{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 29704,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00033665499596014,
      "grad_norm": 2.460483261683304,
      "learning_rate": 9.999999552565099e-06,
      "loss": 0.9384908676147461,
      "num_input_tokens_seen": 81496,
      "step": 5,
      "train_runtime": 11.6714,
      "train_tokens_per_second": 6982.556
    },
    {
      "epoch": 0.00067330999192028,
      "grad_norm": 2.362196423857104,
      "learning_rate": 9.999997734860948e-06,
      "loss": 0.688481616973877,
      "num_input_tokens_seen": 163416,
      "step": 10,
      "train_runtime": 21.2135,
      "train_tokens_per_second": 7703.414
    },
    {
      "epoch": 0.0010099649878804202,
      "grad_norm": 1.8785417074831652,
      "learning_rate": 9.999994518923374e-06,
      "loss": 0.7289166927337647,
      "num_input_tokens_seen": 245336,
      "step": 15,
      "train_runtime": 30.6845,
      "train_tokens_per_second": 7995.434
    },
    {
      "epoch": 0.00134661998384056,
      "grad_norm": 1.9923094571050737,
      "learning_rate": 9.999989904753278e-06,
      "loss": 0.665110731124878,
      "num_input_tokens_seen": 327256,
      "step": 20,
      "train_runtime": 40.2082,
      "train_tokens_per_second": 8139.043
    },
    {
      "epoch": 0.0016832749798007002,
      "grad_norm": 1.705689832146987,
      "learning_rate": 9.999983892351947e-06,
      "loss": 0.6510555267333984,
      "num_input_tokens_seen": 409176,
      "step": 25,
      "train_runtime": 49.7795,
      "train_tokens_per_second": 8219.769
    },
    {
      "epoch": 0.0020199299757608404,
      "grad_norm": 1.8652662845626249,
      "learning_rate": 9.999976481721066e-06,
      "loss": 0.5777565002441406,
      "num_input_tokens_seen": 491096,
      "step": 30,
      "train_runtime": 59.2145,
      "train_tokens_per_second": 8293.512
    },
    {
      "epoch": 0.0023565849717209803,
      "grad_norm": 1.7451159659623492,
      "learning_rate": 9.999967672862704e-06,
      "loss": 0.609865951538086,
      "num_input_tokens_seen": 572544,
      "step": 35,
      "train_runtime": 69.2122,
      "train_tokens_per_second": 8272.299
    },
    {
      "epoch": 0.00269323996768112,
      "grad_norm": 1.4496605310022321,
      "learning_rate": 9.999957465779326e-06,
      "loss": 0.582823657989502,
      "num_input_tokens_seen": 654464,
      "step": 40,
      "train_runtime": 79.6133,
      "train_tokens_per_second": 8220.533
    },
    {
      "epoch": 0.0030298949636412605,
      "grad_norm": 1.9052867312534216,
      "learning_rate": 9.999945860473788e-06,
      "loss": 0.5875959396362305,
      "num_input_tokens_seen": 736384,
      "step": 45,
      "train_runtime": 89.4343,
      "train_tokens_per_second": 8233.803
    },
    {
      "epoch": 0.0033665499596014004,
      "grad_norm": 1.5567881699567276,
      "learning_rate": 9.999932856949333e-06,
      "loss": 0.5394650459289551,
      "num_input_tokens_seen": 818304,
      "step": 50,
      "train_runtime": 98.9081,
      "train_tokens_per_second": 8273.38
    },
    {
      "epoch": 0.0037032049555615404,
      "grad_norm": 1.907482177848805,
      "learning_rate": 9.999918455209599e-06,
      "loss": 0.5791189193725585,
      "num_input_tokens_seen": 899840,
      "step": 55,
      "train_runtime": 108.7409,
      "train_tokens_per_second": 8275.085
    },
    {
      "epoch": 0.004039859951521681,
      "grad_norm": 1.8210052318717191,
      "learning_rate": 9.99990265525861e-06,
      "loss": 0.5762081146240234,
      "num_input_tokens_seen": 981760,
      "step": 60,
      "train_runtime": 118.2267,
      "train_tokens_per_second": 8304.05
    },
