-06,
      "loss": 0.06788015365600586,
      "num_input_tokens_seen": 2838072,
      "step": 1503,
      "train_runtime": 4529.1766,
      "train_tokens_per_second": 626.62
    },
    {
      "epoch": 1.5195855446044984,
      "grad_norm": 1.4195806980133057,
      "learning_rate": 4.830379746835443e-06,
      "loss": 0.049872562289237976,
      "num_input_tokens_seen": 2840010,
      "step": 1504,
      "train_runtime": 4531.4606,
      "train_tokens_per_second": 626.732
    },
    {
      "epoch": 1.520596411422795,
      "grad_norm": 1.5526108741760254,
      "learning_rate": 4.8202531645569625e-06,
      "loss": 0.04352928698062897,
      "num_input_tokens_seen": 2841990,
      "step": 1505,
      "train_runtime": 4533.7361,
      "train_tokens_per_second": 626.854
    },
    {
      "epoch": 1.5216072782410919,
      "grad_norm": 1.3962048292160034,
      "learning_rate": 4.8101265822784815e-06,
      "loss": 0.05957988277077675,
      "num_input_tokens_seen": 2843498,
      "step": 1506,
      "train_runtime": 4535.8841,
      "train_tokens_per_second": 626.889
    },
    {
      "epoch": 1.5226181450593885,
      "grad_norm": 1.6576125621795654,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.07533112168312073,
      "num_input_tokens_seen": 2845090,
      "step": 1507,
      "train_runtime": 4538.0583,
      "train_tokens_per_second": 626.94
    },
    {
      "epoch": 1.5236290118776852,
      "grad_norm": 1.7079765796661377,
      "learning_rate": 4.78987341772152e-06,
      "loss": 0.0601004958152771,
      "num_input_tokens_seen": 2846700,
      "step": 1508,
      "train_runtime": 4540.2674,
      "train_tokens_per_second": 626.989
    },
    {
      "epoch": 1.5246398786959818,
      "grad_norm": 1.2272881269454956,
      "learning_rate": 4.779746835443038e-06,
      "loss": 0.04607832804322243,
      "num_input_tokens_seen": 2848114,
      "step": 1509,
      "train_runtime": 4542.4244,
      "train_tokens_per_second": 627.003
    },
    {
      "epoch": 1.5256507455142785,
      "grad_norm": 1.4930332899093628,
      "learning_rate": 4.769620253164557e-06,
      "loss": 0.07360376417636871,
      "num_input_tokens_seen": 2849846,
      "step": 1510,
      "train_runtime": 4544.6459,
      "train_tokens_per_second": 627.078
    },
    {
      "epoch": 1.5266616123325751,
      "grad_norm": 1.4419664144515991,
      "learning_rate": 4.759493670886076e-06,
      "loss": 0.06184345483779907,
      "num_input_tokens_seen": 2851468,
      "step": 1511,
      "train_runtime": 4546.8862,
      "train_tokens_per_second": 627.125
    },
    {
      "epoch": 1.5276724791508718,
      "grad_norm": 1.348374605178833,
      "learning_rate": 4.749367088607595e-06,
      "loss": 0.06621316820383072,
      "num_input_tokens_seen": 2853370,
      "step": 1512,
      "train_runtime": 4549.1738,
      "train_tokens_per_second": 627.228
    },
    {
      "epoch": 1.5286833459691684,
      "grad_norm": 1.2672215700149536,
      "learning_rate": 4.739240506329114e-06,
      "loss": 0.07050938159227371,
      "num_input_tokens_seen": 2855182,
      "step": 1513,
      "train_runtime": 4551.4166,
      "train_tokens_per_second": 627.317
    },
    {
      "epoch": 1.529694212787465,
      "grad_norm": 1.4655346870422363,
      "learning_rate": 4.729113924050633e-06,
      "loss": 0.06721789389848709,
      "num_input_tokens_seen": 2857214,
      "step": 1514,
      "train_runtime": 4553.7647,
      "train_tokens_per_second": 627.44
    },
    {
      "epoch": 1.530705079605762,
      "grad_norm": 1.517762541770935,
      "learning_rate": 4.7189873417721525e-06,
      "loss": 0.05682816356420517,
      "num_input_tokens_seen": 2859046,
      "step": 1515,
      "train_runtime": 4556.0681,
      "train_tokens_per_second": 627.525
    },
    {
      "epoch": 1.5317159464240586,
      "grad_norm": 1.3225510120391846,
      "learning_rate": 4.7088607594936715e-06,
      "loss": 0.05826298147439957,
      "num_input_tokens_seen": 2861014,
      "step": 1516,
      "train_runtime": 4558.4249,
      "train_tokens_per_second": 627.632
    },
    {
      "epoch": 1.5327268132423553,
      "grad_norm": 1.1289844512939453,
      "learning_rate": 4.69873417721519e-06,
      "loss": 0.0564601793885231,
      "num_input_tokens_seen": 2862968,
      "step": 1517,
      "train_runtime": 4560.7052,
      "train_tokens_per_second": 627.747
    },
    {
      "epoch": 1.5337376800606521,
      "grad_norm": 1.1863211393356323,
      "learning_rate": 4.68860759493671e-06,
      "loss": 0.03920586034655571,
      "num_input_tokens_seen": 2864930,
      "step": 1518,
      "train_runtime": 4562.9942,
      "train_tokens_per_second": 627.862
    },
    {
      "epoch": 1.5347485468789488,
      "grad_norm": 1.074202537536621,
      "learning_rate": 4.678481012658228e-06,
      "loss": 0.04804112762212753,
      "num_input_tokens_seen": 2867000,
      "step": 1519,
      "train_runtime": 4565.318,
      "train_tokens_per_second": 627.996
    },
    {
      "epoch": 1.5357594136972454,
      "grad_norm": 1.1481704711914062,
      "learning_rate": 4.668354430379747e-06,
      "loss": 0.040212418884038925,
      "num_input_tokens_seen": 2869152,
      "step": 1520,
      "train_runtime": 4567.6733,
      "train_tokens_per_second": 628.143
    },
    {
      "epoch": 1.536770280515542,
      "grad_norm": 1.4831442832946777,
      "learning_rate": 4.658227848101266e-06,
      "loss": 0.06600040942430496,
      "num_input_tokens_seen": 2870778,
      "step": 1521,
      "train_runtime": 4569.8853,
      "train_tokens_per_second": 628.195
    },
    {
      "epoch": 1.5377811473338387,
      "grad_norm": 1.2420631647109985,
      "learning_rate": 4.648101265822785e-06,
      "loss": 0.05727258324623108,
      "num_input_tokens_seen": 2872896,
      "step": 1522,
      "train_runtime": 4572.2168,
      "train_tokens_per_second": 628.338
    },
    {
      "epoch": 1.5387920141521354,
      "grad_norm": 1.1956125497817993,
      "learning_rate": 4.637974683544304e-06,
      "loss": 0.05453263595700264,
      "num_input_tokens_seen": 2874902,
      "step": 1523,
      "train_runtime": 4574.5244,
      "train_tokens_per_second": 628.459
    },
    {
      "epoch": 1.539802880970432,
      "grad_norm": 1.2424403429031372,
      "learning_rate": 4.627848101265823e-06,
      "loss": 0.049895938485860825,
      "num_input_tokens_seen": 2876462,
      "step": 1524,
      "train_runtime": 4576.7088,
      "train_tokens_per_second": 628.5
    },
    {
      "epoch": 1.5408137477887287,
      "grad_norm": 1.2138495445251465,
      "learning_rate": 4.617721518987342e-06,
      "loss": 0.06158655881881714,
      "num_input_tokens_seen": 2878202,
      "step": 1525,
      "train_runtime": 4578.9598,
      "train_tokens_per_second": 628.571
    },
    {
      "epoch": 1.5418246146070256,
      "grad_norm": 1.1184782981872559,
      "learning_rate": 4.6075949367088615e-06,
      "loss": 0.05643635615706444,
      "num_input_tokens_seen": 2880316,
      "step": 1526,
      "train_runtime": 4581.2901,
      "train_tokens_per_second": 628.713
    },
    {
      "epoch": 1.5428354814253222,
      "grad_norm": 1.4851226806640625,
      "learning_rate": 4.59746835443038e-06,
      "loss": 0.06179232895374298,
      "num_input_tokens_seen": 2881992,
      "step": 1527,
      "train_runtime": 4583.5051,
      "train_tokens_per_second": 628.775
    },
    {
      "epoch": 1.5438463482436189,
      "grad_norm": 0.94132000207901,
      "learning_rate": 4.587341772151899e-06,
      "loss": 0.03961401805281639,
      "num_input_tokens_seen": 2883762,
      "step": 1528,
      "train_runtime": 4585.7925,
      "train_tokens_per_second": 628.847
    },
    {
      "epoch": 1.5448572150619158,
      "grad_norm": 1.4120358228683472,
      "learning_rate": 4.577215189873418e-06,
      "loss": 0.0553383007645607,
      "num_input_tokens_seen": 2886030,
      "step": 1529,
      "train_runtime": 4588.1973,
      "train_tokens_per_second": 629.012
    },
    {
      "epoch": 1.5458680818802124,
      "grad_norm": 1.3151836395263672,
      "learning_rate": 4.567088607594937e-06,
      "loss": 0.05081061273813248,
      "num_input_tokens_seen": 2887852,
      "step": 1530,
      "train_runtime": 4590.4695,
      "train_tokens_per_second": 629.097
    },
    {
      "epoch": 1.546878948698509,
      "grad_norm": 1.3867336511611938,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.06330150365829468,
      "num_input_tokens_seen": 2889610,
      "step": 1531,
      "train_runtime": 4606.0312,
      "train_tokens_per_second": 627.354
    },
    {
      "epoch": 1.5478898155168057,
      "grad_norm": 1.3176735639572144,
      "learning_rate": 4.546835443037975e-06,
      "loss": 0.06053399667143822,
      "num_input_tokens_seen": 2891706,
      "step": 1532,
      "train_runtime": 4608.3281,
      "train_tokens_per_second": 627.496
    },
    {
      "epoch": 1.5489006823351024,
      "grad_norm": 1.4362438917160034,
      "learning_rate": 4.536708860759494e-06,
      "loss": 0.07300815731287003,
      "num_input_tokens_seen": 2893848,
      "step": 1533,
      "train_runtime": 4610.708,
      "train_tokens_per_second": 627.636
    },
    {
      "epoch": 1.549911549153399,
      "grad_norm": 1.0981767177581787,
      "learning_rate": 4.526582278481013e-06,
      "loss": 0.03816743940114975,
      "num_input_tokens_seen": 2895582,
      "step": 1534,
      "train_runtime": 4612.9444,
      "train_tokens_per_second": 627.708
    },
    {
      "epoch": 1.5509224159716957,
      "grad_norm": 1.4019081592559814,
      "learning_rate": 4.516455696202532e-06,
      "loss": 0.050167202949523926,
      "num_input_tokens_seen": 2897422,
      "step": 1535,
      "train_runtime": 4615.1556,
      "train_tokens_per_second": 627.806
    },
    {
      "epoch": 1.5519332827899923,
      "grad_norm": 1.178011178970337,
      "learning_rate": 4.5063291139240515e-06,
      "loss": 0.033923763781785965,
      "num_input_tokens_seen": 2898578,
      "step": 1536,
      "train_runtime": 4617.2118,
      "train_tokens_per_second": 627.777
    },
    {
      "epoch": 1.552944149608289,
      "grad_norm": 0.9976033568382263,
      "learning_rate": 4.49620253164557e-06,
      "loss": 0.051811426877975464,
      "num_input_tokens_seen": 2900830,
      "step": 1537,
      "train_runtime": 4619.5337,
      "train_tokens_per_second": 627.949
    },
    {
      "epoch": 1.5539550164265858,
      "grad_norm": 1.2337932586669922,
      "learning_rate": 4.486075949367089e-06,
      "loss": 0.05312459170818329,
      "num_input_tokens_seen": 2902758,
      "step": 1538,
      "train_runtime": 4621.804,
      "train_tokens_per_second": 628.057
    },
    {
      "epoch": 1.5549658832448825,
      "grad_norm": 1.2490909099578857,
      "learning_rate": 4.475949367088608e-06,
      "loss": 0.0594743937253952,
      "num_input_tokens_seen": 2904738,
      "step": 1539,
      "train_runtime": 4624.0593,
      "train_tokens_per_second": 628.179
    },
    {
      "epoch": 1.5559767500631791,
      "grad_norm": 1.274011492729187,
      "learning_rate": 4.465822784810127e-06,
      "loss": 0.06778080761432648,
      "num_input_tokens_seen": 2907096,
      "step": 1540,
      "train_runtime": 4626.4602,
      "train_tokens_per_second": 628.363
    },
    {
      "epoch": 1.556987616881476,
      "grad_norm": 1.357280969619751,
      "learning_rate": 4.455696202531646e-06,
      "loss": 0.06358253955841064,
      "num_input_tokens_seen": 2908836,
      "step": 1541,
      "train_runtime": 4628.8142,
      "train_tokens_per_second": 628.419
    },
    {
      "epoch": 1.5579984836997727,
      "grad_norm": 1.592889428138733,
      "learning_rate": 4.445569620253165e-06,
      "loss": 0.05604672431945801,
      "num_input_tokens_seen": 2910510,
      "step": 1542,
      "train_runtime": 4631.0341,
      "train_tokens_per_second": 628.48
    },
    {
      "epoch": 1.5590093505180693,
      "grad_norm": 0.9576203227043152,
      "learning_rate": 4.4354430379746834e-06,
      "loss": 0.03890995681285858,
      "num_input_tokens_seen": 2912316,
      "step": 1543,
      "train_runtime": 4633.265,
      "train_tokens_per_second": 628.567
    },
    {
      "epoch": 1.560020217336366,
      "grad_norm": 1.5377309322357178,
      "learning_rate": 4.425316455696203e-06,
      "loss": 0.0765279158949852,
      "num_input_tokens_seen": 2914070,
      "step": 1544,
      "train_runtime": 4635.5036,
      "train_tokens_per_second": 628.642
    },
    {
      "epoch": 1.5610310841546626,
      "grad_norm": 1.701785922050476,
      "learning_rate": 4.415189873417722e-06,
      "loss": 0.0848785936832428,
      "num_input_tokens_seen": 2916386,
      "step": 1545,
      "train_runtime": 4637.8598,
      "train_tokens_per_second": 628.822
    },
    {
      "epoch": 1.5620419509729593,
      "grad_norm": 1.2688883543014526,
      "learning_rate": 4.405063291139241e-06,
      "loss": 0.060875337570905685,
      "num_input_tokens_seen": 2918148,
      "step": 1546,
      "train_runtime": 4640.1228,
      "train_tokens_per_second": 628.895
    },
    {
      "epoch": 1.563052817791256,
      "grad_norm": 1.1968705654144287,
      "learning_rate": 4.39493670886076e-06,
      "loss": 0.0575110949575901,
      "num_input_tokens_seen": 2920642,
      "step": 1547,
      "train_runtime": 4642.5702,
      "train_tokens_per_second": 629.1
    },
    {
      "epoch": 1.5640636846095526,
      "grad_norm": 1.4697033166885376,
      "learning_rate": 4.384810126582279e-06,
      "loss": 0.06372949481010437,
      "num_input_tokens_seen": 2922254,
      "step": 1548,
      "train_runtime": 4644.8072,
      "train_tokens_per_second": 629.144
    },
    {
      "epoch": 1.5650745514278492,
      "grad_norm": 1.3741692304611206,
      "learning_rate": 4.374683544303798e-06,
      "loss": 0.04079161584377289,
      "num_input_tokens_seen": 2924156,
      "step": 1549,
      "train_runtime": 4647.0781,
      "train_tokens_per_second": 629.246
    },
    {
      "epoch": 1.566085418246146,
      "grad_norm": 1.4274075031280518,
      "learning_rate": 4.364556962025317e-06,
      "loss": 0.07402218133211136,
      "num_input_tokens_seen": 2926290,
      "step": 1550,
      "train_runtime": 4649.4117,
      "train_tokens_per_second": 629.389
    },
    {
      "epoch": 1.5670962850644428,
      "grad_norm": 1.2086807489395142,
      "learning_rate": 4.354430379746836e-06,
      "loss": 0.05297299474477768,
      "num_input_tokens_seen": 2928344,
      "step": 1551,
      "train_runtime": 4651.73,
      "train_tokens_per_second": 629.517
    },
    {
      "epoch": 1.5681071518827394,
      "grad_norm": 0.9295864105224609,
      "learning_rate": 4.344303797468355e-06,
      "loss": 0.032322127372026443,
      "num_input_tokens_seen": 2930198,
      "step": 1552,
      "train_runtime": 4654.0126,
      "train_tokens_per_second": 629.607
    },
    {
      "epoch": 1.5691180187010363,
      "grad_norm": 1.0612962245941162,
      "learning_rate": 4.3341772151898734e-06,
      "loss": 0.03210993483662605,
      "num_input_tokens_seen": 2931694,
      "step": 1553,
      "train_runtime": 4656.1517,
      "train_tokens_per_second": 629.639
    },
    {
      "epoch": 1.570128885519333,
      "grad_norm": 1.5804210901260376,
      "learning_rate": 4.3240506329113925e-06,
      "loss": 0.07168054580688477,
      "num_input_tokens_seen": 2933774,
      "step": 1554,
      "train_runtime": 4658.5312,
      "train_tokens_per_second": 629.764
    },
    {
      "epoch": 1.5711397523376296,
      "grad_norm": 1.3794080018997192,
      "learning_rate": 4.313924050632912e-06,
      "loss": 0.06514356285333633,
      "num_input_tokens_seen": 2935584,
      "step": 1555,
      "train_runtime": 4660.7964,
      "train_tokens_per_second": 629.846
    },
    {
      "epoch": 1.5721506191559262,
      "grad_norm": 1.4102269411087036,
      "learning_rate": 4.303797468354431e-06,
      "loss": 0.05319744348526001,
      "num_input_tokens_seen": 2937110,
      "step": 1556,
      "train_runtime": 4663.002,
      "train_tokens_per_second": 629.875
    },
    {
      "epoch": 1.573161485974223,
      "grad_norm": 1.2388628721237183,
      "learning_rate": 4.29367088607595e-06,
      "loss": 0.04378683865070343,
      "num_input_tokens_seen": 2938708,
      "step": 1557,
      "train_runtime": 4665.1951,
      "train_tokens_per_second": 629.922
    },
    {
      "epoch": 1.5741723527925195,
      "grad_norm": 1.3174517154693604,
      "learning_rate": 4.283544303797469e-06,
      "loss": 0.06327192485332489,
      "num_input_tokens_seen": 2941278,
      "step": 1558,
      "train_runtime": 4667.7011,
      "train_tokens_per_second": 630.134
    },
    {
      "epoch": 1.5751832196108162,
      "grad_norm": 1.4438049793243408,
      "learning_rate": 4.273417721518988e-06,
      "loss": 0.06322846561670303,
      "num_input_tokens_seen": 2942738,
      "step": 1559,
      "train_runtime": 4669.8447,
      "train_tokens_per_second": 630.158
    },
    {
      "epoch": 1.5761940864291129,
      "grad_norm": 1.2885171175003052,
      "learning_rate": 4.263291139240507e-06,
      "loss": 0.05553532764315605,
      "num_input_tokens_seen": 2944420,
      "step": 1560,
      "train_runtime": 4672.0637,
      "train_tokens_per_second": 630.218
    },
    {
      "epoch": 1.5772049532474095,
      "grad_norm": 1.6389775276184082,
      "learning_rate": 4.253164556962025e-06,
      "loss": 0.06942377984523773,
      "num_input_tokens_seen": 2946336,
      "step": 1561,
      "train_runtime": 4687.7107,
      "train_tokens_per_second": 628.523
    },
    {
      "epoch": 1.5782158200657064,
      "grad_norm": 1.2550915479660034,
      "learning_rate": 4.243037974683545e-06,
      "loss": 0.06170507147908211,
      "num_input_tokens_seen": 2947836,
      "step": 1562,
      "train_runtime": 4689.8812,
      "train_tokens_per_second": 628.552
    },
    {
      "epoch": 1.579226686884003,
      "grad_norm": 1.1916841268539429,
      "learning_rate": 4.232911392405063e-06,
      "loss": 0.055288903415203094,
      "num_input_tokens_seen": 2950112,