    {
      "epoch": 0.004376514947481821,
      "grad_norm": 1.8929614784553113,
      "learning_rate": 9.999885457100789e-06,
      "loss": 0.602548885345459,
      "num_input_tokens_seen": 1063680,
      "step": 65,
      "train_runtime": 127.7211,
      "train_tokens_per_second": 8328.147
    },
    {
      "epoch": 0.0047131699434419605,
      "grad_norm": 1.5833757591953261,
      "learning_rate": 9.999866860740943e-06,
      "loss": 0.5119287490844726,
      "num_input_tokens_seen": 1145440,
      "step": 70,
      "train_runtime": 137.9004,
      "train_tokens_per_second": 8306.282
    },
    {
      "epoch": 0.005049824939402101,
      "grad_norm": 1.5922049242118776,
      "learning_rate": 9.999846866184274e-06,
      "loss": 0.5539478778839111,
      "num_input_tokens_seen": 1227360,
      "step": 75,
      "train_runtime": 147.3961,
      "train_tokens_per_second": 8326.95
    },
    {
      "epoch": 0.00538647993536224,
      "grad_norm": 3.8989399112505403,
      "learning_rate": 9.99982547343637e-06,
      "loss": 0.496228551864624,
      "num_input_tokens_seen": 1308192,
      "step": 80,
      "train_runtime": 157.0388,
      "train_tokens_per_second": 8330.372
    },
    {
      "epoch": 0.005723134931322381,
      "grad_norm": 1.7476725058334628,
      "learning_rate": 9.999802682503217e-06,
      "loss": 0.5341177463531495,
      "num_input_tokens_seen": 1390112,
      "step": 85,
      "train_runtime": 166.5412,
      "train_tokens_per_second": 8346.955
    },
    {
      "epoch": 0.006059789927282521,
      "grad_norm": 1.6133094937800356,
      "learning_rate": 9.999778493391186e-06,
      "loss": 0.5331961631774902,
      "num_input_tokens_seen": 1472032,
      "step": 90,
      "train_runtime": 176.0405,
      "train_tokens_per_second": 8361.896
    },
    {
      "epoch": 0.0063964449232426605,
      "grad_norm": 1.4986156430168782,
      "learning_rate": 9.999752906107043e-06,
      "loss": 0.5544253826141358,
      "num_input_tokens_seen": 1553952,
      "step": 95,
      "train_runtime": 185.5396,
      "train_tokens_per_second": 8375.31
    },
    {
      "epoch": 0.006733099919202801,
      "grad_norm": 1.5517342677786692,
      "learning_rate": 9.999725920657944e-06,
      "loss": 0.5102095603942871,
      "num_input_tokens_seen": 1635872,
      "step": 100,
      "train_runtime": 195.0341,
      "train_tokens_per_second": 8387.62
    },
    {
      "epoch": 0.007069754915162941,
      "grad_norm": 1.414420182821726,
      "learning_rate": 9.999697537051432e-06,
      "loss": 0.5597661972045899,
      "num_input_tokens_seen": 1717792,
      "step": 105,
      "train_runtime": 204.535,
      "train_tokens_per_second": 8398.524
    },
    {
      "epoch": 0.007406409911123081,
      "grad_norm": 1.7760433404911615,
      "learning_rate": 9.999667755295449e-06,
      "loss": 0.5204813957214356,
      "num_input_tokens_seen": 1799712,
      "step": 110,
      "train_runtime": 214.0349,
      "train_tokens_per_second": 8408.496
    },
    {
      "epoch": 0.007743064907083221,
      "grad_norm": 1.6255614234807807,
      "learning_rate": 9.999636575398319e-06,
      "loss": 0.47191972732543946,
      "num_input_tokens_seen": 1881632,
      "step": 115,
      "train_runtime": 223.5351,
      "train_tokens_per_second": 8417.612
    },
    {
      "epoch": 0.008079719903043361,
      "grad_norm": 1.5685359694014347,
      "learning_rate": 9.999603997368766e-06,
      "loss": 0.48044939041137696,
      "num_input_tokens_seen": 1963032,
      "step": 120,
      "train_runtime": 233.3292,
      "train_tokens_per_second": 8413.143
    },
    {
      "epoch": 0.008416374899003501,