      "step": 1563,
      "train_runtime": 4692.2551,
      "train_tokens_per_second": 628.719
    },
    {
      "epoch": 1.5802375537022997,
      "grad_norm": 1.5838149785995483,
      "learning_rate": 4.2227848101265825e-06,
      "loss": 0.0642075315117836,
      "num_input_tokens_seen": 2952126,
      "step": 1564,
      "train_runtime": 4694.5231,
      "train_tokens_per_second": 628.845
    },
    {
      "epoch": 1.5812484205205966,
      "grad_norm": 1.3996124267578125,
      "learning_rate": 4.212658227848102e-06,
      "loss": 0.057322438806295395,
      "num_input_tokens_seen": 2953936,
      "step": 1565,
      "train_runtime": 4696.7953,
      "train_tokens_per_second": 628.926
    },
    {
      "epoch": 1.5822592873388932,
      "grad_norm": 1.3723688125610352,
      "learning_rate": 4.202531645569621e-06,
      "loss": 0.061545997858047485,
      "num_input_tokens_seen": 2955706,
      "step": 1566,
      "train_runtime": 4699.0361,
      "train_tokens_per_second": 629.003
    },
    {
      "epoch": 1.5832701541571899,
      "grad_norm": 1.5268528461456299,
      "learning_rate": 4.19240506329114e-06,
      "loss": 0.06614181399345398,
      "num_input_tokens_seen": 2957292,
      "step": 1567,
      "train_runtime": 4701.2707,
      "train_tokens_per_second": 629.041
    },
    {
      "epoch": 1.5842810209754865,
      "grad_norm": 1.491364598274231,
      "learning_rate": 4.182278481012659e-06,
      "loss": 0.06419438868761063,
      "num_input_tokens_seen": 2959516,
      "step": 1568,
      "train_runtime": 4703.6141,
      "train_tokens_per_second": 629.2
    },
    {
      "epoch": 1.5852918877937832,
      "grad_norm": 1.192054271697998,
      "learning_rate": 4.172151898734177e-06,
      "loss": 0.04947643727064133,
      "num_input_tokens_seen": 2961496,
      "step": 1569,
      "train_runtime": 4705.9071,
      "train_tokens_per_second": 629.315
    },
    {
      "epoch": 1.5863027546120798,
      "grad_norm": 1.2916615009307861,
      "learning_rate": 4.162025316455697e-06,
      "loss": 0.046376898884773254,
      "num_input_tokens_seen": 2963170,
      "step": 1570,
      "train_runtime": 4708.1188,
      "train_tokens_per_second": 629.375
    },
    {
      "epoch": 1.5873136214303765,
      "grad_norm": 1.3182624578475952,
      "learning_rate": 4.151898734177215e-06,
      "loss": 0.055604659020900726,
      "num_input_tokens_seen": 2965072,
      "step": 1571,
      "train_runtime": 4710.4542,
      "train_tokens_per_second": 629.466
    },
    {
      "epoch": 1.5883244882486731,
      "grad_norm": 1.2492234706878662,
      "learning_rate": 4.141772151898734e-06,
      "loss": 0.05962325632572174,
      "num_input_tokens_seen": 2967136,
      "step": 1572,
      "train_runtime": 4712.8342,
      "train_tokens_per_second": 629.586
    },
    {
      "epoch": 1.5893353550669698,
      "grad_norm": 1.1575913429260254,
      "learning_rate": 4.131645569620253e-06,
      "loss": 0.04633761942386627,
      "num_input_tokens_seen": 2968980,
      "step": 1573,
      "train_runtime": 4715.0822,
      "train_tokens_per_second": 629.677
    },
    {
      "epoch": 1.5903462218852666,
      "grad_norm": 1.2850408554077148,
      "learning_rate": 4.1215189873417725e-06,
      "loss": 0.056543268263339996,
      "num_input_tokens_seen": 2970630,
      "step": 1574,
      "train_runtime": 4717.2841,
      "train_tokens_per_second": 629.733
    },
    {
      "epoch": 1.5913570887035633,
      "grad_norm": 1.213210105895996,
      "learning_rate": 4.111392405063292e-06,
      "loss": 0.05120829492807388,
      "num_input_tokens_seen": 2972686,
      "step": 1575,
      "train_runtime": 4719.584,
      "train_tokens_per_second": 629.862
    },
    {
      "epoch": 1.59236795552186,
      "grad_norm": 1.2734119892120361,
      "learning_rate": 4.101265822784811e-06,
      "loss": 0.05734711512923241,
      "num_input_tokens_seen": 2974526,
      "step": 1576,
      "train_runtime": 4721.8334,
      "train_tokens_per_second": 629.951
    },
    {
      "epoch": 1.5933788223401568,
      "grad_norm": 1.0458225011825562,
      "learning_rate": 4.09113924050633e-06,
      "loss": 0.0468803271651268,
      "num_input_tokens_seen": 2976356,
      "step": 1577,
      "train_runtime": 4724.1274,
      "train_tokens_per_second": 630.033
    },
    {
      "epoch": 1.5943896891584535,
      "grad_norm": 1.3595490455627441,
      "learning_rate": 4.081012658227849e-06,
      "loss": 0.057377636432647705,
      "num_input_tokens_seen": 2977998,
      "step": 1578,
      "train_runtime": 4726.3364,
      "train_tokens_per_second": 630.086
    },
    {
      "epoch": 1.5954005559767501,
      "grad_norm": 1.2467687129974365,
      "learning_rate": 4.070886075949367e-06,
      "loss": 0.054508354514837265,
      "num_input_tokens_seen": 2979636,
      "step": 1579,
      "train_runtime": 4728.5291,
      "train_tokens_per_second": 630.14
    },
    {
      "epoch": 1.5964114227950468,
      "grad_norm": 1.2042741775512695,
      "learning_rate": 4.060759493670887e-06,
      "loss": 0.059663258492946625,
      "num_input_tokens_seen": 2981328,
      "step": 1580,
      "train_runtime": 4730.7443,
      "train_tokens_per_second": 630.203
    },
    {
      "epoch": 1.5974222896133434,
      "grad_norm": 1.2732118368148804,
      "learning_rate": 4.050632911392405e-06,
      "loss": 0.06218750402331352,
      "num_input_tokens_seen": 2982932,
      "step": 1581,
      "train_runtime": 4732.9191,
      "train_tokens_per_second": 630.252
    },
    {
      "epoch": 1.59843315643164,
      "grad_norm": 1.5464329719543457,
      "learning_rate": 4.040506329113924e-06,
      "loss": 0.06518936902284622,
      "num_input_tokens_seen": 2985058,
      "step": 1582,
      "train_runtime": 4735.2542,
      "train_tokens_per_second": 630.39
    },
    {
      "epoch": 1.5994440232499367,
      "grad_norm": 1.7101939916610718,
      "learning_rate": 4.030379746835443e-06,
      "loss": 0.04698918014764786,
      "num_input_tokens_seen": 2987304,
      "step": 1583,
      "train_runtime": 4737.5879,
      "train_tokens_per_second": 630.554
    },
    {
      "epoch": 1.6004548900682334,
      "grad_norm": 1.2198654413223267,
      "learning_rate": 4.0202531645569625e-06,
      "loss": 0.06318310648202896,
      "num_input_tokens_seen": 2989272,
      "step": 1584,
      "train_runtime": 4739.8983,
      "train_tokens_per_second": 630.662
    },
    {
      "epoch": 1.6004548900682334,
      "eval_loss": 0.3663329482078552,
      "eval_runtime": 54.3212,
      "eval_samples_per_second": 16.182,
      "eval_steps_per_second": 8.1,
      "num_input_tokens_seen": 2989272,
      "step": 1584
    },
    {
      "epoch": 1.60146575688653,
      "grad_norm": 1.1672413349151611,
      "learning_rate": 4.010126582278482e-06,
      "loss": 0.03929019346833229,
      "num_input_tokens_seen": 2991142,
      "step": 1585,
      "train_runtime": 4796.5172,
      "train_tokens_per_second": 623.607
    },
    {
      "epoch": 1.602476623704827,
      "grad_norm": 1.689853310585022,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.058579038828611374,
      "num_input_tokens_seen": 2992704,
      "step": 1586,
      "train_runtime": 4798.7011,
      "train_tokens_per_second": 623.649
    },
    {
      "epoch": 1.6034874905231236,
      "grad_norm": 1.5296273231506348,
      "learning_rate": 3.989873417721519e-06,
      "loss": 0.055761225521564484,
      "num_input_tokens_seen": 2994422,
      "step": 1587,
      "train_runtime": 4800.9454,
      "train_tokens_per_second": 623.715
    },
    {
      "epoch": 1.6044983573414204,
      "grad_norm": 1.5676381587982178,
      "learning_rate": 3.979746835443039e-06,
      "loss": 0.07547661662101746,
      "num_input_tokens_seen": 2996456,
      "step": 1588,
      "train_runtime": 4803.2516,
      "train_tokens_per_second": 623.839
    },
    {
      "epoch": 1.605509224159717,
      "grad_norm": 1.5270965099334717,
      "learning_rate": 3.969620253164557e-06,
      "loss": 0.042157530784606934,
      "num_input_tokens_seen": 2998580,
      "step": 1589,
      "train_runtime": 4805.5859,
      "train_tokens_per_second": 623.978
    },
    {
      "epoch": 1.6065200909780137,
      "grad_norm": 1.384373664855957,
      "learning_rate": 3.959493670886076e-06,
      "loss": 0.051778409630060196,
      "num_input_tokens_seen": 3000296,
      "step": 1590,
      "train_runtime": 4807.8296,
      "train_tokens_per_second": 624.044
    },
    {
      "epoch": 1.6075309577963104,
      "grad_norm": 1.4385806322097778,
      "learning_rate": 3.949367088607595e-06,
      "loss": 0.05921599268913269,
      "num_input_tokens_seen": 3002560,
      "step": 1591,
      "train_runtime": 4823.6043,
      "train_tokens_per_second": 622.472
    },
    {
      "epoch": 1.608541824614607,
      "grad_norm": 1.31643545627594,
      "learning_rate": 3.939240506329114e-06,
      "loss": 0.045938633382320404,
      "num_input_tokens_seen": 3004104,
      "step": 1592,
      "train_runtime": 4825.776,
      "train_tokens_per_second": 622.512
    },
    {
      "epoch": 1.6095526914329037,
      "grad_norm": 1.273729681968689,
      "learning_rate": 3.929113924050633e-06,
      "loss": 0.04683920741081238,
      "num_input_tokens_seen": 3005420,
      "step": 1593,
      "train_runtime": 4827.892,
      "train_tokens_per_second": 622.512
    },
    {
      "epoch": 1.6105635582512003,
      "grad_norm": 1.8324092626571655,
      "learning_rate": 3.9189873417721525e-06,
      "loss": 0.10185383260250092,
      "num_input_tokens_seen": 3006926,
      "step": 1594,
      "train_runtime": 4830.0627,
      "train_tokens_per_second": 622.544
    },
    {
      "epoch": 1.611574425069497,
      "grad_norm": 1.9990055561065674,
      "learning_rate": 3.908860759493672e-06,
      "loss": 0.051357023417949677,
      "num_input_tokens_seen": 3008578,
      "step": 1595,
      "train_runtime": 4832.2417,
      "train_tokens_per_second": 622.605
    },
    {
      "epoch": 1.6125852918877936,
      "grad_norm": 1.311766266822815,
      "learning_rate": 3.898734177215191e-06,
      "loss": 0.06318081170320511,
      "num_input_tokens_seen": 3010530,
      "step": 1596,
      "train_runtime": 4834.5102,
      "train_tokens_per_second": 622.717
    },
    {
      "epoch": 1.6135961587060905,
      "grad_norm": 1.0897035598754883,
      "learning_rate": 3.888607594936709e-06,
      "loss": 0.05656641721725464,
      "num_input_tokens_seen": 3012808,
      "step": 1597,
      "train_runtime": 4836.8865,
      "train_tokens_per_second": 622.882
    },
    {
      "epoch": 1.6146070255243872,
      "grad_norm": 1.5967590808868408,
      "learning_rate": 3.878481012658228e-06,
      "loss": 0.07358765602111816,
      "num_input_tokens_seen": 3014468,
      "step": 1598,
      "train_runtime": 4839.0885,
      "train_tokens_per_second": 622.941
    },
    {
      "epoch": 1.6156178923426838,
      "grad_norm": 1.463265299797058,
      "learning_rate": 3.868354430379747e-06,
      "loss": 0.06203243136405945,
      "num_input_tokens_seen": 3015760,
      "step": 1599,
      "train_runtime": 4841.2142,
      "train_tokens_per_second": 622.935
    },
    {
      "epoch": 1.6166287591609807,
      "grad_norm": 1.2703170776367188,
      "learning_rate": 3.858227848101266e-06,
      "loss": 0.0503770112991333,
      "num_input_tokens_seen": 3017286,
      "step": 1600,
      "train_runtime": 4843.3962,
      "train_tokens_per_second": 622.969
    },
    {
      "epoch": 1.6176396259792774,
      "grad_norm": 1.5702494382858276,
      "learning_rate": 3.848101265822785e-06,
      "loss": 0.08263242244720459,
      "num_input_tokens_seen": 3019240,
      "step": 1601,
      "train_runtime": 4845.8433,
      "train_tokens_per_second": 623.058
    },
    {
      "epoch": 1.618650492797574,
      "grad_norm": 1.310970425605774,
      "learning_rate": 3.837974683544304e-06,
      "loss": 0.06051701307296753,
      "num_input_tokens_seen": 3021418,
      "step": 1602,
      "train_runtime": 4848.2543,
      "train_tokens_per_second": 623.197
    },
    {
      "epoch": 1.6196613596158707,
      "grad_norm": 1.295046091079712,
      "learning_rate": 3.827848101265823e-06,
      "loss": 0.05538259446620941,
      "num_input_tokens_seen": 3023596,
      "step": 1603,
      "train_runtime": 4850.5894,
      "train_tokens_per_second": 623.346
    },
    {
      "epoch": 1.6206722264341673,
      "grad_norm": 1.1572978496551514,
      "learning_rate": 3.8177215189873425e-06,
      "loss": 0.05210157111287117,
      "num_input_tokens_seen": 3025522,
      "step": 1604,
      "train_runtime": 4852.8945,
      "train_tokens_per_second": 623.447
    },
    {
      "epoch": 1.621683093252464,
      "grad_norm": 1.2352598905563354,
      "learning_rate": 3.807594936708861e-06,
      "loss": 0.057181812822818756,
      "num_input_tokens_seen": 3027840,
      "step": 1605,
      "train_runtime": 4855.2586,
      "train_tokens_per_second": 623.621
    },
    {
      "epoch": 1.6226939600707606,
      "grad_norm": 1.310869574546814,
      "learning_rate": 3.7974683544303802e-06,
      "loss": 0.06201940029859543,
      "num_input_tokens_seen": 3029608,
      "step": 1606,
      "train_runtime": 4857.5402,
      "train_tokens_per_second": 623.692
    },
    {
      "epoch": 1.6237048268890573,
      "grad_norm": 1.218132495880127,
      "learning_rate": 3.787341772151899e-06,
      "loss": 0.04639403522014618,
      "num_input_tokens_seen": 3031396,
      "step": 1607,
      "train_runtime": 4859.7466,
      "train_tokens_per_second": 623.777
    },
    {
      "epoch": 1.624715693707354,
      "grad_norm": 1.306732177734375,
      "learning_rate": 3.7772151898734184e-06,
      "loss": 0.057783059775829315,
      "num_input_tokens_seen": 3033048,
      "step": 1608,
      "train_runtime": 4861.9726,
      "train_tokens_per_second": 623.831
    },
    {
      "epoch": 1.6257265605256508,
      "grad_norm": 1.3649557828903198,
      "learning_rate": 3.767088607594937e-06,
      "loss": 0.05704411864280701,
      "num_input_tokens_seen": 3035034,
      "step": 1609,
      "train_runtime": 4864.2512,
      "train_tokens_per_second": 623.947
    },
    {
      "epoch": 1.6267374273439474,
      "grad_norm": 1.366093635559082,
      "learning_rate": 3.756962025316456e-06,
      "loss": 0.0632607713341713,
      "num_input_tokens_seen": 3036868,
      "step": 1610,
      "train_runtime": 4866.5102,
      "train_tokens_per_second": 624.034
    },
    {
      "epoch": 1.627748294162244,
      "grad_norm": 1.4790771007537842,
      "learning_rate": 3.746835443037975e-06,
      "loss": 0.062478166073560715,
      "num_input_tokens_seen": 3038720,
      "step": 1611,
      "train_runtime": 4868.7678,
      "train_tokens_per_second": 624.125
    },
    {
      "epoch": 1.628759160980541,
      "grad_norm": 1.6402771472930908,
      "learning_rate": 3.7367088607594943e-06,
      "loss": 0.06919492036104202,
      "num_input_tokens_seen": 3040610,
      "step": 1612,
      "train_runtime": 4871.0361,
      "train_tokens_per_second": 624.222
    },
    {
      "epoch": 1.6297700277988376,
      "grad_norm": 1.089000940322876,
      "learning_rate": 3.726582278481013e-06,
      "loss": 0.04717552661895752,
      "num_input_tokens_seen": 3042618,
      "step": 1613,
      "train_runtime": 4873.3092,
      "train_tokens_per_second": 624.343
    },
    {
      "epoch": 1.6307808946171343,
      "grad_norm": 1.3077912330627441,
      "learning_rate": 3.716455696202532e-06,
      "loss": 0.05890218913555145,
      "num_input_tokens_seen": 3044814,
      "step": 1614,
      "train_runtime": 4875.6533,
      "train_tokens_per_second": 624.494
    },
    {
      "epoch": 1.631791761435431,
      "grad_norm": 1.5189601182937622,
      "learning_rate": 3.7063291139240507e-06,
      "loss": 0.058564163744449615,
      "num_input_tokens_seen": 3046482,
      "step": 1615,
      "train_runtime": 4877.8602,
      "train_tokens_per_second": 624.553
    },
    {
      "epoch": 1.6328026282537276,
      "grad_norm": 1.5458300113677979,
      "learning_rate": 3.6962025316455702e-06,
      "loss": 0.06204815208911896,
      "num_input_tokens_seen": 3048240,
      "step": 1616,
      "train_runtime": 4880.1552,
      "train_tokens_per_second": 624.619
    },
    {
      "epoch": 1.6338134950720242,
      "grad_norm": 1.033968448638916,
      "learning_rate": 3.686075949367089e-06,
      "loss": 0.05834265798330307,
      "num_input_tokens_seen": 3050738,
      "step": 1617,
      "train_runtime": 4882.561,
      "train_tokens_per_second": 624.823
    },
    {
      "epoch": 1.6348243618903209,
      "grad_norm": 1.3086661100387573,
      "learning_rate": 3.675949367088608e-06,
      "loss": 0.0483185350894928,
      "num_input_tokens_seen": 3052798,
      "step": 1618,
      "train_runtime": 4884.8652,
      "train_tokens_per_second": 624.95
    },
    {
      "epoch": 1.6358352287086175,
      "grad_norm": 1.1004079580307007,
      "learning_rate": 3.665822784810127e-06,
      "loss": 0.038739271461963654,
      "num_input_tokens_seen": 3054350,
      "step": 1619,
      "train_runtime": 4887.0497,
      "train_tokens_per_second": 624.989
    },
    {
      "epoch": 1.6368460955269142,
      "grad_norm": 1.2069733142852783,
      "learning_rate": 3.655696202531646e-06,
      "loss": 0.039483167231082916,
      "num_input_tokens_seen": 3056636,
      "step": 1620,
      "train_runtime": 4889.4276,
      "train_tokens_per_second": 625.152
    },
    {
      "epoch": 1.637856962345211,
      "grad_norm": 1.307073712348938,
      "learning_rate": 3.645569620253165e-06,
      "loss": 0.04968315362930298,
      "num_input_tokens_seen": 3058180,
      "step": 1621,
      "train_runtime": 4905.4757,
      "train_tokens_per_second": 623.422
    },
    {
      "epoch": 1.6388678291635077,
      "grad_norm": 1.0826923847198486,
      "learning_rate": 3.6354430379746835e-06,
      "loss": 0.04820631444454193,
      "num_input_tokens_seen": 3059906,
      "step": 1622,
      "train_runtime": 4907.6825,
      "train_tokens_per_second": 623.493
    },
    {
      "epoch": 1.6398786959818044,
      "grad_norm": 0.9074928164482117,
      "learning_rate": 3.625316455696203e-06,
      "loss": 0.03573989123106003,
      "num_input_tokens_seen": 3062134,
      "step": 1623,
      "train_runtime": 4910.0854,
      "train_tokens_per_second": 623.642
    },
    {
      "epoch": 1.6408895628001012,
      "grad_norm": 1.1247352361679077,
      "learning_rate": 3.6151898734177216e-06,
      "loss": 0.049534816294908524,
      "num_input_tokens_seen": 3063896,
      "step": 1624,
      "train_runtime": 4912.3056,
      "train_tokens_per_second": 623.719
    },
    {
      "epoch": 1.6419004296183979,
      "grad_norm": 1.4930373430252075,
      "learning_rate": 3.6050632911392407e-06,
      "loss": 0.06582847237586975,
      "num_input_tokens_seen": 3065622,
      "step": 1625,
      "train_runtime": 4914.5498,
      "train_tokens_per_second": 623.785
    },
    {
      "epoch": 1.6429112964366945,
      "grad_norm": 2.20893931388855,
      "learning_rate": 3.5949367088607594e-06,
      "loss": 0.0630914717912674,
      "num_input_tokens_seen": 3067090,
      "step": 1626,
      "train_runtime": 4916.7014,
      "train_tokens_per_second": 623.811
    },
    {
      "epoch": 1.6439221632549912,
      "grad_norm": 1.1391489505767822,
      "learning_rate": 3.584810126582279e-06,
      "loss": 0.05367257446050644,
      "num_input_tokens_seen": 3069452,
      "step": 1627,
      "train_runtime": 4919.1085,
      "train_tokens_per_second": 623.985
    },
    {
      "epoch": 1.6449330300732878,
      "grad_norm": 1.2861955165863037,
      "learning_rate": 3.5746835443037975e-06,
      "loss": 0.04748687148094177,
      "num_input_tokens_seen": 3071050,
      "step": 1628,
      "train_runtime": 4921.3293,
      "train_tokens_per_second": 624.029
    },
    {
      "epoch": 1.6459438968915845,
      "grad_norm": 1.6114451885223389,
      "learning_rate": 3.5645569620253166e-06,
      "loss": 0.07197785377502441,
      "num_input_tokens_seen": 3072606,
      "step": 1629,
      "train_runtime": 4923.5241,
      "train_tokens_per_second": 624.066
    },
    {
      "epoch": 1.6469547637098811,
      "grad_norm": 1.5271601676940918,
      "learning_rate": 3.5544303797468353e-06,
      "loss": 0.04787354916334152,
      "num_input_tokens_seen": 3074244,
      "step": 1630,
      "train_runtime": 4925.7525,
      "train_tokens_per_second": 624.117
    },
    {
      "epoch": 1.6479656305281778,
      "grad_norm": 1.7118289470672607,
      "learning_rate": 3.544303797468355e-06,
      "loss": 0.06489066034555435,
      "num_input_tokens_seen": 3075800,
      "step": 1631,
      "train_runtime": 4928.0998,
      "train_tokens_per_second": 624.135
    },
    {
      "epoch": 1.6489764973464744,
      "grad_norm": 1.116047739982605,
      "learning_rate": 3.5341772151898735e-06,
      "loss": 0.03874336928129196,
      "num_input_tokens_seen": 3077748,
      "step": 1632,
      "train_runtime": 4930.3864,
      "train_tokens_per_second": 624.241
    },
    {
      "epoch": 1.6499873641647713,
      "grad_norm": 1.237392544746399,
      "learning_rate": 3.5240506329113925e-06,
      "loss": 0.05634452402591705,
      "num_input_tokens_seen": 3079672,
      "step": 1633,
      "train_runtime": 4932.703,
      "train_tokens_per_second": 624.338
    },
    {
      "epoch": 1.650998230983068,
      "grad_norm": 1.5994071960449219,
      "learning_rate": 3.5139240506329116e-06,
      "loss": 0.07179383933544159,
      "num_input_tokens_seen": 3081838,
      "step": 1634,
      "train_runtime": 4935.0457,
      "train_tokens_per_second": 624.48
    },
    {
      "epoch": 1.6520090978013646,
      "grad_norm": 1.5029454231262207,
      "learning_rate": 3.5037974683544307e-06,
      "loss": 0.07771960645914078,
      "num_input_tokens_seen": 3084034,
      "step": 1635,
      "train_runtime": 4937.3827,
      "train_tokens_per_second": 624.629
    },
    {
      "epoch": 1.6530199646196615,
      "grad_norm": 1.5678901672363281,
      "learning_rate": 3.4936708860759494e-06,
      "loss": 0.06559330224990845,
      "num_input_tokens_seen": 3085876,
      "step": 1636,
      "train_runtime": 4939.6535,
      "train_tokens_per_second": 624.715
    },
    {
      "epoch": 1.6540308314379581,
      "grad_norm": 1.4560083150863647,
      "learning_rate": 3.4835443037974685e-06,
      "loss": 0.07017797976732254,
      "num_input_tokens_seen": 3087968,
      "step": 1637,
      "train_runtime": 4941.9676,
      "train_tokens_per_second": 624.846
    },
    {
      "epoch": 1.6550416982562548,
      "grad_norm": 1.0919362306594849,
      "learning_rate": 3.4734177215189875e-06,
      "loss": 0.05355212837457657,
      "num_input_tokens_seen": 3089862,
      "step": 1638,
      "train_runtime": 4944.2331,
      "train_tokens_per_second": 624.943
    },
    {
      "epoch": 1.6560525650745515,
      "grad_norm": 1.4916249513626099,
      "learning_rate": 3.4632911392405066e-06,
      "loss": 0.07239791750907898,
      "num_input_tokens_seen": 3091530,
      "step": 1639,
      "train_runtime": 4946.4187,
      "train_tokens_per_second": 625.004
    },
    {
      "epoch": 1.657063431892848,
      "grad_norm": 1.4590784311294556,
      "learning_rate": 3.4531645569620253e-06,
      "loss": 0.04920034110546112,
      "num_input_tokens_seen": 3093300,
      "step": 1640,
      "train_runtime": 4948.6907,
      "train_tokens_per_second": 625.074
    },
    {
      "epoch": 1.6580742987111448,
      "grad_norm": 1.2773500680923462,
      "learning_rate": 3.443037974683545e-06,
      "loss": 0.0706145167350769,
      "num_input_tokens_seen": 3095484,
      "step": 1641,
      "train_runtime": 4951.032,
      "train_tokens_per_second": 625.22
    },
    {
      "epoch": 1.6590851655294414,
      "grad_norm": 1.154941201210022,
      "learning_rate": 3.4329113924050635e-06,
      "loss": 0.04785441607236862,
      "num_input_tokens_seen": 3097440,
      "step": 1642,
      "train_runtime": 4953.3447,
      "train_tokens_per_second": 625.323
    },
    {
      "epoch": 1.660096032347738,
      "grad_norm": 1.2175999879837036,
      "learning_rate": 3.4227848101265825e-06,
      "loss": 0.05075257271528244,
      "num_input_tokens_seen": 3099356,
      "step": 1643,
      "train_runtime": 4955.7301,
      "train_tokens_per_second": 625.409
    },
    {
      "epoch": 1.6611068991660347,
      "grad_norm": 1.4382654428482056,
      "learning_rate": 3.412658227848101e-06,
      "loss": 0.06688381731510162,
      "num_input_tokens_seen": 3101190,
      "step": 1644,
      "train_runtime": 4957.9881,
      "train_tokens_per_second": 625.494
    },
    {
      "epoch": 1.6621177659843316,
      "grad_norm": 1.05448317527771,
      "learning_rate": 3.4025316455696207e-06,
      "loss": 0.04826867952942848,
      "num_input_tokens_seen": 3103408,
      "step": 1645,
      "train_runtime": 4960.3376,
      "train_tokens_per_second": 625.645
    },
    {
      "epoch": 1.6631286328026282,
      "grad_norm": 1.277748942375183,
      "learning_rate": 3.3924050632911394e-06,
      "loss": 0.05567628890275955,
      "num_input_tokens_seen": 3105644,
      "step": 1646,
      "train_runtime": 4962.6731,
      "train_tokens_per_second": 625.801
    },
    {
      "epoch": 1.664139499620925,
      "grad_norm": 1.3424869775772095,
      "learning_rate": 3.3822784810126585e-06,
      "loss": 0.052945174276828766,
      "num_input_tokens_seen": 3107626,
      "step": 1647,
      "train_runtime": 4964.9861,
      "train_tokens_per_second": 625.908
    },
    {
      "epoch": 1.6651503664392218,
      "grad_norm": 1.553688406944275,
      "learning_rate": 3.372151898734177e-06,
      "loss": 0.06445041298866272,
      "num_input_tokens_seen": 3109686,
      "step": 1648,
      "train_runtime": 4967.3237,
      "train_tokens_per_second": 626.028
    },
    {
      "epoch": 1.6661612332575184,
      "grad_norm": 1.3637549877166748,
      "learning_rate": 3.3620253164556966e-06,
      "loss": 0.06860950589179993,
      "num_input_tokens_seen": 3111442,
      "step": 1649,
      "train_runtime": 4969.585,
      "train_tokens_per_second": 626.097
    },
    {
      "epoch": 1.667172100075815,
      "grad_norm": 1.2801460027694702,
      "learning_rate": 3.3518987341772153e-06,
      "loss": 0.05368033051490784,
      "num_input_tokens_seen": 3113542,
      "step": 1650,
      "train_runtime": 4971.9122,
      "train_tokens_per_second": 626.226
    },
    {
      "epoch": 1.6681829668941117,
      "grad_norm": 1.313769817352295,
      "learning_rate": 3.3417721518987344e-06,
      "loss": 0.05135171115398407,
      "num_input_tokens_seen": 3115364,
      "step": 1651,
      "train_runtime": 4988.095,
      "train_tokens_per_second": 624.56
    },
    {
      "epoch": 1.6691938337124084,
      "grad_norm": 1.1594148874282837,
      "learning_rate": 3.331645569620253e-06,
      "loss": 0.04917239770293236,
      "num_input_tokens_seen": 3117738,
      "step": 1652,
      "train_runtime": 4990.4879,
      "train_tokens_per_second": 624.736
    },
    {
      "epoch": 1.670204700530705,
      "grad_norm": 1.0684016942977905,
      "learning_rate": 3.3215189873417725e-06,
      "loss": 0.04415421560406685,
      "num_input_tokens_seen": 3119876,
      "step": 1653,
      "train_runtime": 4992.8306,
      "train_tokens_per_second": 624.871
    },
    {
      "epoch": 1.6712155673490017,
      "grad_norm": 1.3335574865341187,
      "learning_rate": 3.311392405063291e-06,
      "loss": 0.05189182609319687,
      "num_input_tokens_seen": 3121986,
      "step": 1654,
      "train_runtime": 4995.1492,
      "train_tokens_per_second": 625.004
    },
    {
      "epoch": 1.6722264341672983,
      "grad_norm": 1.289138913154602,
      "learning_rate": 3.3012658227848103e-06,
      "loss": 0.06688280403614044,
      "num_input_tokens_seen": 3123822,
      "step": 1655,
      "train_runtime": 4997.4292,
      "train_tokens_per_second": 625.086
    },
    {
      "epoch": 1.6732373009855952,
      "grad_norm": 1.097314476966858,
      "learning_rate": 3.2911392405063294e-06,
      "loss": 0.03791508451104164,
      "num_input_tokens_seen": 3125864,
      "step": 1656,
      "train_runtime": 4999.7012,
      "train_tokens_per_second": 625.21
    },
    {
      "epoch": 1.6742481678038919,
      "grad_norm": 1.4409215450286865,
      "learning_rate": 3.2810126582278485e-06,
      "loss": 0.03573412075638771,
      "num_input_tokens_seen": 3127242,
      "step": 1657,
      "train_runtime": 5001.8208,
      "train_tokens_per_second": 625.221
    },
    {
      "epoch": 1.6752590346221885,
      "grad_norm": 1.55764639377594,
      "learning_rate": 3.270886075949367e-06,
      "loss": 0.058477044105529785,
      "num_input_tokens_seen": 3128716,
      "step": 1658,
      "train_runtime": 5004.0028,
      "train_tokens_per_second": 625.243
    },
    {
      "epoch": 1.6762699014404854,
      "grad_norm": 1.7484968900680542,
      "learning_rate": 3.260759493670886e-06,
      "loss": 0.07414916157722473,
      "num_input_tokens_seen": 3130486,
      "step": 1659,
      "train_runtime": 5006.2512,
      "train_tokens_per_second": 625.315
    },
    {
      "epoch": 1.677280768258782,
      "grad_norm": 1.5339076519012451,
      "learning_rate": 3.2506329113924053e-06,
      "loss": 0.0380026251077652,
      "num_input_tokens_seen": 3132030,
      "step": 1660,
      "train_runtime": 5008.4587,
      "train_tokens_per_second": 625.348
    },
    {
      "epoch": 1.6782916350770787,
      "grad_norm": 1.1547660827636719,
      "learning_rate": 3.2405063291139244e-06,
      "loss": 0.04821736738085747,
      "num_input_tokens_seen": 3133638,
      "step": 1661,
      "train_runtime": 5010.793,
      "train_tokens_per_second": 625.378
    },
    {
      "epoch": 1.6793025018953753,
      "grad_norm": 1.6885863542556763,
      "learning_rate": 3.230379746835443e-06,
      "loss": 0.0635613203048706,
      "num_input_tokens_seen": 3135314,
      "step": 1662,
      "train_runtime": 5013.0182,
      "train_tokens_per_second": 625.434
    },
    {
      "epoch": 1.680313368713672,
      "grad_norm": 1.4018542766571045,
      "learning_rate": 3.2202531645569625e-06,
      "loss": 0.06594564020633698,
      "num_input_tokens_seen": 3137340,
      "step": 1663,
      "train_runtime": 5015.2867,
      "train_tokens_per_second": 625.555
    },
    {
      "epoch": 1.6813242355319686,
      "grad_norm": 1.5609691143035889,
      "learning_rate": 3.210126582278481e-06,
      "loss": 0.06660561263561249,
      "num_input_tokens_seen": 3139916,
      "step": 1664,
      "train_runtime": 5017.7223,
      "train_tokens_per_second": 625.765
    },
    {
      "epoch": 1.6823351023502653,
      "grad_norm": 1.3404815196990967,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 0.05947958678007126,
      "num_input_tokens_seen": 3141946,
      "step": 1665,
      "train_runtime": 5020.0333,
      "train_tokens_per_second": 625.882
    },
    {
      "epoch": 1.683345969168562,
      "grad_norm": 1.3477376699447632,
      "learning_rate": 3.189873417721519e-06,
      "loss": 0.04895316809415817,
      "num_input_tokens_seen": 3143520,
      "step": 1666,
      "train_runtime": 5022.2131,
      "train_tokens_per_second": 625.923
    },
    {
      "epoch": 1.6843568359868586,
      "grad_norm": 1.264715313911438,
      "learning_rate": 3.1797468354430384e-06,
      "loss": 0.03985775634646416,
      "num_input_tokens_seen": 3145024,
      "step": 1667,
      "train_runtime": 5024.3936,
      "train_tokens_per_second": 625.951
    },
    {
      "epoch": 1.6853677028051555,
      "grad_norm": 1.588227391242981,
      "learning_rate": 3.169620253164557e-06,
      "loss": 0.07643485069274902,
      "num_input_tokens_seen": 3146530,
      "step": 1668,
      "train_runtime": 5026.5474,
      "train_tokens_per_second": 625.982
    },
    {
      "epoch": 1.6863785696234521,
      "grad_norm": 1.2682186365127563,
      "learning_rate": 3.159493670886076e-06,
      "loss": 0.054399602115154266,
      "num_input_tokens_seen": 3148500,
      "step": 1669,
      "train_runtime": 5028.8301,
      "train_tokens_per_second": 626.09
    },
    {
      "epoch": 1.6873894364417488,
      "grad_norm": 1.1590795516967773,
      "learning_rate": 3.149367088607595e-06,
      "loss": 0.05733440816402435,
      "num_input_tokens_seen": 3150822,
      "step": 1670,
      "train_runtime": 5031.1993,
      "train_tokens_per_second": 626.257
    },
    {
      "epoch": 1.6884003032600456,
      "grad_norm": 1.8241840600967407,
      "learning_rate": 3.1392405063291144e-06,
      "loss": 0.06027050316333771,
      "num_input_tokens_seen": 3152442,
      "step": 1671,
      "train_runtime": 5033.4142,
      "train_tokens_per_second": 626.303
    },
    {
      "epoch": 1.6894111700783423,
      "grad_norm": 1.4099621772766113,
      "learning_rate": 3.129113924050633e-06,
      "loss": 0.07096248120069504,
      "num_input_tokens_seen": 3154712,
      "step": 1672,
      "train_runtime": 5035.7455,
      "train_tokens_per_second": 626.464
    },
    {
      "epoch": 1.690422036896639,
      "grad_norm": 1.4060488939285278,
      "learning_rate": 3.118987341772152e-06,
      "loss": 0.04511949419975281,
      "num_input_tokens_seen": 3156586,
      "step": 1673,
      "train_runtime": 5037.9832,
      "train_tokens_per_second": 626.557
    },
    {
      "epoch": 1.6914329037149356,
      "grad_norm": 1.4204593896865845,
      "learning_rate": 3.1088607594936708e-06,
      "loss": 0.07037386298179626,
      "num_input_tokens_seen": 3158526,
      "step": 1674,
      "train_runtime": 5040.2504,
      "train_tokens_per_second": 626.661
    },
    {
      "epoch": 1.6924437705332323,
      "grad_norm": 1.5116983652114868,
      "learning_rate": 3.0987341772151903e-06,
      "loss": 0.06725286692380905,
      "num_input_tokens_seen": 3160372,
      "step": 1675,
      "train_runtime": 5042.5054,
      "train_tokens_per_second": 626.746
    },
    {
      "epoch": 1.693454637351529,
      "grad_norm": 1.4775930643081665,
      "learning_rate": 3.088607594936709e-06,
      "loss": 0.06278984248638153,
      "num_input_tokens_seen": 3162204,
      "step": 1676,
      "train_runtime": 5044.7467,
      "train_tokens_per_second": 626.831
    },
    {
      "epoch": 1.6944655041698256,
      "grad_norm": 1.424811840057373,
      "learning_rate": 3.078481012658228e-06,
      "loss": 0.06108666583895683,
      "num_input_tokens_seen": 3164144,
      "step": 1677,
      "train_runtime": 5047.0594,
      "train_tokens_per_second": 626.928
    },
    {
      "epoch": 1.6954763709881222,
      "grad_norm": 1.261054515838623,
      "learning_rate": 3.068354430379747e-06,
      "loss": 0.04168813303112984,
      "num_input_tokens_seen": 3165916,
      "step": 1678,
      "train_runtime": 5049.315,
      "train_tokens_per_second": 626.999
    },
    {
      "epoch": 1.6964872378064189,
      "grad_norm": 1.656855821609497,
      "learning_rate": 3.058227848101266e-06,
      "loss": 0.05792325735092163,
      "num_input_tokens_seen": 3167602,
      "step": 1679,
      "train_runtime": 5051.5183,
      "train_tokens_per_second": 627.059
    },
    {
      "epoch": 1.6974981046247157,
      "grad_norm": 1.3091037273406982,
      "learning_rate": 3.048101265822785e-06,
      "loss": 0.05761159956455231,
      "num_input_tokens_seen": 3169826,
      "step": 1680,
      "train_runtime": 5053.849,
      "train_tokens_per_second": 627.21
    },
    {
      "epoch": 1.6985089714430124,
      "grad_norm": 1.2030730247497559,
      "learning_rate": 3.037974683544304e-06,
      "loss": 0.038950107991695404,
      "num_input_tokens_seen": 3171628,
      "step": 1681,
      "train_runtime": 5068.3366,
      "train_tokens_per_second": 625.773
    },
    {
      "epoch": 1.699519838261309,
      "grad_norm": 1.358215570449829,
      "learning_rate": 3.027848101265823e-06,
      "loss": 0.0661626011133194,
      "num_input_tokens_seen": 3173482,
      "step": 1682,
      "train_runtime": 5070.612,
      "train_tokens_per_second": 625.858
    },