      "grad_norm": 1.5940737801537561,
      "learning_rate": 9.999570021215895e-06,
      "loss": 0.5110182762145996,
      "num_input_tokens_seen": 2044952,
      "step": 125,
      "train_runtime": 242.8286,
      "train_tokens_per_second": 8421.381
    },
    {
      "epoch": 0.008753029894963642,
      "grad_norm": 1.5042862892535902,
      "learning_rate": 9.99953464694921e-06,
      "loss": 0.5660683155059815,
      "num_input_tokens_seen": 2126872,
      "step": 130,
      "train_runtime": 252.3352,
      "train_tokens_per_second": 8428.758
    },
    {
      "epoch": 0.009089684890923782,
      "grad_norm": 1.8038655213424397,
      "learning_rate": 9.999497874578606e-06,
      "loss": 0.5527376174926758,
      "num_input_tokens_seen": 2208792,
      "step": 135,
      "train_runtime": 261.85,
      "train_tokens_per_second": 8435.332
    },
    {
      "epoch": 0.009426339886883921,
      "grad_norm": 1.5356420208237254,
      "learning_rate": 9.999459704114361e-06,
      "loss": 0.4918352603912354,
      "num_input_tokens_seen": 2290712,
      "step": 140,
      "train_runtime": 271.3485,
      "train_tokens_per_second": 8441.955
    },
    {
      "epoch": 0.00976299488284406,
      "grad_norm": 1.42284237982547,
      "learning_rate": 9.999420135567153e-06,
      "loss": 0.5440413951873779,
      "num_input_tokens_seen": 2372632,
      "step": 145,
      "train_runtime": 280.8421,
      "train_tokens_per_second": 8448.279
    },
    {
      "epoch": 0.010099649878804202,
      "grad_norm": 1.7076102890810851,
      "learning_rate": 9.999379168948047e-06,
      "loss": 0.5866301536560059,
      "num_input_tokens_seen": 2454552,
      "step": 150,
      "train_runtime": 290.4493,
      "train_tokens_per_second": 8450.879
    },
    {
      "epoch": 0.010436304874764341,
      "grad_norm": 1.7879663390540963,
      "learning_rate": 9.999336804268496e-06,
      "loss": 0.4952958106994629,
      "num_input_tokens_seen": 2536472,
      "step": 155,
      "train_runtime": 299.9489,
      "train_tokens_per_second": 8456.346
    },
    {
      "epoch": 0.01077295987072448,
      "grad_norm": 1.6727669597050367,
      "learning_rate": 9.999293041540351e-06,
      "loss": 0.5209620475769043,
      "num_input_tokens_seen": 2618392,
      "step": 160,
      "train_runtime": 309.4412,
      "train_tokens_per_second": 8461.678
    },
    {
      "epoch": 0.011109614866684622,
      "grad_norm": 1.5192318100076687,
      "learning_rate": 9.999247880775849e-06,
      "loss": 0.5333065509796142,
      "num_input_tokens_seen": 2700312,
      "step": 165,
      "train_runtime": 318.9366,
      "train_tokens_per_second": 8466.61
    },
    {
      "epoch": 0.011446269862644761,
      "grad_norm": 1.8192000644830455,
      "learning_rate": 9.999201321987617e-06,
      "loss": 0.4927525043487549,
      "num_input_tokens_seen": 2782232,
      "step": 170,
      "train_runtime": 328.5332,
      "train_tokens_per_second": 8468.647
    },
    {
      "epoch": 0.011782924858604901,
      "grad_norm": 1.7177540644461324,
      "learning_rate": 9.999153365188676e-06,
      "loss": 0.5407365798950196,
      "num_input_tokens_seen": 2864152,
      "step": 175,
      "train_runtime": 338.1317,
      "train_tokens_per_second": 8470.522
    },
    {
      "epoch": 0.012119579854565042,
      "grad_norm": 1.9257743952127973,
      "learning_rate": 9.99910401039244e-06,
      "loss": 0.5079368591308594,
      "num_input_tokens_seen": 2945832,
      "step": 180,
      "train_runtime": 348.1093,
      "train_tokens_per_second": 8462.377
    },
    {
      "epoch": 0.012456234850525182,
      "grad_norm": 1.8439849270039301,
      "learning_rate": 9.999053257612708e-06,