    {
      "epoch": 1.700530705079606,
      "grad_norm": 1.5518194437026978,
      "learning_rate": 3.017721518987342e-06,
      "loss": 0.06892532855272293,
      "num_input_tokens_seen": 3175088,
      "step": 1683,
      "train_runtime": 5072.7876,
      "train_tokens_per_second": 625.906
    },
    {
      "epoch": 1.7015415718979026,
      "grad_norm": 1.2152044773101807,
      "learning_rate": 3.0075949367088608e-06,
      "loss": 0.06139761954545975,
      "num_input_tokens_seen": 3177244,
      "step": 1684,
      "train_runtime": 5075.1136,
      "train_tokens_per_second": 626.044
    },
    {
      "epoch": 1.7025524387161992,
      "grad_norm": 1.1380729675292969,
      "learning_rate": 2.9974683544303803e-06,
      "loss": 0.03789081424474716,
      "num_input_tokens_seen": 3178878,
      "step": 1685,
      "train_runtime": 5077.3501,
      "train_tokens_per_second": 626.09
    },
    {
      "epoch": 1.7035633055344959,
      "grad_norm": 1.2746224403381348,
      "learning_rate": 2.987341772151899e-06,
      "loss": 0.061715319752693176,
      "num_input_tokens_seen": 3181054,
      "step": 1686,
      "train_runtime": 5079.6889,
      "train_tokens_per_second": 626.23
    },
    {
      "epoch": 1.7045741723527925,
      "grad_norm": 1.2717055082321167,
      "learning_rate": 2.977215189873418e-06,
      "loss": 0.07180582731962204,
      "num_input_tokens_seen": 3182894,
      "step": 1687,
      "train_runtime": 5081.9371,
      "train_tokens_per_second": 626.315
    },
    {
      "epoch": 1.7055850391710892,
      "grad_norm": 1.3061388731002808,
      "learning_rate": 2.9670886075949367e-06,
      "loss": 0.06722988933324814,
      "num_input_tokens_seen": 3185000,
      "step": 1688,
      "train_runtime": 5084.3054,
      "train_tokens_per_second": 626.438
    },
    {
      "epoch": 1.7065959059893858,
      "grad_norm": 1.494813323020935,
      "learning_rate": 2.956962025316456e-06,
      "loss": 0.05606045573949814,
      "num_input_tokens_seen": 3186718,
      "step": 1689,
      "train_runtime": 5086.7546,
      "train_tokens_per_second": 626.474
    },
    {
      "epoch": 1.7076067728076825,
      "grad_norm": 1.5084346532821655,
      "learning_rate": 2.946835443037975e-06,
      "loss": 0.05965816229581833,
      "num_input_tokens_seen": 3188424,
      "step": 1690,
      "train_runtime": 5088.9936,
      "train_tokens_per_second": 626.533
    },
    {
      "epoch": 1.7086176396259791,
      "grad_norm": 1.051021695137024,
      "learning_rate": 2.936708860759494e-06,
      "loss": 0.04397488012909889,
      "num_input_tokens_seen": 3190914,
      "step": 1691,
      "train_runtime": 5091.3892,
      "train_tokens_per_second": 626.728
    },
    {
      "epoch": 1.709628506444276,
      "grad_norm": 1.4790111780166626,
      "learning_rate": 2.9265822784810126e-06,
      "loss": 0.051483381539583206,
      "num_input_tokens_seen": 3192874,
      "step": 1692,
      "train_runtime": 5093.8331,
      "train_tokens_per_second": 626.812
    },
    {
      "epoch": 1.7106393732625726,
      "grad_norm": 1.1525030136108398,
      "learning_rate": 2.916455696202532e-06,
      "loss": 0.05331018939614296,
      "num_input_tokens_seen": 3194340,
      "step": 1693,
      "train_runtime": 5096.0103,
      "train_tokens_per_second": 626.832
    },
    {
      "epoch": 1.7116502400808693,
      "grad_norm": 1.2143023014068604,
      "learning_rate": 2.9063291139240508e-06,
      "loss": 0.042922090739011765,
      "num_input_tokens_seen": 3196300,
      "step": 1694,
      "train_runtime": 5098.283,
      "train_tokens_per_second": 626.937
    },
    {
      "epoch": 1.7126611068991662,
      "grad_norm": 1.183375597000122,
      "learning_rate": 2.89620253164557e-06,
      "loss": 0.06555414944887161,
      "num_input_tokens_seen": 3198282,
      "step": 1695,
      "train_runtime": 5100.5932,
      "train_tokens_per_second": 627.041
    },
    {
      "epoch": 1.7136719737174628,
      "grad_norm": 1.7683194875717163,
      "learning_rate": 2.8860759493670885e-06,
      "loss": 0.07557274401187897,
      "num_input_tokens_seen": 3199958,
      "step": 1696,
      "train_runtime": 5102.7968,
      "train_tokens_per_second": 627.099
    },
    {
      "epoch": 1.7146828405357595,
      "grad_norm": 1.642382264137268,
      "learning_rate": 2.875949367088608e-06,
      "loss": 0.07335276901721954,
      "num_input_tokens_seen": 3201796,
      "step": 1697,
      "train_runtime": 5105.0784,
      "train_tokens_per_second": 627.179
    },
    {
      "epoch": 1.7156937073540561,
      "grad_norm": 1.3908863067626953,
      "learning_rate": 2.8658227848101267e-06,
      "loss": 0.06027093529701233,
      "num_input_tokens_seen": 3203628,
      "step": 1698,
      "train_runtime": 5107.3208,
      "train_tokens_per_second": 627.262
    },
    {
      "epoch": 1.7167045741723528,
      "grad_norm": 1.5782723426818848,
      "learning_rate": 2.8556962025316458e-06,
      "loss": 0.07139906287193298,
      "num_input_tokens_seen": 3205424,
      "step": 1699,
      "train_runtime": 5109.5423,
      "train_tokens_per_second": 627.341
    },
    {
      "epoch": 1.7177154409906494,
      "grad_norm": 1.1833198070526123,
      "learning_rate": 2.845569620253165e-06,
      "loss": 0.05544673651456833,
      "num_input_tokens_seen": 3207514,
      "step": 1700,
      "train_runtime": 5111.8374,
      "train_tokens_per_second": 627.468
    },
    {
      "epoch": 1.718726307808946,
      "grad_norm": 1.814977765083313,
      "learning_rate": 2.835443037974684e-06,
      "loss": 0.08291111141443253,
      "num_input_tokens_seen": 3209430,
      "step": 1701,
      "train_runtime": 5114.1121,
      "train_tokens_per_second": 627.563
    },
    {
      "epoch": 1.7197371746272427,
      "grad_norm": 1.644702672958374,
      "learning_rate": 2.8253164556962026e-06,
      "loss": 0.05391887202858925,
      "num_input_tokens_seen": 3211230,
      "step": 1702,
      "train_runtime": 5116.3445,
      "train_tokens_per_second": 627.641
    },
    {
      "epoch": 1.7207480414455394,
      "grad_norm": 1.7358590364456177,
      "learning_rate": 2.8151898734177217e-06,
      "loss": 0.06020115688443184,
      "num_input_tokens_seen": 3213296,
      "step": 1703,
      "train_runtime": 5118.6199,
      "train_tokens_per_second": 627.766
    },
    {
      "epoch": 1.7217589082638363,
      "grad_norm": 1.419994831085205,
      "learning_rate": 2.8050632911392408e-06,
      "loss": 0.07038295269012451,
      "num_input_tokens_seen": 3215426,
      "step": 1704,
      "train_runtime": 5120.9568,
      "train_tokens_per_second": 627.896
    },
    {
      "epoch": 1.722769775082133,
      "grad_norm": 1.4363977909088135,
      "learning_rate": 2.79493670886076e-06,
      "loss": 0.055247142910957336,
      "num_input_tokens_seen": 3217030,
      "step": 1705,
      "train_runtime": 5123.1464,
      "train_tokens_per_second": 627.94
    },
    {
      "epoch": 1.7237806419004296,
      "grad_norm": 1.2219488620758057,
      "learning_rate": 2.7848101265822785e-06,
      "loss": 0.04963585361838341,
      "num_input_tokens_seen": 3219282,
      "step": 1706,
      "train_runtime": 5125.5511,
      "train_tokens_per_second": 628.085
    },
    {
      "epoch": 1.7247915087187264,
      "grad_norm": 1.0972269773483276,
      "learning_rate": 2.774683544303798e-06,
      "loss": 0.05038131773471832,
      "num_input_tokens_seen": 3221324,
      "step": 1707,
      "train_runtime": 5127.9116,
      "train_tokens_per_second": 628.194
    },
    {
      "epoch": 1.725802375537023,
      "grad_norm": 1.1392947435379028,
      "learning_rate": 2.7645569620253167e-06,
      "loss": 0.04151773452758789,
      "num_input_tokens_seen": 3223002,
      "step": 1708,
      "train_runtime": 5130.1239,
      "train_tokens_per_second": 628.25
    },
    {
      "epoch": 1.7268132423553197,
      "grad_norm": 1.592848539352417,
      "learning_rate": 2.7544303797468358e-06,
      "loss": 0.06350857764482498,
      "num_input_tokens_seen": 3224970,
      "step": 1709,
      "train_runtime": 5132.3997,
      "train_tokens_per_second": 628.355
    },
    {
      "epoch": 1.7278241091736164,
      "grad_norm": 1.2553719282150269,
      "learning_rate": 2.7443037974683544e-06,
      "loss": 0.04228603467345238,
      "num_input_tokens_seen": 3226402,
      "step": 1710,
      "train_runtime": 5134.5792,
      "train_tokens_per_second": 628.367
    },
    {
      "epoch": 1.728834975991913,
      "grad_norm": 1.384685754776001,
      "learning_rate": 2.734177215189874e-06,
      "loss": 0.050136856734752655,
      "num_input_tokens_seen": 3228294,
      "step": 1711,
      "train_runtime": 5149.0504,
      "train_tokens_per_second": 626.969
    },
    {
      "epoch": 1.7298458428102097,
      "grad_norm": 1.3079723119735718,
      "learning_rate": 2.7240506329113926e-06,
      "loss": 0.06112975254654884,
      "num_input_tokens_seen": 3230478,
      "step": 1712,
      "train_runtime": 5151.3796,
      "train_tokens_per_second": 627.109
    },
    {
      "epoch": 1.7308567096285064,
      "grad_norm": 1.4471968412399292,
      "learning_rate": 2.7139240506329117e-06,
      "loss": 0.04918985068798065,
      "num_input_tokens_seen": 3232430,
      "step": 1713,
      "train_runtime": 5153.6547,
      "train_tokens_per_second": 627.211
    },
    {
      "epoch": 1.731867576446803,
      "grad_norm": 1.5963494777679443,
      "learning_rate": 2.7037974683544303e-06,
      "loss": 0.07741352170705795,
      "num_input_tokens_seen": 3234332,
      "step": 1714,
      "train_runtime": 5155.94,
      "train_tokens_per_second": 627.302
    },
    {
      "epoch": 1.7328784432650999,
      "grad_norm": 1.1868281364440918,
      "learning_rate": 2.69367088607595e-06,
      "loss": 0.05764024332165718,
      "num_input_tokens_seen": 3236438,
      "step": 1715,
      "train_runtime": 5158.2071,
      "train_tokens_per_second": 627.435
    },
    {
      "epoch": 1.7338893100833965,
      "grad_norm": 1.416505217552185,
      "learning_rate": 2.6835443037974685e-06,
      "loss": 0.04790296405553818,
      "num_input_tokens_seen": 3238184,
      "step": 1716,
      "train_runtime": 5160.4459,
      "train_tokens_per_second": 627.501
    },
    {
      "epoch": 1.7349001769016932,
      "grad_norm": 1.231259822845459,
      "learning_rate": 2.6734177215189876e-06,
      "loss": 0.04259142279624939,
      "num_input_tokens_seen": 3239792,
      "step": 1717,
      "train_runtime": 5162.6736,
      "train_tokens_per_second": 627.542
    },
    {
      "epoch": 1.73591104371999,
      "grad_norm": 1.1704325675964355,
      "learning_rate": 2.6632911392405062e-06,
      "loss": 0.057569827884435654,
      "num_input_tokens_seen": 3241846,
      "step": 1718,
      "train_runtime": 5164.9773,
      "train_tokens_per_second": 627.659
    },
    {
      "epoch": 1.7369219105382867,
      "grad_norm": 1.3232004642486572,
      "learning_rate": 2.6531645569620257e-06,
      "loss": 0.05082739517092705,
      "num_input_tokens_seen": 3243330,
      "step": 1719,
      "train_runtime": 5167.19,
      "train_tokens_per_second": 627.678
    },
    {
      "epoch": 1.7379327773565834,
      "grad_norm": 1.7098069190979004,
      "learning_rate": 2.6430379746835444e-06,
      "loss": 0.046800512820482254,
      "num_input_tokens_seen": 3244718,
      "step": 1720,
      "train_runtime": 5169.358,
      "train_tokens_per_second": 627.683
    },
    {
      "epoch": 1.73894364417488,
      "grad_norm": 1.6553046703338623,
      "learning_rate": 2.6329113924050635e-06,
      "loss": 0.0658150166273117,
      "num_input_tokens_seen": 3246278,
      "step": 1721,
      "train_runtime": 5171.5428,
      "train_tokens_per_second": 627.719
    },
    {
      "epoch": 1.7399545109931767,
      "grad_norm": 1.56905996799469,
      "learning_rate": 2.6227848101265826e-06,
      "loss": 0.08340397477149963,
      "num_input_tokens_seen": 3248164,
      "step": 1722,
      "train_runtime": 5173.8844,
      "train_tokens_per_second": 627.8
    },
    {
      "epoch": 1.7409653778114733,
      "grad_norm": 1.3230301141738892,
      "learning_rate": 2.6126582278481017e-06,
      "loss": 0.05231938511133194,
      "num_input_tokens_seen": 3249786,
      "step": 1723,
      "train_runtime": 5176.1478,
      "train_tokens_per_second": 627.839
    },
    {
      "epoch": 1.74197624462977,
      "grad_norm": 1.3642750978469849,
      "learning_rate": 2.6025316455696203e-06,
      "loss": 0.05207236483693123,
      "num_input_tokens_seen": 3251246,
      "step": 1724,
      "train_runtime": 5178.3586,
      "train_tokens_per_second": 627.853
    },
    {
      "epoch": 1.7429871114480666,
      "grad_norm": 1.70603346824646,
      "learning_rate": 2.5924050632911394e-06,
      "loss": 0.06655357033014297,
      "num_input_tokens_seen": 3252628,
      "step": 1725,
      "train_runtime": 5180.5002,
      "train_tokens_per_second": 627.86
    },
    {
      "epoch": 1.7439979782663633,
      "grad_norm": 1.0853768587112427,
      "learning_rate": 2.5822784810126585e-06,
      "loss": 0.03993096202611923,
      "num_input_tokens_seen": 3254830,
      "step": 1726,
      "train_runtime": 5182.9408,
      "train_tokens_per_second": 627.989
    },
    {
      "epoch": 1.7450088450846601,
      "grad_norm": 1.2915998697280884,
      "learning_rate": 2.5721518987341776e-06,
      "loss": 0.055095624178647995,
      "num_input_tokens_seen": 3256576,
      "step": 1727,
      "train_runtime": 5185.1588,
      "train_tokens_per_second": 628.057
    },
    {
      "epoch": 1.7460197119029568,
      "grad_norm": 0.9929714798927307,
      "learning_rate": 2.5620253164556962e-06,
      "loss": 0.039859525859355927,
      "num_input_tokens_seen": 3258778,
      "step": 1728,
      "train_runtime": 5187.5361,
      "train_tokens_per_second": 628.194
    },
    {
      "epoch": 1.7470305787212534,
      "grad_norm": 1.6721854209899902,
      "learning_rate": 2.5518987341772157e-06,
      "loss": 0.06250029057264328,
      "num_input_tokens_seen": 3260680,
      "step": 1729,
      "train_runtime": 5189.8551,
      "train_tokens_per_second": 628.28
    },
    {
      "epoch": 1.7480414455395503,
      "grad_norm": 1.2347322702407837,
      "learning_rate": 2.5417721518987344e-06,
      "loss": 0.041486263275146484,
      "num_input_tokens_seen": 3262762,
      "step": 1730,
      "train_runtime": 5192.1666,
      "train_tokens_per_second": 628.401
    },
    {
      "epoch": 1.749052312357847,
      "grad_norm": 1.375963568687439,
      "learning_rate": 2.5316455696202535e-06,
      "loss": 0.06132054701447487,
      "num_input_tokens_seen": 3264636,
      "step": 1731,
      "train_runtime": 5194.4166,
      "train_tokens_per_second": 628.489
    },
    {
      "epoch": 1.7500631791761436,
      "grad_norm": 1.9958546161651611,
      "learning_rate": 2.521518987341772e-06,
      "loss": 0.05005979910492897,
      "num_input_tokens_seen": 3266290,
      "step": 1732,
      "train_runtime": 5196.622,
      "train_tokens_per_second": 628.541
    },
    {
      "epoch": 1.7510740459944403,
      "grad_norm": 1.1845694780349731,
      "learning_rate": 2.5113924050632917e-06,
      "loss": 0.04483165964484215,
      "num_input_tokens_seen": 3268124,
      "step": 1733,
      "train_runtime": 5198.8951,
      "train_tokens_per_second": 628.619
    },
    {
      "epoch": 1.752084912812737,
      "grad_norm": 1.1084177494049072,
      "learning_rate": 2.5012658227848103e-06,
      "loss": 0.0325375497341156,
      "num_input_tokens_seen": 3269868,
      "step": 1734,
      "train_runtime": 5201.125,
      "train_tokens_per_second": 628.685
    },
    {
      "epoch": 1.7530957796310336,
      "grad_norm": 1.0399636030197144,
      "learning_rate": 2.4911392405063294e-06,
      "loss": 0.029853176325559616,
      "num_input_tokens_seen": 3271730,
      "step": 1735,
      "train_runtime": 5203.4105,
      "train_tokens_per_second": 628.766
    },
    {
      "epoch": 1.7541066464493302,
      "grad_norm": 1.1212575435638428,
      "learning_rate": 2.481012658227848e-06,
      "loss": 0.04909026622772217,
      "num_input_tokens_seen": 3273410,
      "step": 1736,
      "train_runtime": 5205.6414,
      "train_tokens_per_second": 628.82
    },
    {
      "epoch": 1.7551175132676269,
      "grad_norm": 1.4462919235229492,
      "learning_rate": 2.470886075949367e-06,
      "loss": 0.03970807418227196,
      "num_input_tokens_seen": 3274634,
      "step": 1737,
      "train_runtime": 5207.7367,
      "train_tokens_per_second": 628.802
    },
    {
      "epoch": 1.7561283800859235,
      "grad_norm": 1.2837393283843994,
      "learning_rate": 2.4607594936708862e-06,
      "loss": 0.05064672231674194,
      "num_input_tokens_seen": 3276240,
      "step": 1738,
      "train_runtime": 5209.9421,
      "train_tokens_per_second": 628.844
    },
    {
      "epoch": 1.7571392469042204,
      "grad_norm": 1.4514061212539673,
      "learning_rate": 2.4506329113924053e-06,
      "loss": 0.06642775982618332,
      "num_input_tokens_seen": 3278222,
      "step": 1739,
      "train_runtime": 5212.204,
      "train_tokens_per_second": 628.951
    },
    {
      "epoch": 1.758150113722517,
      "grad_norm": 1.253423810005188,
      "learning_rate": 2.440506329113924e-06,
      "loss": 0.05761011317372322,
      "num_input_tokens_seen": 3279978,
      "step": 1740,
      "train_runtime": 5214.4107,
      "train_tokens_per_second": 629.022
    },
    {
      "epoch": 1.7591609805408137,
      "grad_norm": 1.2656490802764893,
      "learning_rate": 2.430379746835443e-06,
      "loss": 0.04442978650331497,
      "num_input_tokens_seen": 3282112,
      "step": 1741,
      "train_runtime": 5230.228,
      "train_tokens_per_second": 627.528
    },
    {
      "epoch": 1.7601718473591106,
      "grad_norm": 1.2130296230316162,
      "learning_rate": 2.420253164556962e-06,
      "loss": 0.05412264168262482,
      "num_input_tokens_seen": 3284090,
      "step": 1742,
      "train_runtime": 5232.6443,
      "train_tokens_per_second": 627.616
    },
    {
      "epoch": 1.7611827141774072,
      "grad_norm": 1.0931416749954224,
      "learning_rate": 2.4101265822784812e-06,
      "loss": 0.04760541766881943,
      "num_input_tokens_seen": 3285966,
      "step": 1743,
      "train_runtime": 5234.9581,
      "train_tokens_per_second": 627.697
    },
    {
      "epoch": 1.762193580995704,
      "grad_norm": 1.3005732297897339,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.06743331998586655,
      "num_input_tokens_seen": 3287836,
      "step": 1744,
      "train_runtime": 5237.2041,
      "train_tokens_per_second": 627.785
    },
    {
      "epoch": 1.7632044478140005,
      "grad_norm": 1.2369167804718018,
      "learning_rate": 2.389873417721519e-06,
      "loss": 0.057023998349905014,
      "num_input_tokens_seen": 3289794,
      "step": 1745,
      "train_runtime": 5239.5151,
      "train_tokens_per_second": 627.881
    },
    {
      "epoch": 1.7642153146322972,
      "grad_norm": 1.1120179891586304,
      "learning_rate": 2.379746835443038e-06,
      "loss": 0.03772059828042984,
      "num_input_tokens_seen": 3291814,
      "step": 1746,
      "train_runtime": 5241.8309,
      "train_tokens_per_second": 627.989
    },
    {
      "epoch": 1.7652261814505938,
      "grad_norm": 1.1966243982315063,
      "learning_rate": 2.369620253164557e-06,
      "loss": 0.05821807682514191,
      "num_input_tokens_seen": 3293728,
      "step": 1747,
      "train_runtime": 5244.1125,
      "train_tokens_per_second": 628.081
    },
    {
      "epoch": 1.7662370482688905,
      "grad_norm": 1.4127540588378906,
      "learning_rate": 2.3594936708860762e-06,
      "loss": 0.05692339688539505,
      "num_input_tokens_seen": 3295612,
      "step": 1748,
      "train_runtime": 5246.3544,
      "train_tokens_per_second": 628.172
    },
    {
      "epoch": 1.7672479150871871,
      "grad_norm": 1.5254579782485962,
      "learning_rate": 2.349367088607595e-06,
      "loss": 0.06370013952255249,
      "num_input_tokens_seen": 3297530,
      "step": 1749,
      "train_runtime": 5248.7038,
      "train_tokens_per_second": 628.256
    },
    {
      "epoch": 1.7682587819054838,
      "grad_norm": 1.0889806747436523,
      "learning_rate": 2.339240506329114e-06,
      "loss": 0.04300104081630707,
      "num_input_tokens_seen": 3299672,
      "step": 1750,
      "train_runtime": 5251.0411,
      "train_tokens_per_second": 628.384
    },
    {
      "epoch": 1.7692696487237807,
      "grad_norm": 1.6858285665512085,
      "learning_rate": 2.329113924050633e-06,
      "loss": 0.05766045302152634,
      "num_input_tokens_seen": 3301324,
      "step": 1751,
      "train_runtime": 5253.3303,
      "train_tokens_per_second": 628.425
    },
    {
      "epoch": 1.7702805155420773,
      "grad_norm": 1.1065605878829956,
      "learning_rate": 2.318987341772152e-06,
      "loss": 0.05296618491411209,
      "num_input_tokens_seen": 3303024,
      "step": 1752,
      "train_runtime": 5255.5401,
      "train_tokens_per_second": 628.484
    },
    {
      "epoch": 1.771291382360374,
      "grad_norm": 1.292311668395996,
      "learning_rate": 2.308860759493671e-06,
      "loss": 0.04545572027564049,
      "num_input_tokens_seen": 3304390,
      "step": 1753,
      "train_runtime": 5257.6699,
      "train_tokens_per_second": 628.489
    },
    {
      "epoch": 1.7723022491786709,
      "grad_norm": 1.6802729368209839,
      "learning_rate": 2.29873417721519e-06,
      "loss": 0.06228732317686081,
      "num_input_tokens_seen": 3306056,
      "step": 1754,
      "train_runtime": 5259.8827,
      "train_tokens_per_second": 628.542
    },
    {
      "epoch": 1.7733131159969675,
      "grad_norm": 1.3820728063583374,
      "learning_rate": 2.288607594936709e-06,
      "loss": 0.043544918298721313,
      "num_input_tokens_seen": 3308636,
      "step": 1755,
      "train_runtime": 5262.3307,
      "train_tokens_per_second": 628.74
    },
    {
      "epoch": 1.7743239828152642,
      "grad_norm": 1.2809100151062012,
      "learning_rate": 2.278481012658228e-06,
      "loss": 0.04450707137584686,
      "num_input_tokens_seen": 3310440,
      "step": 1756,
      "train_runtime": 5264.5934,
      "train_tokens_per_second": 628.812
    },
    {
      "epoch": 1.7753348496335608,
      "grad_norm": 1.3307225704193115,
      "learning_rate": 2.268354430379747e-06,
      "loss": 0.05021868646144867,
      "num_input_tokens_seen": 3312156,
      "step": 1757,
      "train_runtime": 5266.8612,
      "train_tokens_per_second": 628.867
    },
    {
      "epoch": 1.7763457164518575,
      "grad_norm": 1.0402779579162598,
      "learning_rate": 2.258227848101266e-06,
      "loss": 0.03799012303352356,
      "num_input_tokens_seen": 3314202,
      "step": 1758,
      "train_runtime": 5269.1894,
      "train_tokens_per_second": 628.978
    },
    {
      "epoch": 1.7773565832701541,
      "grad_norm": 1.2089042663574219,
      "learning_rate": 2.248101265822785e-06,
      "loss": 0.04334067553281784,
      "num_input_tokens_seen": 3315796,
      "step": 1759,
      "train_runtime": 5271.3738,
      "train_tokens_per_second": 629.019
    },
    {
      "epoch": 1.7783674500884508,
      "grad_norm": 1.2877362966537476,
      "learning_rate": 2.237974683544304e-06,
      "loss": 0.049931056797504425,
      "num_input_tokens_seen": 3317610,
      "step": 1760,
      "train_runtime": 5273.6526,
      "train_tokens_per_second": 629.092
    },
    {
      "epoch": 1.7793783169067474,
      "grad_norm": 1.2813947200775146,
      "learning_rate": 2.227848101265823e-06,
      "loss": 0.059495292603969574,
      "num_input_tokens_seen": 3319370,
      "step": 1761,
      "train_runtime": 5275.9368,
      "train_tokens_per_second": 629.153
    },
    {
      "epoch": 1.780389183725044,
      "grad_norm": 1.388823390007019,
      "learning_rate": 2.2177215189873417e-06,
      "loss": 0.052768148481845856,
      "num_input_tokens_seen": 3321328,
      "step": 1762,
      "train_runtime": 5278.245,
      "train_tokens_per_second": 629.249
    },
    {
      "epoch": 1.781400050543341,
      "grad_norm": 1.5318822860717773,
      "learning_rate": 2.207594936708861e-06,
      "loss": 0.06609731167554855,
      "num_input_tokens_seen": 3322902,
      "step": 1763,
      "train_runtime": 5280.3977,
      "train_tokens_per_second": 629.29
    },
    {
      "epoch": 1.7824109173616376,
      "grad_norm": 1.3329148292541504,
      "learning_rate": 2.19746835443038e-06,
      "loss": 0.06167571619153023,
      "num_input_tokens_seen": 3324686,
      "step": 1764,
      "train_runtime": 5282.6133,
      "train_tokens_per_second": 629.364
    },
    {
      "epoch": 1.7834217841799342,
      "grad_norm": 1.3537893295288086,
      "learning_rate": 2.187341772151899e-06,
      "loss": 0.06361069530248642,
      "num_input_tokens_seen": 3327138,
      "step": 1765,
      "train_runtime": 5285.0111,
      "train_tokens_per_second": 629.542
    },
    {
      "epoch": 1.7844326509982311,
      "grad_norm": 1.1344093084335327,
      "learning_rate": 2.177215189873418e-06,
      "loss": 0.057892173528671265,
      "num_input_tokens_seen": 3329122,
      "step": 1766,
      "train_runtime": 5287.3169,
      "train_tokens_per_second": 629.643
    },
    {
      "epoch": 1.7854435178165278,
      "grad_norm": 1.3750102519989014,
      "learning_rate": 2.1670886075949367e-06,
      "loss": 0.04363403469324112,
      "num_input_tokens_seen": 3330766,
      "step": 1767,
      "train_runtime": 5289.5256,
      "train_tokens_per_second": 629.691
    },
    {
      "epoch": 1.7864543846348244,
      "grad_norm": 1.705133080482483,
      "learning_rate": 2.156962025316456e-06,
      "loss": 0.053902529180049896,
      "num_input_tokens_seen": 3332672,
      "step": 1768,
      "train_runtime": 5291.781,
      "train_tokens_per_second": 629.783
    },
    {
      "epoch": 1.787465251453121,
      "grad_norm": 1.1496018171310425,
      "learning_rate": 2.146835443037975e-06,
      "loss": 0.04963824525475502,
      "num_input_tokens_seen": 3334644,
      "step": 1769,
      "train_runtime": 5294.1344,
      "train_tokens_per_second": 629.875
    },
    {
      "epoch": 1.7884761182714177,
      "grad_norm": 1.2758835554122925,
      "learning_rate": 2.136708860759494e-06,
      "loss": 0.0505569651722908,
      "num_input_tokens_seen": 3336452,
      "step": 1770,
      "train_runtime": 5296.3473,
      "train_tokens_per_second": 629.953
    },
    {
      "epoch": 1.7894869850897144,
      "grad_norm": 1.2700601816177368,
      "learning_rate": 2.1265822784810126e-06,
      "loss": 0.05532461032271385,
      "num_input_tokens_seen": 3338456,
      "step": 1771,
      "train_runtime": 5312.2418,
      "train_tokens_per_second": 628.446
    },
    {
      "epoch": 1.790497851908011,
      "grad_norm": 1.4618710279464722,
      "learning_rate": 2.1164556962025317e-06,
      "loss": 0.07050764560699463,
      "num_input_tokens_seen": 3340018,
      "step": 1772,
      "train_runtime": 5314.3885,
      "train_tokens_per_second": 628.486
    },
    {
      "epoch": 1.7915087187263077,
      "grad_norm": 1.1822164058685303,
      "learning_rate": 2.106329113924051e-06,
      "loss": 0.048912353813648224,
      "num_input_tokens_seen": 3342170,
      "step": 1773,
      "train_runtime": 5316.7959,
      "train_tokens_per_second": 628.606
    },
    {
      "epoch": 1.7925195855446043,
      "grad_norm": 1.1533139944076538,
      "learning_rate": 2.09620253164557e-06,
      "loss": 0.04874260723590851,
      "num_input_tokens_seen": 3343816,
      "step": 1774,
      "train_runtime": 5319.0819,
      "train_tokens_per_second": 628.645
    },
    {
      "epoch": 1.7935304523629012,
      "grad_norm": 1.4660189151763916,
      "learning_rate": 2.0860759493670885e-06,
      "loss": 0.05181489884853363,
      "num_input_tokens_seen": 3345358,
      "step": 1775,
      "train_runtime": 5321.3081,
      "train_tokens_per_second": 628.672
    },
    {
      "epoch": 1.7945413191811979,
      "grad_norm": 1.4509342908859253,
      "learning_rate": 2.0759493670886076e-06,
      "loss": 0.05990975350141525,
      "num_input_tokens_seen": 3347366,
      "step": 1776,
      "train_runtime": 5323.6016,
      "train_tokens_per_second": 628.778
    },
    {
      "epoch": 1.7955521859994947,
      "grad_norm": 1.3399603366851807,
      "learning_rate": 2.0658227848101267e-06,
      "loss": 0.053806304931640625,
      "num_input_tokens_seen": 3349976,
      "step": 1777,
      "train_runtime": 5326.008,
      "train_tokens_per_second": 628.984
    },
    {
      "epoch": 1.7965630528177914,
      "grad_norm": 1.341980218887329,
      "learning_rate": 2.055696202531646e-06,
      "loss": 0.06572294980287552,
      "num_input_tokens_seen": 3351784,
      "step": 1778,
      "train_runtime": 5328.2431,
      "train_tokens_per_second": 629.06
    },
    {
      "epoch": 1.797573919636088,
      "grad_norm": 1.309593677520752,
      "learning_rate": 2.045569620253165e-06,
      "loss": 0.06417813897132874,
      "num_input_tokens_seen": 3353932,
      "step": 1779,
      "train_runtime": 5330.5755,
      "train_tokens_per_second": 629.188
    },
    {
      "epoch": 1.7985847864543847,
      "grad_norm": 1.408837080001831,
      "learning_rate": 2.0354430379746835e-06,
      "loss": 0.0630507543683052,
      "num_input_tokens_seen": 3355726,
      "step": 1780,
      "train_runtime": 5332.8093,
      "train_tokens_per_second": 629.26
    },
    {
      "epoch": 1.7995956532726813,
      "grad_norm": 1.135152816772461,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.04721590131521225,
      "num_input_tokens_seen": 3358448,
      "step": 1781,
      "train_runtime": 5335.2893,
      "train_tokens_per_second": 629.478
    },
    {
      "epoch": 1.800606520090978,
      "grad_norm": 1.1626096963882446,
      "learning_rate": 2.0151898734177217e-06,
      "loss": 0.060310959815979004,
      "num_input_tokens_seen": 3360568,
      "step": 1782,
      "train_runtime": 5337.6136,
      "train_tokens_per_second": 629.601
    },
    {
      "epoch": 1.800606520090978,
      "eval_loss": 0.3664248585700989,
      "eval_runtime": 54.5014,
      "eval_samples_per_second": 16.128,
      "eval_steps_per_second": 8.073,
      "num_input_tokens_seen": 3360568,
      "step": 1782
    },
    {
      "epoch": 1.8016173869092746,
      "grad_norm": 1.2499885559082031,
      "learning_rate": 2.005063291139241e-06,
      "loss": 0.05978626385331154,
      "num_input_tokens_seen": 3362796,
      "step": 1783,
      "train_runtime": 5394.4877,
      "train_tokens_per_second": 623.376
    },
    {
      "epoch": 1.8026282537275713,
      "grad_norm": 1.3499104976654053,
      "learning_rate": 1.9949367088607595e-06,
      "loss": 0.06265848875045776,
      "num_input_tokens_seen": 3364652,
      "step": 1784,
      "train_runtime": 5396.7521,
      "train_tokens_per_second": 623.459
    },
    {
      "epoch": 1.803639120545868,
      "grad_norm": 1.4324194192886353,
      "learning_rate": 1.9848101265822785e-06,
      "loss": 0.05359485000371933,
      "num_input_tokens_seen": 3366096,
      "step": 1785,
      "train_runtime": 5398.8732,
      "train_tokens_per_second": 623.481
    },
    {
      "epoch": 1.8046499873641648,
      "grad_norm": 1.269586205482483,
      "learning_rate": 1.9746835443037976e-06,
      "loss": 0.043103210628032684,
      "num_input_tokens_seen": 3368244,
      "step": 1786,
      "train_runtime": 5401.2311,
      "train_tokens_per_second": 623.607
    },
    {
      "epoch": 1.8056608541824615,
      "grad_norm": 1.6268213987350464,
      "learning_rate": 1.9645569620253167e-06,
      "loss": 0.07587891817092896,
      "num_input_tokens_seen": 3369938,
      "step": 1787,
      "train_runtime": 5403.4442,
      "train_tokens_per_second": 623.665
    },
    {
      "epoch": 1.8066717210007581,
      "grad_norm": 2.2663137912750244,
      "learning_rate": 1.954430379746836e-06,
      "loss": 0.05486909672617912,
      "num_input_tokens_seen": 3371544,
      "step": 1788,
      "train_runtime": 5405.5838,
      "train_tokens_per_second": 623.715
    },
    {
      "epoch": 1.807682587819055,
      "grad_norm": 1.3724452257156372,
      "learning_rate": 1.9443037974683544e-06,
      "loss": 0.042518265545368195,
      "num_input_tokens_seen": 3372914,
      "step": 1789,
      "train_runtime": 5407.693,
      "train_tokens_per_second": 623.725
    },
    {
      "epoch": 1.8086934546373516,
      "grad_norm": 1.1299023628234863,
      "learning_rate": 1.9341772151898735e-06,
      "loss": 0.0554213710129261,
      "num_input_tokens_seen": 3374998,
      "step": 1790,
      "train_runtime": 5410.0205,
      "train_tokens_per_second": 623.842
    },
    {
      "epoch": 1.8097043214556483,
      "grad_norm": 1.0861995220184326,
      "learning_rate": 1.9240506329113926e-06,
      "loss": 0.05097381770610809,
      "num_input_tokens_seen": 3376898,
      "step": 1791,
      "train_runtime": 5412.2847,
      "train_tokens_per_second": 623.932
    },
    {
      "epoch": 1.810715188273945,
      "grad_norm": 1.3107178211212158,
      "learning_rate": 1.9139240506329117e-06,
      "loss": 0.0591639019548893,
      "num_input_tokens_seen": 3379128,
      "step": 1792,
      "train_runtime": 5414.7192,
      "train_tokens_per_second": 624.063
    },
    {
      "epoch": 1.8117260550922416,
      "grad_norm": 1.7104859352111816,
      "learning_rate": 1.9037974683544306e-06,
      "loss": 0.05437002703547478,
      "num_input_tokens_seen": 3380466,
      "step": 1793,
      "train_runtime": 5416.8642,
      "train_tokens_per_second": 624.063
    },
    {
      "epoch": 1.8127369219105383,
      "grad_norm": 1.2663828134536743,
      "learning_rate": 1.8936708860759494e-06,
      "loss": 0.056174956262111664,
      "num_input_tokens_seen": 3381916,
      "step": 1794,
      "train_runtime": 5419.0256,
      "train_tokens_per_second": 624.082
    },
    {
      "epoch": 1.813747788728835,
      "grad_norm": 1.0927003622055054,
      "learning_rate": 1.8835443037974685e-06,
      "loss": 0.04781580716371536,
      "num_input_tokens_seen": 3384378,
      "step": 1795,
      "train_runtime": 5421.4335,
      "train_tokens_per_second": 624.259
    },
    {
      "epoch": 1.8147586555471316,
      "grad_norm": 1.5116275548934937,
      "learning_rate": 1.8734177215189874e-06,
      "loss": 0.0798468217253685,
      "num_input_tokens_seen": 3386200,
      "step": 1796,
      "train_runtime": 5423.6687,
      "train_tokens_per_second": 624.338
    },
    {
      "epoch": 1.8157695223654282,
      "grad_norm": 1.0812292098999023,
      "learning_rate": 1.8632911392405065e-06,
      "loss": 0.04019061475992203,
      "num_input_tokens_seen": 3388380,
      "step": 1797,
      "train_runtime": 5426.0056,
      "train_tokens_per_second": 624.47
    },
    {
      "epoch": 1.816780389183725,
      "grad_norm": 1.4491227865219116,
      "learning_rate": 1.8531645569620254e-06,
      "loss": 0.039338499307632446,
      "num_input_tokens_seen": 3390244,
      "step": 1798,
      "train_runtime": 5428.3781,
      "train_tokens_per_second": 624.541
    },
    {
      "epoch": 1.8177912560020217,
      "grad_norm": 1.4811445474624634,
      "learning_rate": 1.8430379746835444e-06,
      "loss": 0.06751224398612976,
      "num_input_tokens_seen": 3392044,
      "step": 1799,
      "train_runtime": 5430.6209,
      "train_tokens_per_second": 624.614
    },
    {
      "epoch": 1.8188021228203184,
      "grad_norm": 1.068530559539795,
      "learning_rate": 1.8329113924050635e-06,
      "loss": 0.04439627379179001,
      "num_input_tokens_seen": 3394136,
      "step": 1800,
      "train_runtime": 5432.9258,
      "train_tokens_per_second": 624.734
    },
    {
      "epoch": 1.8198129896386153,
      "grad_norm": 1.0574533939361572,
      "learning_rate": 1.8227848101265824e-06,
      "loss": 0.04897860065102577,
      "num_input_tokens_seen": 3396196,
      "step": 1801,
      "train_runtime": 5448.6542,
      "train_tokens_per_second": 623.309
    },
    {
      "epoch": 1.820823856456912,
      "grad_norm": 1.2108123302459717,