      "loss": 0.48992328643798827,
      "num_input_tokens_seen": 3027752,
      "step": 185,
      "train_runtime": 357.6147,
      "train_tokens_per_second": 8466.519
    },
    {
      "epoch": 0.012792889846485321,
      "grad_norm": 1.6334293356385416,
      "learning_rate": 9.999001106863672e-06,
      "loss": 0.5860056400299072,
      "num_input_tokens_seen": 3109672,
      "step": 190,
      "train_runtime": 367.1133,
      "train_tokens_per_second": 8470.607
    },
    {
      "epoch": 0.013129544842445462,
      "grad_norm": 1.5995124582894358,
      "learning_rate": 9.998947558159916e-06,
      "loss": 0.5179404735565185,
      "num_input_tokens_seen": 3191592,
      "step": 195,
      "train_runtime": 376.8028,
      "train_tokens_per_second": 8470.192
    },
    {
      "epoch": 0.013466199838405602,
      "grad_norm": 1.6738904223536097,
      "learning_rate": 9.99889261151642e-06,
      "loss": 0.5358803749084473,
      "num_input_tokens_seen": 3273232,
      "step": 200,
      "train_runtime": 386.7377,
      "train_tokens_per_second": 8463.7
    },
    {
      "epoch": 0.013802854834365741,
      "grad_norm": 1.4475817592439468,
      "learning_rate": 9.998836266948542e-06,
      "loss": 0.45974125862121584,
      "num_input_tokens_seen": 3355152,
      "step": 205,
      "train_runtime": 396.3795,
      "train_tokens_per_second": 8464.495
    },
    {
      "epoch": 0.014139509830325882,
      "grad_norm": 1.5473848323239392,
      "learning_rate": 9.998778524472043e-06,
      "loss": 0.5021562576293945,
      "num_input_tokens_seen": 3437072,
      "step": 210,
      "train_runtime": 405.9166,
      "train_tokens_per_second": 8467.433
    },
    {
      "epoch": 0.014476164826286022,
      "grad_norm": 1.910746680484798,
      "learning_rate": 9.99871938410307e-06,
      "loss": 0.49573307037353515,
      "num_input_tokens_seen": 3518992,
      "step": 215,
      "train_runtime": 415.5245,
      "train_tokens_per_second": 8468.795
    },
    {
      "epoch": 0.014812819822246161,
      "grad_norm": 1.59442781821053,
      "learning_rate": 9.99865884585816e-06,
      "loss": 0.5263462066650391,
      "num_input_tokens_seen": 3600264,
      "step": 220,
      "train_runtime": 425.2896,
      "train_tokens_per_second": 8465.44
    },
    {
      "epoch": 0.015149474818206303,
      "grad_norm": 1.6760205948235518,
      "learning_rate": 9.998596909754245e-06,
      "loss": 0.5159770011901855,
      "num_input_tokens_seen": 3682184,
      "step": 225,
      "train_runtime": 434.8551,
      "train_tokens_per_second": 8467.611
    },
    {
      "epoch": 0.015486129814166442,
      "grad_norm": 1.6537246034201292,
      "learning_rate": 9.998533575808642e-06,
      "loss": 0.5088547706604004,
      "num_input_tokens_seen": 3763952,
      "step": 230,
      "train_runtime": 444.9303,
      "train_tokens_per_second": 8459.645
    },
    {
      "epoch": 0.015822784810126583,
      "grad_norm": 1.6613299102875403,
      "learning_rate": 9.998468844039064e-06,
      "loss": 0.4808199405670166,
      "num_input_tokens_seen": 3845872,
      "step": 235,
      "train_runtime": 454.5475,
      "train_tokens_per_second": 8460.881
    },
    {
      "epoch": 0.016159439806086723,
      "grad_norm": 1.3130596864365738,
      "learning_rate": 9.998402714463613e-06,
      "loss": 0.5082195281982422,
      "num_input_tokens_seen": 3927792,
      "step": 240,
      "train_runtime": 464.0433,
      "train_tokens_per_second": 8464.279
    },
    {
      "epoch": 0.016496094802046862,
      "grad_norm": 1.678808224786986,
      "learning_rate": 9.998335187100782e-06,
      "loss": 0.568327522277832,