      "learning_rate": 1.8126582278481015e-06,
      "loss": 0.04313907027244568,
      "num_input_tokens_seen": 3398414,
      "step": 1802,
      "train_runtime": 5450.9941,
      "train_tokens_per_second": 623.448
    },
    {
      "epoch": 1.8218347232752086,
      "grad_norm": 1.490373134613037,
      "learning_rate": 1.8025316455696204e-06,
      "loss": 0.05872609466314316,
      "num_input_tokens_seen": 3400244,
      "step": 1803,
      "train_runtime": 5453.2413,
      "train_tokens_per_second": 623.527
    },
    {
      "epoch": 1.8228455900935052,
      "grad_norm": 1.4431567192077637,
      "learning_rate": 1.7924050632911394e-06,
      "loss": 0.06668207794427872,
      "num_input_tokens_seen": 3401862,
      "step": 1804,
      "train_runtime": 5455.4959,
      "train_tokens_per_second": 623.566
    },
    {
      "epoch": 1.8238564569118019,
      "grad_norm": 1.7261453866958618,
      "learning_rate": 1.7822784810126583e-06,
      "loss": 0.05830330401659012,
      "num_input_tokens_seen": 3403396,
      "step": 1805,
      "train_runtime": 5457.6708,
      "train_tokens_per_second": 623.599
    },
    {
      "epoch": 1.8248673237300985,
      "grad_norm": 1.2453124523162842,
      "learning_rate": 1.7721518987341774e-06,
      "loss": 0.04764130711555481,
      "num_input_tokens_seen": 3405348,
      "step": 1806,
      "train_runtime": 5459.987,
      "train_tokens_per_second": 623.692
    },
    {
      "epoch": 1.8258781905483952,
      "grad_norm": 1.1063189506530762,
      "learning_rate": 1.7620253164556963e-06,
      "loss": 0.04134732857346535,
      "num_input_tokens_seen": 3407222,
      "step": 1807,
      "train_runtime": 5462.2645,
      "train_tokens_per_second": 623.775
    },
    {
      "epoch": 1.8268890573666918,
      "grad_norm": 1.3576112985610962,
      "learning_rate": 1.7518987341772154e-06,
      "loss": 0.0450436994433403,
      "num_input_tokens_seen": 3408828,
      "step": 1808,
      "train_runtime": 5464.4541,
      "train_tokens_per_second": 623.819
    },
    {
      "epoch": 1.8278999241849885,
      "grad_norm": 1.13437819480896,
      "learning_rate": 1.7417721518987342e-06,
      "loss": 0.03500455617904663,
      "num_input_tokens_seen": 3410766,
      "step": 1809,
      "train_runtime": 5466.7321,
      "train_tokens_per_second": 623.913
    },
    {
      "epoch": 1.8289107910032854,
      "grad_norm": 1.3162808418273926,
      "learning_rate": 1.7316455696202533e-06,
      "loss": 0.03751616179943085,
      "num_input_tokens_seen": 3412244,
      "step": 1810,
      "train_runtime": 5468.9127,
      "train_tokens_per_second": 623.935
    },
    {
      "epoch": 1.829921657821582,
      "grad_norm": 1.3975801467895508,
      "learning_rate": 1.7215189873417724e-06,
      "loss": 0.05604361742734909,
      "num_input_tokens_seen": 3413644,
      "step": 1811,
      "train_runtime": 5471.2383,
      "train_tokens_per_second": 623.925
    },
    {
      "epoch": 1.8309325246398787,
      "grad_norm": 1.4612771272659302,
      "learning_rate": 1.7113924050632913e-06,
      "loss": 0.06397251039743423,
      "num_input_tokens_seen": 3415632,
      "step": 1812,
      "train_runtime": 5473.6343,
      "train_tokens_per_second": 624.015
    },
    {
      "epoch": 1.8319433914581755,
      "grad_norm": 1.4182970523834229,
      "learning_rate": 1.7012658227848104e-06,
      "loss": 0.06215008720755577,
      "num_input_tokens_seen": 3417608,
      "step": 1813,
      "train_runtime": 5475.9048,
      "train_tokens_per_second": 624.117
    },
    {
      "epoch": 1.8329542582764722,
      "grad_norm": 1.336040735244751,
      "learning_rate": 1.6911392405063292e-06,
      "loss": 0.08397555351257324,
      "num_input_tokens_seen": 3419590,
      "step": 1814,
      "train_runtime": 5478.2125,
      "train_tokens_per_second": 624.216
    },
    {
      "epoch": 1.8339651250947688,
      "grad_norm": 1.0550512075424194,
      "learning_rate": 1.6810126582278483e-06,
      "loss": 0.04642835259437561,
      "num_input_tokens_seen": 3421706,
      "step": 1815,
      "train_runtime": 5480.5151,
      "train_tokens_per_second": 624.34
    },
    {
      "epoch": 1.8349759919130655,
      "grad_norm": 1.1457314491271973,
      "learning_rate": 1.6708860759493672e-06,
      "loss": 0.065874382853508,
      "num_input_tokens_seen": 3423888,
      "step": 1816,
      "train_runtime": 5482.819,
      "train_tokens_per_second": 624.476
    },
    {
      "epoch": 1.8359868587313621,
      "grad_norm": 1.548355221748352,
      "learning_rate": 1.6607594936708863e-06,
      "loss": 0.06853128969669342,
      "num_input_tokens_seen": 3425318,
      "step": 1817,
      "train_runtime": 5484.9924,
      "train_tokens_per_second": 624.489
    },
    {
      "epoch": 1.8369977255496588,
      "grad_norm": 1.3600349426269531,
      "learning_rate": 1.6506329113924051e-06,
      "loss": 0.046113211661577225,
      "num_input_tokens_seen": 3426994,
      "step": 1818,
      "train_runtime": 5487.1693,
      "train_tokens_per_second": 624.547
    },
    {
      "epoch": 1.8380085923679554,
      "grad_norm": 1.5665621757507324,
      "learning_rate": 1.6405063291139242e-06,
      "loss": 0.0590553916990757,
      "num_input_tokens_seen": 3428430,
      "step": 1819,
      "train_runtime": 5489.2871,
      "train_tokens_per_second": 624.567
    },
    {
      "epoch": 1.839019459186252,
      "grad_norm": 1.5991250276565552,
      "learning_rate": 1.630379746835443e-06,
      "loss": 0.050970666110515594,
      "num_input_tokens_seen": 3429958,
      "step": 1820,
      "train_runtime": 5491.4639,
      "train_tokens_per_second": 624.598
    },
    {
      "epoch": 1.8400303260045487,
      "grad_norm": 1.3793953657150269,
      "learning_rate": 1.6202531645569622e-06,
      "loss": 0.0674283429980278,
      "num_input_tokens_seen": 3431730,
      "step": 1821,
      "train_runtime": 5493.7033,
      "train_tokens_per_second": 624.666
    },
    {
      "epoch": 1.8410411928228456,
      "grad_norm": 1.6066606044769287,
      "learning_rate": 1.6101265822784813e-06,
      "loss": 0.07037971913814545,
      "num_input_tokens_seen": 3433280,
      "step": 1822,
      "train_runtime": 5495.9491,
      "train_tokens_per_second": 624.693
    },
    {
      "epoch": 1.8420520596411423,
      "grad_norm": 1.3880949020385742,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 0.05144776403903961,
      "num_input_tokens_seen": 3435124,
      "step": 1823,
      "train_runtime": 5498.2228,
      "train_tokens_per_second": 624.77
    },
    {
      "epoch": 1.843062926459439,
      "grad_norm": 1.7201285362243652,
      "learning_rate": 1.5898734177215192e-06,
      "loss": 0.0744432657957077,
      "num_input_tokens_seen": 3436512,
      "step": 1824,
      "train_runtime": 5500.3675,
      "train_tokens_per_second": 624.779
    },
    {
      "epoch": 1.8440737932777358,
      "grad_norm": 1.615710735321045,
      "learning_rate": 1.579746835443038e-06,
      "loss": 0.05054347589612007,
      "num_input_tokens_seen": 3437948,
      "step": 1825,
      "train_runtime": 5502.5096,
      "train_tokens_per_second": 624.796
    },
    {
      "epoch": 1.8450846600960324,
      "grad_norm": 1.3520798683166504,
      "learning_rate": 1.5696202531645572e-06,
      "loss": 0.06495378166437149,
      "num_input_tokens_seen": 3439876,
      "step": 1826,
      "train_runtime": 5504.8221,
      "train_tokens_per_second": 624.884
    },
    {
      "epoch": 1.846095526914329,
      "grad_norm": 1.4349499940872192,
      "learning_rate": 1.559493670886076e-06,
      "loss": 0.06821642071008682,
      "num_input_tokens_seen": 3441730,
      "step": 1827,
      "train_runtime": 5507.0642,
      "train_tokens_per_second": 624.966
    },
    {
      "epoch": 1.8471063937326258,
      "grad_norm": 1.3526065349578857,
      "learning_rate": 1.5493670886075951e-06,
      "loss": 0.04434981942176819,
      "num_input_tokens_seen": 3443580,
      "step": 1828,
      "train_runtime": 5509.279,
      "train_tokens_per_second": 625.051
    },
    {
      "epoch": 1.8481172605509224,
      "grad_norm": 1.303549885749817,
      "learning_rate": 1.539240506329114e-06,
      "loss": 0.04251484572887421,
      "num_input_tokens_seen": 3445306,
      "step": 1829,
      "train_runtime": 5511.5501,
      "train_tokens_per_second": 625.107
    },
    {
      "epoch": 1.849128127369219,
      "grad_norm": 1.7012320756912231,
      "learning_rate": 1.529113924050633e-06,
      "loss": 0.06697791814804077,
      "num_input_tokens_seen": 3447488,
      "step": 1830,
      "train_runtime": 5513.924,
      "train_tokens_per_second": 625.233
    },
    {
      "epoch": 1.8501389941875157,
      "grad_norm": 1.9567233324050903,
      "learning_rate": 1.518987341772152e-06,
      "loss": 0.06960449367761612,
      "num_input_tokens_seen": 3449220,
      "step": 1831,
      "train_runtime": 5528.436,
      "train_tokens_per_second": 623.905
    },
    {
      "epoch": 1.8511498610058124,
      "grad_norm": 1.2680948972702026,
      "learning_rate": 1.508860759493671e-06,
      "loss": 0.053070344030857086,
      "num_input_tokens_seen": 3451032,
      "step": 1832,
      "train_runtime": 5530.7009,
      "train_tokens_per_second": 623.977
    },
    {
      "epoch": 1.852160727824109,
      "grad_norm": 1.1677778959274292,
      "learning_rate": 1.4987341772151901e-06,
      "loss": 0.03939846530556679,
      "num_input_tokens_seen": 3452824,
      "step": 1833,
      "train_runtime": 5533.0059,
      "train_tokens_per_second": 624.041
    },
    {
      "epoch": 1.8531715946424059,
      "grad_norm": 1.1770761013031006,
      "learning_rate": 1.488607594936709e-06,
      "loss": 0.03609131649136543,
      "num_input_tokens_seen": 3454654,
      "step": 1834,
      "train_runtime": 5535.3736,
      "train_tokens_per_second": 624.105
    },
    {
      "epoch": 1.8541824614607025,
      "grad_norm": 1.313709020614624,
      "learning_rate": 1.478481012658228e-06,
      "loss": 0.041569799184799194,
      "num_input_tokens_seen": 3456194,
      "step": 1835,
      "train_runtime": 5537.5806,
      "train_tokens_per_second": 624.134
    },
    {
      "epoch": 1.8551933282789992,
      "grad_norm": 1.3749877214431763,
      "learning_rate": 1.468354430379747e-06,
      "loss": 0.035944532603025436,
      "num_input_tokens_seen": 3458628,
      "step": 1836,
      "train_runtime": 5540.0225,
      "train_tokens_per_second": 624.299
    },
    {
      "epoch": 1.856204195097296,
      "grad_norm": 1.559613823890686,
      "learning_rate": 1.458227848101266e-06,
      "loss": 0.06211206316947937,
      "num_input_tokens_seen": 3460506,
      "step": 1837,
      "train_runtime": 5542.3069,
      "train_tokens_per_second": 624.38
    },
    {
      "epoch": 1.8572150619155927,
      "grad_norm": 1.2114509344100952,
      "learning_rate": 1.448101265822785e-06,
      "loss": 0.04860848933458328,
      "num_input_tokens_seen": 3462502,
      "step": 1838,
      "train_runtime": 5544.6184,
      "train_tokens_per_second": 624.48
    },
    {
      "epoch": 1.8582259287338894,
      "grad_norm": 1.2931417226791382,
      "learning_rate": 1.437974683544304e-06,
      "loss": 0.045285362750291824,
      "num_input_tokens_seen": 3464010,
      "step": 1839,
      "train_runtime": 5546.7942,
      "train_tokens_per_second": 624.507
    },
    {
      "epoch": 1.859236795552186,
      "grad_norm": 1.4669591188430786,
      "learning_rate": 1.4278481012658229e-06,
      "loss": 0.04686623066663742,
      "num_input_tokens_seen": 3465982,
      "step": 1840,
      "train_runtime": 5549.0982,
      "train_tokens_per_second": 624.603
    },
    {
      "epoch": 1.8602476623704827,
      "grad_norm": 1.4337525367736816,
      "learning_rate": 1.417721518987342e-06,
      "loss": 0.07382497191429138,
      "num_input_tokens_seen": 3468036,
      "step": 1841,
      "train_runtime": 5551.3966,
      "train_tokens_per_second": 624.714
    },
    {
      "epoch": 1.8612585291887793,
      "grad_norm": 1.317531704902649,
      "learning_rate": 1.4075949367088608e-06,
      "loss": 0.054642219096422195,
      "num_input_tokens_seen": 3469660,
      "step": 1842,
      "train_runtime": 5553.5939,
      "train_tokens_per_second": 624.759
    },
    {
      "epoch": 1.862269396007076,
      "grad_norm": 1.6953511238098145,
      "learning_rate": 1.39746835443038e-06,
      "loss": 0.07365219295024872,
      "num_input_tokens_seen": 3471300,
      "step": 1843,
      "train_runtime": 5555.807,
      "train_tokens_per_second": 624.806
    },
    {
      "epoch": 1.8632802628253726,
      "grad_norm": 1.3182090520858765,
      "learning_rate": 1.387341772151899e-06,
      "loss": 0.05989851802587509,
      "num_input_tokens_seen": 3472838,
      "step": 1844,
      "train_runtime": 5557.9917,
      "train_tokens_per_second": 624.837
    },
    {
      "epoch": 1.8642911296436695,
      "grad_norm": 1.4372217655181885,
      "learning_rate": 1.3772151898734179e-06,
      "loss": 0.05546453222632408,
      "num_input_tokens_seen": 3474660,
      "step": 1845,
      "train_runtime": 5560.2293,
      "train_tokens_per_second": 624.913
    },
    {
      "epoch": 1.8653019964619661,
      "grad_norm": 1.42632257938385,
      "learning_rate": 1.367088607594937e-06,
      "loss": 0.0642927885055542,
      "num_input_tokens_seen": 3476294,
      "step": 1846,
      "train_runtime": 5562.4262,
      "train_tokens_per_second": 624.96
    },
    {
      "epoch": 1.8663128632802628,
      "grad_norm": 1.1571648120880127,
      "learning_rate": 1.3569620253164558e-06,
      "loss": 0.03720973804593086,
      "num_input_tokens_seen": 3477836,
      "step": 1847,
      "train_runtime": 5564.5947,
      "train_tokens_per_second": 624.994
    },
    {
      "epoch": 1.8673237300985597,
      "grad_norm": 1.2866696119308472,
      "learning_rate": 1.346835443037975e-06,
      "loss": 0.05578981712460518,
      "num_input_tokens_seen": 3479554,
      "step": 1848,
      "train_runtime": 5566.8306,
      "train_tokens_per_second": 625.051
    },
    {
      "epoch": 1.8683345969168563,
      "grad_norm": 1.5326685905456543,
      "learning_rate": 1.3367088607594938e-06,
      "loss": 0.06295924633741379,
      "num_input_tokens_seen": 3481364,
      "step": 1849,
      "train_runtime": 5569.099,
      "train_tokens_per_second": 625.122
    },
    {
      "epoch": 1.869345463735153,
      "grad_norm": 1.3938623666763306,
      "learning_rate": 1.3265822784810129e-06,
      "loss": 0.07753098011016846,
      "num_input_tokens_seen": 3483210,
      "step": 1850,
      "train_runtime": 5571.3639,
      "train_tokens_per_second": 625.199
    },
    {
      "epoch": 1.8703563305534496,
      "grad_norm": 0.940585196018219,
      "learning_rate": 1.3164556962025317e-06,
      "loss": 0.03267010673880577,
      "num_input_tokens_seen": 3484998,
      "step": 1851,
      "train_runtime": 5573.5967,
      "train_tokens_per_second": 625.269
    },
    {
      "epoch": 1.8713671973717463,
      "grad_norm": 1.0479862689971924,
      "learning_rate": 1.3063291139240508e-06,
      "loss": 0.03430189564824104,
      "num_input_tokens_seen": 3486614,
      "step": 1852,
      "train_runtime": 5575.8042,
      "train_tokens_per_second": 625.311
    },
    {
      "epoch": 1.872378064190043,
      "grad_norm": 1.2628525495529175,
      "learning_rate": 1.2962025316455697e-06,
      "loss": 0.04698391258716583,
      "num_input_tokens_seen": 3488406,
      "step": 1853,
      "train_runtime": 5578.019,
      "train_tokens_per_second": 625.384
    },
    {
      "epoch": 1.8733889310083396,
      "grad_norm": 1.6127231121063232,
      "learning_rate": 1.2860759493670888e-06,
      "loss": 0.06976167857646942,
      "num_input_tokens_seen": 3489924,
      "step": 1854,
      "train_runtime": 5580.2856,
      "train_tokens_per_second": 625.402
    },
    {
      "epoch": 1.8743997978266362,
      "grad_norm": 1.2900846004486084,
      "learning_rate": 1.2759493670886079e-06,
      "loss": 0.04649154469370842,
      "num_input_tokens_seen": 3492204,
      "step": 1855,
      "train_runtime": 5582.6512,
      "train_tokens_per_second": 625.546
    },
    {
      "epoch": 1.875410664644933,
      "grad_norm": 1.1668962240219116,
      "learning_rate": 1.2658227848101267e-06,
      "loss": 0.02965341880917549,
      "num_input_tokens_seen": 3493908,
      "step": 1856,
      "train_runtime": 5584.8966,
      "train_tokens_per_second": 625.599
    },
    {
      "epoch": 1.8764215314632298,
      "grad_norm": 1.1431705951690674,
      "learning_rate": 1.2556962025316458e-06,
      "loss": 0.05246175453066826,
      "num_input_tokens_seen": 3495948,
      "step": 1857,
      "train_runtime": 5587.168,
      "train_tokens_per_second": 625.71
    },
    {
      "epoch": 1.8774323982815264,
      "grad_norm": 1.3699880838394165,
      "learning_rate": 1.2455696202531647e-06,
      "loss": 0.05064959079027176,
      "num_input_tokens_seen": 3498066,
      "step": 1858,
      "train_runtime": 5589.4946,
      "train_tokens_per_second": 625.829
    },
    {
      "epoch": 1.878443265099823,
      "grad_norm": 1.0862371921539307,
      "learning_rate": 1.2354430379746836e-06,
      "loss": 0.05020325630903244,
      "num_input_tokens_seen": 3500240,
      "step": 1859,
      "train_runtime": 5591.8246,
      "train_tokens_per_second": 625.957
    },
    {
      "epoch": 1.87945413191812,
      "grad_norm": 1.525439739227295,
      "learning_rate": 1.2253164556962027e-06,
      "loss": 0.06935298442840576,
      "num_input_tokens_seen": 3502098,
      "step": 1860,
      "train_runtime": 5594.1119,
      "train_tokens_per_second": 626.033
    },
    {
      "epoch": 1.8804649987364166,
      "grad_norm": 1.3590785264968872,
      "learning_rate": 1.2151898734177215e-06,
      "loss": 0.041813626885414124,
      "num_input_tokens_seen": 3504010,
      "step": 1861,
      "train_runtime": 5608.8547,
      "train_tokens_per_second": 624.728
    },
    {
      "epoch": 1.8814758655547132,
      "grad_norm": 1.380476474761963,
      "learning_rate": 1.2050632911392406e-06,