      "num_input_tokens_seen": 4009712,
      "step": 245,
      "train_runtime": 473.6743,
      "train_tokens_per_second": 8465.124
    },
    {
      "epoch": 0.016832749798007002,
      "grad_norm": 1.3976624147643106,
      "learning_rate": 9.998266261969453e-06,
      "loss": 0.5340527534484864,
      "num_input_tokens_seen": 4091632,
      "step": 250,
      "train_runtime": 483.2966,
      "train_tokens_per_second": 8466.089
    },
    {
      "epoch": 0.01716940479396714,
      "grad_norm": 1.6859484837546082,
      "learning_rate": 9.998195939088905e-06,
      "loss": 0.5228978633880615,
      "num_input_tokens_seen": 4173552,
      "step": 255,
      "train_runtime": 492.915,
      "train_tokens_per_second": 8467.083
    },
    {
      "epoch": 0.017506059789927284,
      "grad_norm": 1.517987502630828,
      "learning_rate": 9.998124218478797e-06,
      "loss": 0.4781132698059082,
      "num_input_tokens_seen": 4255472,
      "step": 260,
      "train_runtime": 502.4175,
      "train_tokens_per_second": 8469.992
    },
    {
      "epoch": 0.017842714785887424,
      "grad_norm": 1.8456594240545514,
      "learning_rate": 9.998051100159192e-06,
      "loss": 0.5052493572235107,
      "num_input_tokens_seen": 4337392,
      "step": 265,
      "train_runtime": 512.0325,
      "train_tokens_per_second": 8470.93
    },
    {
      "epoch": 0.018179369781847563,
      "grad_norm": 1.6278165421414545,
      "learning_rate": 9.997976584150533e-06,
      "loss": 0.5074337005615235,
      "num_input_tokens_seen": 4418696,
      "step": 270,
      "train_runtime": 521.9342,
      "train_tokens_per_second": 8466.002
    },
    {
      "epoch": 0.018516024777807703,
      "grad_norm": 1.5791239749740076,
      "learning_rate": 9.997900670473662e-06,
      "loss": 0.4905442237854004,
      "num_input_tokens_seen": 4500616,
      "step": 275,
      "train_runtime": 531.5369,
      "train_tokens_per_second": 8467.175
    },
    {
      "epoch": 0.018852679773767842,
      "grad_norm": 1.957691814855914,
      "learning_rate": 9.997823359149804e-06,
      "loss": 0.46509718894958496,
      "num_input_tokens_seen": 4582536,
      "step": 280,
      "train_runtime": 541.0369,
      "train_tokens_per_second": 8469.914
    },
    {
      "epoch": 0.01918933476972798,
      "grad_norm": 1.6839820922276703,
      "learning_rate": 9.997744650200579e-06,
      "loss": 0.5586575508117676,
      "num_input_tokens_seen": 4664456,
      "step": 285,
      "train_runtime": 550.6246,
      "train_tokens_per_second": 8471.208
    },
    {
      "epoch": 0.01952598976568812,
      "grad_norm": 1.9530963316687509,
      "learning_rate": 9.997664543648e-06,
      "loss": 0.4851670265197754,
      "num_input_tokens_seen": 4746376,
      "step": 290,
      "train_runtime": 560.2233,
      "train_tokens_per_second": 8472.293
    },
    {
      "epoch": 0.019862644761648264,
      "grad_norm": 1.4988948075922344,
      "learning_rate": 9.997583039514469e-06,
      "loss": 0.5149078369140625,
      "num_input_tokens_seen": 4828296,
      "step": 295,
      "train_runtime": 569.7912,
      "train_tokens_per_second": 8473.799
    },
    {
      "epoch": 0.020199299757608404,
      "grad_norm": 1.6086969806615214,
      "learning_rate": 9.997500137822776e-06,
      "loss": 0.5039136409759521,
      "num_input_tokens_seen": 4910216,
      "step": 300,
      "train_runtime": 579.3729,
      "train_tokens_per_second": 8475.053
    },
    {
      "epoch": 0.020535954753568543,
      "grad_norm": 2.492104329619339,
      "learning_rate": 9.997415838596106e-06,
      "loss": 0.4922384738922119,
      "num_input_tokens_seen": 4991992,
      "step": 305,
      "train_runtime": 589.5005,
      "train_tokens_per_second": 8468.173
    },
    {
      "epoch": 0.020872609749528682,
      "grad_norm": 1.570732064839653,
      "learning_rate": 9.997330141858032e-06,
      "loss": 0.539021635055542,
      "num_input_tokens_seen": 5073912,
      "step": 310,
      "train_runtime": 599.0972,
      "train_tokens_per_second": 8469.263
    },
    {
      "epoch": 0.021209264745488822,
      "grad_norm": 1.4516500755537687,
      "learning_rate": 9.997243047632519e-06,
      "loss": 0.5187999248504639,
      "num_input_tokens_seen": 5155832,
      "step": 315,
      "train_runtime": 608.6024,
      "train_tokens_per_second": 8471.594
    },
    {
      "epoch": 0.02154591974144896,
      "grad_norm": 1.5070900114520722,
      "learning_rate": 9.997154555943923e-06,
      "loss": 0.4746111869812012,
      "num_input_tokens_seen": 5237752,
      "step": 320,
      "train_runtime": 618.2342,
      "train_tokens_per_second": 8472.116
    },
    {
      "epoch": 0.021882574737409104,
      "grad_norm": 1.6233504331387838,
      "learning_rate": 9.997064666816992e-06,
      "loss": 0.4469716548919678,
      "num_input_tokens_seen": 5319536,
      "step": 325,
      "train_runtime": 628.2523,
      "train_tokens_per_second": 8467.197
    },
    {
      "epoch": 0.022219229733369244,
      "grad_norm": 1.631296232880249,
      "learning_rate": 9.996973380276858e-06,
      "loss": 0.6025129318237304,
      "num_input_tokens_seen": 5401456,
      "step": 330,
      "train_runtime": 637.8987,
      "train_tokens_per_second": 8467.577
    },
    {
      "epoch": 0.022555884729329383,
      "grad_norm": 1.5724805154706403,
      "learning_rate": 9.996880696349053e-06,
      "loss": 0.5071794033050537,
      "num_input_tokens_seen": 5483152,
      "step": 335,
      "train_runtime": 647.501,
      "train_tokens_per_second": 8468.176
    },
    {
      "epoch": 0.022892539725289523,
      "grad_norm": 1.7318437570286755,
      "learning_rate": 9.996786615059497e-06,
      "loss": 0.5633901596069336,
      "num_input_tokens_seen": 5565072,
      "step": 340,
      "train_runtime": 657.1064,
      "train_tokens_per_second": 8469.058
    },
    {
      "epoch": 0.023229194721249662,
      "grad_norm": 1.3463749029504954,
      "learning_rate": 9.996691136434498e-06,
      "loss": 0.44980955123901367,
      "num_input_tokens_seen": 5646992,
      "step": 345,
      "train_runtime": 666.6154,
      "train_tokens_per_second": 8471.14
    },
    {
      "epoch": 0.023565849717209802,
      "grad_norm": 1.442452458730031,
      "learning_rate": 9.996594260500753e-06,
      "loss": 0.5020045280456543,
      "num_input_tokens_seen": 5728912,
      "step": 350,
      "train_runtime": 676.2449,
      "train_tokens_per_second": 8471.653
    },
    {
      "epoch": 0.023902504713169945,
      "grad_norm": 1.550311844324363,
      "learning_rate": 9.996495987285358e-06,
      "loss": 0.5040197849273682,
      "num_input_tokens_seen": 5810832,
      "step": 355,
      "train_runtime": 685.8573,
      "train_tokens_per_second": 8472.363
    },
    {
      "epoch": 0.024239159709130084,
      "grad_norm": 1.6445100382528421,
      "learning_rate": 9.996396316815793e-06,
      "loss": 0.4483181953430176,
      "num_input_tokens_seen": 5892752,
      "step": 360,
      "train_runtime": 695.3736,
      "train_tokens_per_second": 8474.224
    },
    {
      "epoch": 0.024575814705090224,
      "grad_norm": 1.5103348998471153,
      "learning_rate": 9.99629524911993e-06,
      "loss": 0.5478151321411133,
      "num_input_tokens_seen": 5974672,
      "step": 365,
      "train_runtime": 704.9864,