      "loss": 0.07701250910758972,
      "num_input_tokens_seen": 3506308,
      "step": 1862,
      "train_runtime": 5611.2132,
      "train_tokens_per_second": 624.875
    },
    {
      "epoch": 1.88248673237301,
      "grad_norm": 1.5668699741363525,
      "learning_rate": 1.1949367088607595e-06,
      "loss": 0.04789157584309578,
      "num_input_tokens_seen": 3508312,
      "step": 1863,
      "train_runtime": 5613.5129,
      "train_tokens_per_second": 624.976
    },
    {
      "epoch": 1.8834975991913065,
      "grad_norm": 1.7109662294387817,
      "learning_rate": 1.1848101265822786e-06,
      "loss": 0.0572693832218647,
      "num_input_tokens_seen": 3510196,
      "step": 1864,
      "train_runtime": 5615.7769,
      "train_tokens_per_second": 625.06
    },
    {
      "epoch": 1.8845084660096032,
      "grad_norm": 1.3793543577194214,
      "learning_rate": 1.1746835443037974e-06,
      "loss": 0.05042321979999542,
      "num_input_tokens_seen": 3511816,
      "step": 1865,
      "train_runtime": 5617.9865,
      "train_tokens_per_second": 625.102
    },
    {
      "epoch": 1.8855193328278999,
      "grad_norm": 1.190150260925293,
      "learning_rate": 1.1645569620253165e-06,
      "loss": 0.04872255399823189,
      "num_input_tokens_seen": 3513736,
      "step": 1866,
      "train_runtime": 5620.2597,
      "train_tokens_per_second": 625.191
    },
    {
      "epoch": 1.8865301996461965,
      "grad_norm": 1.5015819072723389,
      "learning_rate": 1.1544303797468354e-06,
      "loss": 0.05711166188120842,
      "num_input_tokens_seen": 3515274,
      "step": 1867,
      "train_runtime": 5622.4345,
      "train_tokens_per_second": 625.223
    },
    {
      "epoch": 1.8875410664644932,
      "grad_norm": 1.3962602615356445,
      "learning_rate": 1.1443037974683545e-06,
      "loss": 0.055755674839019775,
      "num_input_tokens_seen": 3517364,
      "step": 1868,
      "train_runtime": 5624.7426,
      "train_tokens_per_second": 625.338
    },
    {
      "epoch": 1.88855193328279,
      "grad_norm": 1.265474557876587,
      "learning_rate": 1.1341772151898736e-06,
      "loss": 0.060417093336582184,
      "num_input_tokens_seen": 3519344,
      "step": 1869,
      "train_runtime": 5627.2291,
      "train_tokens_per_second": 625.413
    },
    {
      "epoch": 1.8895628001010867,
      "grad_norm": 1.1677736043930054,
      "learning_rate": 1.1240506329113924e-06,
      "loss": 0.041969869285821915,
      "num_input_tokens_seen": 3521038,
      "step": 1870,
      "train_runtime": 5629.4312,
      "train_tokens_per_second": 625.47
    },
    {
      "epoch": 1.8905736669193833,
      "grad_norm": 1.2462869882583618,
      "learning_rate": 1.1139240506329115e-06,
      "loss": 0.05771699920296669,
      "num_input_tokens_seen": 3522880,
      "step": 1871,
      "train_runtime": 5631.6695,
      "train_tokens_per_second": 625.548
    },
    {
      "epoch": 1.8915845337376802,
      "grad_norm": 1.385130763053894,
      "learning_rate": 1.1037974683544304e-06,
      "loss": 0.07197462767362595,
      "num_input_tokens_seen": 3525122,
      "step": 1872,
      "train_runtime": 5634.1757,
      "train_tokens_per_second": 625.668
    },
    {
      "epoch": 1.8925954005559769,
      "grad_norm": 1.5400617122650146,
      "learning_rate": 1.0936708860759495e-06,
      "loss": 0.06472428143024445,
      "num_input_tokens_seen": 3526860,
      "step": 1873,
      "train_runtime": 5636.4081,
      "train_tokens_per_second": 625.728
    },
    {
      "epoch": 1.8936062673742735,
      "grad_norm": 1.3437457084655762,
      "learning_rate": 1.0835443037974684e-06,
      "loss": 0.057449765503406525,
      "num_input_tokens_seen": 3528716,
      "step": 1874,
      "train_runtime": 5638.6949,
      "train_tokens_per_second": 625.804
    },
    {
      "epoch": 1.8946171341925702,
      "grad_norm": 1.5442469120025635,
      "learning_rate": 1.0734177215189874e-06,
      "loss": 0.07203658670186996,
      "num_input_tokens_seen": 3530244,
      "step": 1875,
      "train_runtime": 5640.8552,
      "train_tokens_per_second": 625.835
    },
    {
      "epoch": 1.8956280010108668,
      "grad_norm": 1.6463550329208374,
      "learning_rate": 1.0632911392405063e-06,
      "loss": 0.05945805460214615,
      "num_input_tokens_seen": 3531840,
      "step": 1876,
      "train_runtime": 5643.0617,
      "train_tokens_per_second": 625.873
    },
    {
      "epoch": 1.8966388678291635,
      "grad_norm": 1.2262964248657227,
      "learning_rate": 1.0531645569620254e-06,
      "loss": 0.04987657815217972,
      "num_input_tokens_seen": 3533806,
      "step": 1877,
      "train_runtime": 5645.3716,
      "train_tokens_per_second": 625.965
    },
    {
      "epoch": 1.8976497346474601,
      "grad_norm": 1.6830824613571167,
      "learning_rate": 1.0430379746835443e-06,
      "loss": 0.06614528596401215,
      "num_input_tokens_seen": 3535284,
      "step": 1878,
      "train_runtime": 5647.5175,
      "train_tokens_per_second": 625.989
    },
    {
      "epoch": 1.8986606014657568,
      "grad_norm": 1.456224799156189,
      "learning_rate": 1.0329113924050634e-06,
      "loss": 0.08167030662298203,
      "num_input_tokens_seen": 3537260,
      "step": 1879,
      "train_runtime": 5649.7835,
      "train_tokens_per_second": 626.088
    },
    {
      "epoch": 1.8996714682840534,
      "grad_norm": 1.2950639724731445,
      "learning_rate": 1.0227848101265824e-06,
      "loss": 0.04762929677963257,
      "num_input_tokens_seen": 3539170,
      "step": 1880,
      "train_runtime": 5652.0587,
      "train_tokens_per_second": 626.174
    },
    {
      "epoch": 1.9006823351023503,
      "grad_norm": 1.28152596950531,
      "learning_rate": 1.0126582278481013e-06,
      "loss": 0.05366584658622742,
      "num_input_tokens_seen": 3540982,
      "step": 1881,
      "train_runtime": 5654.3758,
      "train_tokens_per_second": 626.237
    },
    {
      "epoch": 1.901693201920647,
      "grad_norm": 1.4385573863983154,
      "learning_rate": 1.0025316455696204e-06,
      "loss": 0.07367570698261261,
      "num_input_tokens_seen": 3543062,
      "step": 1882,
      "train_runtime": 5656.6719,
      "train_tokens_per_second": 626.351
    },
    {
      "epoch": 1.9027040687389436,
      "grad_norm": 1.0679799318313599,
      "learning_rate": 9.924050632911393e-07,
      "loss": 0.04781443998217583,
      "num_input_tokens_seen": 3545132,
      "step": 1883,
      "train_runtime": 5658.94,
      "train_tokens_per_second": 626.466
    },
    {
      "epoch": 1.9037149355572405,
      "grad_norm": 1.4399374723434448,
      "learning_rate": 9.822784810126584e-07,
      "loss": 0.04740697517991066,
      "num_input_tokens_seen": 3546714,
      "step": 1884,
      "train_runtime": 5661.1141,
      "train_tokens_per_second": 626.505
    },
    {
      "epoch": 1.9047258023755371,
      "grad_norm": 1.315198540687561,
      "learning_rate": 9.721518987341772e-07,
      "loss": 0.04880426079034805,
      "num_input_tokens_seen": 3548604,
      "step": 1885,
      "train_runtime": 5663.3878,
      "train_tokens_per_second": 626.587
    },
    {
      "epoch": 1.9057366691938338,
      "grad_norm": 1.624666452407837,
      "learning_rate": 9.620253164556963e-07,
      "loss": 0.07246091216802597,
      "num_input_tokens_seen": 3550542,
      "step": 1886,
      "train_runtime": 5665.6596,
      "train_tokens_per_second": 626.678
    },
    {
      "epoch": 1.9067475360121304,
      "grad_norm": 1.181203842163086,
      "learning_rate": 9.518987341772153e-07,
      "loss": 0.05116025358438492,
      "num_input_tokens_seen": 3552096,
      "step": 1887,
      "train_runtime": 5667.868,
      "train_tokens_per_second": 626.708
    },
    {
      "epoch": 1.907758402830427,
      "grad_norm": 1.6072170734405518,
      "learning_rate": 9.417721518987343e-07,
      "loss": 0.06815660744905472,
      "num_input_tokens_seen": 3553786,
      "step": 1888,
      "train_runtime": 5670.0852,
      "train_tokens_per_second": 626.761
    },
    {
      "epoch": 1.9087692696487237,
      "grad_norm": 1.5671355724334717,
      "learning_rate": 9.316455696202532e-07,
      "loss": 0.07512356340885162,
      "num_input_tokens_seen": 3555838,
      "step": 1889,
      "train_runtime": 5672.4116,
      "train_tokens_per_second": 626.865
    },
    {
      "epoch": 1.9097801364670204,
      "grad_norm": 1.0043747425079346,
      "learning_rate": 9.215189873417722e-07,
      "loss": 0.047433193773031235,
      "num_input_tokens_seen": 3557936,
      "step": 1890,
      "train_runtime": 5674.7416,
      "train_tokens_per_second": 626.978
    },
    {
      "epoch": 1.910791003285317,
      "grad_norm": 1.227074146270752,
      "learning_rate": 9.113924050632912e-07,
      "loss": 0.06004814803600311,
      "num_input_tokens_seen": 3559842,
      "step": 1891,
      "train_runtime": 5689.1579,
      "train_tokens_per_second": 625.724
    },
    {
      "epoch": 1.9118018701036137,
      "grad_norm": 1.4453916549682617,
      "learning_rate": 9.012658227848102e-07,
      "loss": 0.06125418096780777,
      "num_input_tokens_seen": 3561686,
      "step": 1892,
      "train_runtime": 5691.395,
      "train_tokens_per_second": 625.802
    },
    {
      "epoch": 1.9128127369219106,
      "grad_norm": 1.3768304586410522,
      "learning_rate": 8.911392405063292e-07,
      "loss": 0.058950942009687424,
      "num_input_tokens_seen": 3563160,
      "step": 1893,
      "train_runtime": 5693.5401,
      "train_tokens_per_second": 625.825
    },
    {
      "epoch": 1.9138236037402072,
      "grad_norm": 1.6235307455062866,
      "learning_rate": 8.810126582278481e-07,
      "loss": 0.06537044048309326,
      "num_input_tokens_seen": 3564766,
      "step": 1894,
      "train_runtime": 5695.7154,
      "train_tokens_per_second": 625.868
    },
    {
      "epoch": 1.9148344705585039,
      "grad_norm": 1.2254753112792969,
      "learning_rate": 8.708860759493671e-07,
      "loss": 0.04504122585058212,
      "num_input_tokens_seen": 3566548,
      "step": 1895,
      "train_runtime": 5697.9596,
      "train_tokens_per_second": 625.934
    },
    {
      "epoch": 1.9158453373768007,
      "grad_norm": 1.1283148527145386,
      "learning_rate": 8.607594936708862e-07,
      "loss": 0.05272985249757767,
      "num_input_tokens_seen": 3568356,
      "step": 1896,
      "train_runtime": 5700.2025,
      "train_tokens_per_second": 626.005
    },
    {
      "epoch": 1.9168562041950974,
      "grad_norm": 1.8246378898620605,
      "learning_rate": 8.506329113924052e-07,
      "loss": 0.08242052793502808,
      "num_input_tokens_seen": 3570256,
      "step": 1897,
      "train_runtime": 5702.4811,
      "train_tokens_per_second": 626.088
    },
    {
      "epoch": 1.917867071013394,
      "grad_norm": 1.414107084274292,
      "learning_rate": 8.405063291139242e-07,
      "loss": 0.07147019356489182,
      "num_input_tokens_seen": 3572150,
      "step": 1898,
      "train_runtime": 5704.7521,
      "train_tokens_per_second": 626.171
    },
    {
      "epoch": 1.9188779378316907,
      "grad_norm": 1.3033477067947388,
      "learning_rate": 8.303797468354431e-07,
      "loss": 0.07025488466024399,
      "num_input_tokens_seen": 3574394,
      "step": 1899,
      "train_runtime": 5707.1842,
      "train_tokens_per_second": 626.297
    },
    {
      "epoch": 1.9198888046499873,
      "grad_norm": 1.1754509210586548,
      "learning_rate": 8.202531645569621e-07,
      "loss": 0.04968511313199997,
      "num_input_tokens_seen": 3576864,
      "step": 1900,
      "train_runtime": 5709.6344,
      "train_tokens_per_second": 626.461
    },
    {
      "epoch": 1.920899671468284,
      "grad_norm": 1.4616328477859497,
      "learning_rate": 8.101265822784811e-07,
      "loss": 0.05543573200702667,
      "num_input_tokens_seen": 3578852,
      "step": 1901,
      "train_runtime": 5711.9452,
      "train_tokens_per_second": 626.556
    },
    {
      "epoch": 1.9219105382865806,
      "grad_norm": 2.396864414215088,
      "learning_rate": 8.000000000000001e-07,
      "loss": 0.06355445086956024,
      "num_input_tokens_seen": 3580110,
      "step": 1902,
      "train_runtime": 5714.2485,
      "train_tokens_per_second": 626.523
    },
    {
      "epoch": 1.9229214051048773,
      "grad_norm": 1.4294058084487915,
      "learning_rate": 7.89873417721519e-07,
      "loss": 0.060292791575193405,
      "num_input_tokens_seen": 3582264,
      "step": 1903,
      "train_runtime": 5716.7676,
      "train_tokens_per_second": 626.624
    },
    {
      "epoch": 1.9239322719231742,
      "grad_norm": 1.4013493061065674,
      "learning_rate": 7.79746835443038e-07,
      "loss": 0.06805936992168427,
      "num_input_tokens_seen": 3583906,
      "step": 1904,
      "train_runtime": 5718.9845,
      "train_tokens_per_second": 626.668
    },
    {
      "epoch": 1.9249431387414708,
      "grad_norm": 1.514113187789917,
      "learning_rate": 7.69620253164557e-07,
      "loss": 0.07049944251775742,
      "num_input_tokens_seen": 3585726,
      "step": 1905,
      "train_runtime": 5721.2633,
      "train_tokens_per_second": 626.737
    },
    {
      "epoch": 1.9259540055597675,
      "grad_norm": 2.08992075920105,
      "learning_rate": 7.59493670886076e-07,
      "loss": 0.05363238602876663,
      "num_input_tokens_seen": 3587382,
      "step": 1906,
      "train_runtime": 5723.5139,
      "train_tokens_per_second": 626.78
    },
    {
      "epoch": 1.9269648723780644,
      "grad_norm": 1.1218788623809814,
      "learning_rate": 7.493670886075951e-07,
      "loss": 0.0493963249027729,
      "num_input_tokens_seen": 3589402,
      "step": 1907,
      "train_runtime": 5725.8213,
      "train_tokens_per_second": 626.88
    },
    {
      "epoch": 1.927975739196361,
      "grad_norm": 1.096281886100769,
      "learning_rate": 7.39240506329114e-07,
      "loss": 0.0370219424366951,
      "num_input_tokens_seen": 3590808,
      "step": 1908,
      "train_runtime": 5728.0114,
      "train_tokens_per_second": 626.886
    },
    {
      "epoch": 1.9289866060146577,
      "grad_norm": 1.1913776397705078,
      "learning_rate": 7.29113924050633e-07,
      "loss": 0.058016881346702576,
      "num_input_tokens_seen": 3593456,
      "step": 1909,
      "train_runtime": 5730.4887,
      "train_tokens_per_second": 627.077
    },
    {
      "epoch": 1.9299974728329543,
      "grad_norm": 1.1618238687515259,
      "learning_rate": 7.18987341772152e-07,
      "loss": 0.05671081691980362,
      "num_input_tokens_seen": 3595598,
      "step": 1910,
      "train_runtime": 5732.7933,
      "train_tokens_per_second": 627.198
    },
    {
      "epoch": 1.931008339651251,
      "grad_norm": 1.2383519411087036,
      "learning_rate": 7.08860759493671e-07,
      "loss": 0.05270574986934662,
      "num_input_tokens_seen": 3597450,
      "step": 1911,
      "train_runtime": 5735.041,
      "train_tokens_per_second": 627.275
    },
    {
      "epoch": 1.9320192064695476,
      "grad_norm": 1.1728078126907349,
      "learning_rate": 6.9873417721519e-07,
      "loss": 0.05217333883047104,
      "num_input_tokens_seen": 3599768,
      "step": 1912,
      "train_runtime": 5737.4268,
      "train_tokens_per_second": 627.419
    },
    {
      "epoch": 1.9330300732878443,
      "grad_norm": 1.3664581775665283,
      "learning_rate": 6.886075949367089e-07,
      "loss": 0.05507992208003998,
      "num_input_tokens_seen": 3601768,
      "step": 1913,
      "train_runtime": 5739.7315,
      "train_tokens_per_second": 627.515
    },
    {
      "epoch": 1.934040940106141,
      "grad_norm": 1.2706823348999023,
      "learning_rate": 6.784810126582279e-07,
      "loss": 0.059434473514556885,
      "num_input_tokens_seen": 3604040,
      "step": 1914,
      "train_runtime": 5742.1541,
      "train_tokens_per_second": 627.646
    },
    {
      "epoch": 1.9350518069244376,
      "grad_norm": 1.2227834463119507,
      "learning_rate": 6.683544303797469e-07,
      "loss": 0.05084661394357681,
      "num_input_tokens_seen": 3605882,
      "step": 1915,
      "train_runtime": 5744.4029,
      "train_tokens_per_second": 627.721
    },
    {
      "epoch": 1.9360626737427344,
      "grad_norm": 1.2547407150268555,
      "learning_rate": 6.582278481012659e-07,
      "loss": 0.05846727639436722,
      "num_input_tokens_seen": 3607992,
      "step": 1916,
      "train_runtime": 5746.7956,
      "train_tokens_per_second": 627.827
    },
    {
      "epoch": 1.937073540561031,
      "grad_norm": 1.4522916078567505,
      "learning_rate": 6.481012658227849e-07,
      "loss": 0.04499681293964386,
      "num_input_tokens_seen": 3609462,
      "step": 1917,
      "train_runtime": 5748.9518,
      "train_tokens_per_second": 627.847
    },
    {
      "epoch": 1.9380844073793277,
      "grad_norm": 1.0351849794387817,
      "learning_rate": 6.379746835443039e-07,
      "loss": 0.04698522388935089,
      "num_input_tokens_seen": 3611516,
      "step": 1918,
      "train_runtime": 5751.3202,
      "train_tokens_per_second": 627.946
    },
    {
      "epoch": 1.9390952741976246,
      "grad_norm": 1.4694240093231201,
      "learning_rate": 6.278481012658229e-07,
      "loss": 0.043037835508584976,
      "num_input_tokens_seen": 3613642,
      "step": 1919,
      "train_runtime": 5753.6937,
      "train_tokens_per_second": 628.056
    },
    {
      "epoch": 1.9401061410159213,
      "grad_norm": 1.187623143196106,
      "learning_rate": 6.177215189873418e-07,
      "loss": 0.06462160497903824,
      "num_input_tokens_seen": 3615978,
      "step": 1920,
      "train_runtime": 5756.1132,
      "train_tokens_per_second": 628.198
    },
    {
      "epoch": 1.941117007834218,
      "grad_norm": 1.4466322660446167,
      "learning_rate": 6.075949367088608e-07,
      "loss": 0.060681525617837906,
      "num_input_tokens_seen": 3617712,
      "step": 1921,
      "train_runtime": 5773.3292,
      "train_tokens_per_second": 626.625
    },
    {
      "epoch": 1.9421278746525146,
      "grad_norm": 1.4236836433410645,
      "learning_rate": 5.974683544303797e-07,
      "loss": 0.06719941645860672,
      "num_input_tokens_seen": 3619592,
      "step": 1922,
      "train_runtime": 5775.6404,
      "train_tokens_per_second": 626.7
    },
    {
      "epoch": 1.9431387414708112,
      "grad_norm": 1.1856107711791992,
      "learning_rate": 5.873417721518987e-07,
      "loss": 0.047833673655986786,
      "num_input_tokens_seen": 3621670,
      "step": 1923,
      "train_runtime": 5778.0237,
      "train_tokens_per_second": 626.801
    },
    {
      "epoch": 1.9441496082891079,
      "grad_norm": 1.4113465547561646,
      "learning_rate": 5.772151898734177e-07,
      "loss": 0.06201735883951187,
      "num_input_tokens_seen": 3623864,
      "step": 1924,
      "train_runtime": 5780.3819,
      "train_tokens_per_second": 626.925
    },
    {
      "epoch": 1.9451604751074045,
      "grad_norm": 1.6672515869140625,
      "learning_rate": 5.670886075949368e-07,
      "loss": 0.06638215482234955,
      "num_input_tokens_seen": 3625582,
      "step": 1925,
      "train_runtime": 5782.6588,
      "train_tokens_per_second": 626.975
    },
    {
      "epoch": 1.9461713419257012,
      "grad_norm": 1.30471670627594,
      "learning_rate": 5.569620253164558e-07,
      "loss": 0.04921981319785118,
      "num_input_tokens_seen": 3627172,
      "step": 1926,
      "train_runtime": 5784.8724,
      "train_tokens_per_second": 627.01
    },
    {
      "epoch": 1.9471822087439978,
      "grad_norm": 1.170035719871521,
      "learning_rate": 5.468354430379747e-07,
      "loss": 0.033231914043426514,
      "num_input_tokens_seen": 3628574,
      "step": 1927,
      "train_runtime": 5787.0262,
      "train_tokens_per_second": 627.019
    },
    {
      "epoch": 1.9481930755622947,
      "grad_norm": 1.3314383029937744,
      "learning_rate": 5.367088607594937e-07,
      "loss": 0.05085628479719162,
      "num_input_tokens_seen": 3630570,
      "step": 1928,
      "train_runtime": 5789.3341,
      "train_tokens_per_second": 627.114
    },
    {
      "epoch": 1.9492039423805914,
      "grad_norm": 1.1564934253692627,
      "learning_rate": 5.265822784810127e-07,
      "loss": 0.0381021611392498,
      "num_input_tokens_seen": 3632622,
      "step": 1929,
      "train_runtime": 5791.6456,
      "train_tokens_per_second": 627.218
    },
    {
      "epoch": 1.950214809198888,
      "grad_norm": 1.094068169593811,
      "learning_rate": 5.164556962025317e-07,
      "loss": 0.0400090217590332,
      "num_input_tokens_seen": 3634376,
      "step": 1930,
      "train_runtime": 5793.8926,
      "train_tokens_per_second": 627.277
    },
    {
      "epoch": 1.9512256760171849,
      "grad_norm": 1.881561040878296,
      "learning_rate": 5.063291139240507e-07,
      "loss": 0.047163210809230804,
      "num_input_tokens_seen": 3636136,
      "step": 1931,
      "train_runtime": 5796.1417,
      "train_tokens_per_second": 627.337
    },
    {
      "epoch": 1.9522365428354815,
      "grad_norm": 1.5697989463806152,
      "learning_rate": 4.962025316455696e-07,
      "loss": 0.04581909626722336,
      "num_input_tokens_seen": 3637648,
      "step": 1932,
      "train_runtime": 5798.3317,
      "train_tokens_per_second": 627.361
    },
    {
      "epoch": 1.9532474096537782,
      "grad_norm": 1.2818243503570557,
      "learning_rate": 4.860759493670886e-07,
      "loss": 0.040838710963726044,
      "num_input_tokens_seen": 3639374,
      "step": 1933,
      "train_runtime": 5800.5559,
      "train_tokens_per_second": 627.418
    },
    {
      "epoch": 1.9542582764720748,
      "grad_norm": 1.1898375749588013,
      "learning_rate": 4.7594936708860764e-07,
      "loss": 0.045235615223646164,
      "num_input_tokens_seen": 3640998,
      "step": 1934,
      "train_runtime": 5802.7767,
      "train_tokens_per_second": 627.458
    },
    {
      "epoch": 1.9552691432903715,
      "grad_norm": 1.100252389907837,
      "learning_rate": 4.658227848101266e-07,
      "loss": 0.04474688321352005,
      "num_input_tokens_seen": 3643372,
      "step": 1935,
      "train_runtime": 5805.1409,
      "train_tokens_per_second": 627.611
    },
    {
      "epoch": 1.9562800101086681,
      "grad_norm": 1.3565642833709717,
      "learning_rate": 4.556962025316456e-07,
      "loss": 0.06523145735263824,
      "num_input_tokens_seen": 3645376,
      "step": 1936,
      "train_runtime": 5807.382,
      "train_tokens_per_second": 627.714
    },
    {
      "epoch": 1.9572908769269648,
      "grad_norm": 1.2880141735076904,
      "learning_rate": 4.455696202531646e-07,
      "loss": 0.060890838503837585,
      "num_input_tokens_seen": 3647236,
      "step": 1937,
      "train_runtime": 5809.6287,
      "train_tokens_per_second": 627.792
    },
    {
      "epoch": 1.9583017437452614,
      "grad_norm": 1.5652706623077393,
      "learning_rate": 4.3544303797468356e-07,
      "loss": 0.07239330559968948,
      "num_input_tokens_seen": 3649178,
      "step": 1938,
      "train_runtime": 5811.9006,
      "train_tokens_per_second": 627.88
    },
    {
      "epoch": 1.959312610563558,
      "grad_norm": 1.3191746473312378,
      "learning_rate": 4.253164556962026e-07,
      "loss": 0.059902891516685486,
      "num_input_tokens_seen": 3651538,
      "step": 1939,
      "train_runtime": 5814.2999,
      "train_tokens_per_second": 628.027
    },
    {
      "epoch": 1.960323477381855,
      "grad_norm": 0.9755680561065674,
      "learning_rate": 4.1518987341772157e-07,
      "loss": 0.03792475908994675,
      "num_input_tokens_seen": 3653874,
      "step": 1940,
      "train_runtime": 5816.6629,
      "train_tokens_per_second": 628.174
    },
    {
      "epoch": 1.9613343442001516,
      "grad_norm": 1.5447108745574951,
      "learning_rate": 4.0506329113924055e-07,
      "loss": 0.055281493812799454,
      "num_input_tokens_seen": 3655382,
      "step": 1941,
      "train_runtime": 5818.8473,
      "train_tokens_per_second": 628.197
    },
    {
      "epoch": 1.9623452110184483,
      "grad_norm": 1.597838282585144,
      "learning_rate": 3.949367088607595e-07,
      "loss": 0.044146083295345306,
      "num_input_tokens_seen": 3656910,
      "step": 1942,
      "train_runtime": 5821.0344,
      "train_tokens_per_second": 628.223
    },
    {
      "epoch": 1.9633560778367452,
      "grad_norm": 1.4328070878982544,
      "learning_rate": 3.848101265822785e-07,
      "loss": 0.07204818725585938,
      "num_input_tokens_seen": 3658812,
      "step": 1943,
      "train_runtime": 5823.3391,
      "train_tokens_per_second": 628.301
    },
    {
      "epoch": 1.9643669446550418,
      "grad_norm": 1.06763756275177,
      "learning_rate": 3.7468354430379753e-07,
      "loss": 0.04172516614198685,
      "num_input_tokens_seen": 3660618,
      "step": 1944,
      "train_runtime": 5825.5817,
      "train_tokens_per_second": 628.37
    },
    {
      "epoch": 1.9653778114733385,
      "grad_norm": 1.1053352355957031,
      "learning_rate": 3.645569620253165e-07,
      "loss": 0.03737139329314232,
      "num_input_tokens_seen": 3662816,
      "step": 1945,
      "train_runtime": 5827.8859,
      "train_tokens_per_second": 628.498
    },
    {
      "epoch": 1.966388678291635,
      "grad_norm": 1.413617730140686,
      "learning_rate": 3.544303797468355e-07,
      "loss": 0.045816756784915924,
      "num_input_tokens_seen": 3664510,
      "step": 1946,
      "train_runtime": 5830.1387,
      "train_tokens_per_second": 628.546
    },
    {
      "epoch": 1.9673995451099318,
      "grad_norm": 1.3344900608062744,
      "learning_rate": 3.4430379746835447e-07,
      "loss": 0.053659211844205856,
      "num_input_tokens_seen": 3666856,
      "step": 1947,
      "train_runtime": 5832.5396,
      "train_tokens_per_second": 628.689
    },
    {
      "epoch": 1.9684104119282284,
      "grad_norm": 1.4741601943969727,
      "learning_rate": 3.3417721518987345e-07,
      "loss": 0.05253228545188904,
      "num_input_tokens_seen": 3668630,
      "step": 1948,
      "train_runtime": 5834.7812,
      "train_tokens_per_second": 628.752
    },
    {
      "epoch": 1.969421278746525,
      "grad_norm": 1.300722599029541,
      "learning_rate": 3.240506329113924e-07,
      "loss": 0.0691450908780098,
      "num_input_tokens_seen": 3670330,
      "step": 1949,
      "train_runtime": 5837.012,
      "train_tokens_per_second": 628.803
    },
    {
      "epoch": 1.9704321455648217,
      "grad_norm": 1.6856491565704346,
      "learning_rate": 3.1392405063291146e-07,
      "loss": 0.044100165367126465,
      "num_input_tokens_seen": 3672008,
      "step": 1950,
      "train_runtime": 5839.1846,
      "train_tokens_per_second": 628.856
    },
    {
      "epoch": 1.9714430123831184,
      "grad_norm": 1.366133689880371,
      "learning_rate": 3.037974683544304e-07,
      "loss": 0.05769915133714676,
      "num_input_tokens_seen": 3674074,
      "step": 1951,
      "train_runtime": 5853.8862,
      "train_tokens_per_second": 627.63
    },
    {
      "epoch": 1.9724538792014152,
      "grad_norm": 1.6894501447677612,
      "learning_rate": 2.9367088607594936e-07,
      "loss": 0.07025322318077087,
      "num_input_tokens_seen": 3676084,
      "step": 1952,
      "train_runtime": 5856.157,
      "train_tokens_per_second": 627.73
    },
    {
      "epoch": 1.973464746019712,
      "grad_norm": 1.2029181718826294,
      "learning_rate": 2.835443037974684e-07,
      "loss": 0.041464902460575104,
      "num_input_tokens_seen": 3678154,
      "step": 1953,
      "train_runtime": 5858.4511,
      "train_tokens_per_second": 627.837
    },
    {
      "epoch": 1.9744756128380085,
      "grad_norm": 1.5632749795913696,
      "learning_rate": 2.7341772151898737e-07,
      "loss": 0.08382153511047363,
      "num_input_tokens_seen": 3680106,
      "step": 1954,
      "train_runtime": 5860.7258,
      "train_tokens_per_second": 627.927
    },
    {
      "epoch": 1.9754864796563054,
      "grad_norm": 1.28147292137146,
      "learning_rate": 2.6329113924050635e-07,
      "loss": 0.030513249337673187,
      "num_input_tokens_seen": 3682152,
      "step": 1955,
      "train_runtime": 5863.0484,
      "train_tokens_per_second": 628.027
    },
    {
      "epoch": 1.976497346474602,
      "grad_norm": 1.233107089996338,
      "learning_rate": 2.5316455696202533e-07,
      "loss": 0.045211970806121826,
      "num_input_tokens_seen": 3683756,
      "step": 1956,
      "train_runtime": 5865.3086,
      "train_tokens_per_second": 628.058
    },
    {
      "epoch": 1.9775082132928987,
      "grad_norm": 1.6602846384048462,
      "learning_rate": 2.430379746835443e-07,
      "loss": 0.06595426797866821,
      "num_input_tokens_seen": 3685400,
      "step": 1957,
      "train_runtime": 5867.5371,
      "train_tokens_per_second": 628.1
    },
    {
      "epoch": 1.9785190801111954,
      "grad_norm": 1.6605432033538818,
      "learning_rate": 2.329113924050633e-07,
      "loss": 0.07228749990463257,
      "num_input_tokens_seen": 3687110,
      "step": 1958,
      "train_runtime": 5869.757,
      "train_tokens_per_second": 628.154
    },
    {
      "epoch": 1.979529946929492,
      "grad_norm": 1.2205003499984741,
      "learning_rate": 2.227848101265823e-07,
      "loss": 0.045379042625427246,
      "num_input_tokens_seen": 3688784,
      "step": 1959,
      "train_runtime": 5871.9878,
      "train_tokens_per_second": 628.2
    },
    {
      "epoch": 1.9805408137477887,
      "grad_norm": 1.359030842781067,
      "learning_rate": 2.126582278481013e-07,
      "loss": 0.07342518866062164,
      "num_input_tokens_seen": 3690750,
      "step": 1960,
      "train_runtime": 5874.3306,
      "train_tokens_per_second": 628.284
    },
    {
      "epoch": 1.9815516805660853,
      "grad_norm": 1.3784176111221313,
      "learning_rate": 2.0253164556962027e-07,
      "loss": 0.0556640587747097,
      "num_input_tokens_seen": 3692700,
      "step": 1961,
      "train_runtime": 5876.6067,
      "train_tokens_per_second": 628.373
    },
    {
      "epoch": 1.982562547384382,
      "grad_norm": 1.2333017587661743,
      "learning_rate": 1.9240506329113925e-07,
      "loss": 0.04461481422185898,
      "num_input_tokens_seen": 3694398,
      "step": 1962,
      "train_runtime": 5879.1261,
      "train_tokens_per_second": 628.392
    },
    {
      "epoch": 1.9835734142026786,
      "grad_norm": 2.013918876647949,
      "learning_rate": 1.8227848101265826e-07,
      "loss": 0.07952126860618591,
      "num_input_tokens_seen": 3695908,
      "step": 1963,
      "train_runtime": 5881.3504,
      "train_tokens_per_second": 628.411
    },
    {
      "epoch": 1.9845842810209755,
      "grad_norm": 1.2891860008239746,
      "learning_rate": 1.7215189873417723e-07,
      "loss": 0.05664373189210892,
      "num_input_tokens_seen": 3698094,
      "step": 1964,
      "train_runtime": 5883.6952,
      "train_tokens_per_second": 628.533
    },
    {
      "epoch": 1.9855951478392722,
      "grad_norm": 1.332255244255066,
      "learning_rate": 1.620253164556962e-07,
      "loss": 0.06749482452869415,
      "num_input_tokens_seen": 3700184,
      "step": 1965,
      "train_runtime": 5886.0278,
      "train_tokens_per_second": 628.639
    },
    {
      "epoch": 1.986606014657569,
      "grad_norm": 1.4188288450241089,
      "learning_rate": 1.518987341772152e-07,
      "loss": 0.055493321269750595,
      "num_input_tokens_seen": 3701876,
      "step": 1966,
      "train_runtime": 5888.2697,
      "train_tokens_per_second": 628.687
    },
    {
      "epoch": 1.9876168814758657,
      "grad_norm": 1.320956826210022,
      "learning_rate": 1.417721518987342e-07,
      "loss": 0.06059660762548447,
      "num_input_tokens_seen": 3703512,
      "step": 1967,
      "train_runtime": 5890.4485,
      "train_tokens_per_second": 628.732
    },
    {
      "epoch": 1.9886277482941623,
      "grad_norm": 33.71479415893555,
      "learning_rate": 1.3164556962025317e-07,
      "loss": 0.04406532645225525,
      "num_input_tokens_seen": 3705336,
      "step": 1968,
      "train_runtime": 5892.6835,
      "train_tokens_per_second": 628.803
    },
    {
      "epoch": 1.989638615112459,
      "grad_norm": 1.292183518409729,
      "learning_rate": 1.2151898734177215e-07,
      "loss": 0.05553293600678444,
      "num_input_tokens_seen": 3707076,
      "step": 1969,
      "train_runtime": 5894.9226,
      "train_tokens_per_second": 628.859
    },
    {
      "epoch": 1.9906494819307556,
      "grad_norm": 1.9007545709609985,
      "learning_rate": 1.1139240506329114e-07,
      "loss": 0.07554098963737488,
      "num_input_tokens_seen": 3708656,
      "step": 1970,
      "train_runtime": 5897.1043,
      "train_tokens_per_second": 628.894
    },
    {
      "epoch": 1.9916603487490523,
      "grad_norm": 1.814345359802246,
      "learning_rate": 1.0126582278481014e-07,
      "loss": 0.06582854688167572,
      "num_input_tokens_seen": 3710238,
      "step": 1971,
      "train_runtime": 5899.2792,
      "train_tokens_per_second": 628.931
    },
    {
      "epoch": 1.992671215567349,
      "grad_norm": 1.0943604707717896,
      "learning_rate": 9.113924050632913e-08,
      "loss": 0.051717597991228104,
      "num_input_tokens_seen": 3712330,
      "step": 1972,
      "train_runtime": 5901.5442,
      "train_tokens_per_second": 629.044
    },
    {
      "epoch": 1.9936820823856456,
      "grad_norm": 1.5968142747879028,
      "learning_rate": 8.10126582278481e-08,
      "loss": 0.06050843745470047,
      "num_input_tokens_seen": 3713826,
      "step": 1973,
      "train_runtime": 5903.7162,
      "train_tokens_per_second": 629.066
    },
    {
      "epoch": 1.9946929492039422,
      "grad_norm": 1.145068645477295,
      "learning_rate": 7.08860759493671e-08,
      "loss": 0.04888053983449936,
      "num_input_tokens_seen": 3716112,
      "step": 1974,
      "train_runtime": 5906.0585,
      "train_tokens_per_second": 629.203
    },
    {
      "epoch": 1.9957038160222391,
      "grad_norm": 1.3995318412780762,
      "learning_rate": 6.075949367088608e-08,
      "loss": 0.061537325382232666,
      "num_input_tokens_seen": 3717602,
      "step": 1975,
      "train_runtime": 5908.2381,
      "train_tokens_per_second": 629.223
    },
    {
      "epoch": 1.9967146828405358,
      "grad_norm": 0.9206666350364685,
      "learning_rate": 5.063291139240507e-08,
      "loss": 0.02986668422818184,
      "num_input_tokens_seen": 3719202,
      "step": 1976,
      "train_runtime": 5910.4093,
      "train_tokens_per_second": 629.263
    },
    {
      "epoch": 1.9977255496588324,
      "grad_norm": 1.5160229206085205,
      "learning_rate": 4.050632911392405e-08,
      "loss": 0.06555500626564026,
      "num_input_tokens_seen": 3720932,
      "step": 1977,
      "train_runtime": 5912.6278,
      "train_tokens_per_second": 629.32
    },
    {
      "epoch": 1.9987364164771293,
      "grad_norm": 1.421866536140442,
      "learning_rate": 3.037974683544304e-08,
      "loss": 0.0551639124751091,
      "num_input_tokens_seen": 3722462,
      "step": 1978,
      "train_runtime": 5914.7797,
      "train_tokens_per_second": 629.349
    },
    {
      "epoch": 1.999747283295426,
      "grad_norm": 1.2375504970550537,
      "learning_rate": 2.0253164556962027e-08,
      "loss": 0.06062401831150055,
      "num_input_tokens_seen": 3724864,
      "step": 1979,
      "train_runtime": 5917.1893,
      "train_tokens_per_second": 629.499
    },
    {
      "epoch": 2.0,
      "grad_norm": 13.45600414276123,
      "learning_rate": 1.0126582278481013e-08,
      "loss": 0.10668642073869705,
      "num_input_tokens_seen": 3724995,
      "step": 1980,
      "train_runtime": 5917.7901,
      "train_tokens_per_second": 629.457
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.365520179271698,
      "eval_runtime": 54.2315,
      "eval_samples_per_second": 16.208,
      "eval_steps_per_second": 8.113,
      "num_input_tokens_seen": 3724995,
      "step": 1980
    }
  ],
  "logging_steps": 1,
  "max_steps": 1980,
  "num_input_tokens_seen": 3724995,
  "num_train_epochs": 2,
  "save_steps": 30,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.804686815910976e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}