      "train_tokens_per_second": 8474.876
    },
    {
      "epoch": 0.024912469701050363,
      "grad_norm": 1.589060394649911,
      "learning_rate": 9.996192784226033e-06,
      "loss": 0.4783379077911377,
      "num_input_tokens_seen": 6056592,
      "step": 370,
      "train_runtime": 714.5948,
      "train_tokens_per_second": 8475.561
    },
    {
      "epoch": 0.025249124697010503,
      "grad_norm": 1.539399245282955,
      "learning_rate": 9.996088922162755e-06,
      "loss": 0.49391946792602537,
      "num_input_tokens_seen": 6138512,
      "step": 375,
      "train_runtime": 724.1427,
      "train_tokens_per_second": 8476.937
    },
    {
      "epoch": 0.025585779692970642,
      "grad_norm": 1.6245183387616275,
      "learning_rate": 9.995983662959141e-06,
      "loss": 0.46095852851867675,
      "num_input_tokens_seen": 6220216,
      "step": 380,
      "train_runtime": 734.1846,
      "train_tokens_per_second": 8472.278
    },
    {
      "epoch": 0.025922434688930785,
      "grad_norm": 1.5399780285394966,
      "learning_rate": 9.995877006644627e-06,
      "loss": 0.4956529140472412,
      "num_input_tokens_seen": 6301760,
      "step": 385,
      "train_runtime": 744.1863,
      "train_tokens_per_second": 8467.987
    },
    {
      "epoch": 0.026259089684890925,
      "grad_norm": 1.6419338800523573,
      "learning_rate": 9.995768953249038e-06,
      "loss": 0.5725302696228027,
      "num_input_tokens_seen": 6383680,
      "step": 390,
      "train_runtime": 753.7907,
      "train_tokens_per_second": 8468.77
    },
    {
      "epoch": 0.026595744680851064,
      "grad_norm": 1.442232288599664,
      "learning_rate": 9.995659502802594e-06,
      "loss": 0.47279911041259765,
      "num_input_tokens_seen": 6465600,
      "step": 395,
      "train_runtime": 763.2963,
      "train_tokens_per_second": 8470.63
    },
    {
      "epoch": 0.026932399676811204,
      "grad_norm": 1.4296907627490385,
      "learning_rate": 9.995548655335899e-06,
      "loss": 0.48623270988464357,
      "num_input_tokens_seen": 6547520,
      "step": 400,
      "train_runtime": 772.9075,
      "train_tokens_per_second": 8471.285
    },
    {
      "epoch": 0.027269054672771343,
      "grad_norm": 1.5246175800637782,
      "learning_rate": 9.995436410879951e-06,
      "loss": 0.5539831161499024,
      "num_input_tokens_seen": 6629440,
      "step": 405,
      "train_runtime": 782.506,
      "train_tokens_per_second": 8472.063
    },
    {
      "epoch": 0.027605709668731482,
      "grad_norm": 1.4695619773077833,
      "learning_rate": 9.99532276946614e-06,
      "loss": 0.5077695846557617,
      "num_input_tokens_seen": 6711360,
      "step": 410,
      "train_runtime": 792.0107,
      "train_tokens_per_second": 8473.824
    },
    {
      "epoch": 0.027942364664691625,
      "grad_norm": 1.6121338787384554,
      "learning_rate": 9.995207731126246e-06,
      "loss": 0.48922386169433596,
      "num_input_tokens_seen": 6793280,
      "step": 415,
      "train_runtime": 801.6394,
      "train_tokens_per_second": 8474.234
    },
    {
      "epoch": 0.028279019660651765,
      "grad_norm": 2.0516685707854863,
      "learning_rate": 9.99509129589244e-06,
      "loss": 0.5176673889160156,
      "num_input_tokens_seen": 6875200,
      "step": 420,
      "train_runtime": 811.2429,
      "train_tokens_per_second": 8474.897
    },
    {
      "epoch": 0.028615674656611904,
      "grad_norm": 1.4338444068227223,
      "learning_rate": 9.994973463797279e-06,
      "loss": 0.4645550727844238,
      "num_input_tokens_seen": 6957120,
      "step": 425,
      "train_runtime": 820.7996,
      "train_tokens_per_second": 8476.028
    },
    {
      "epoch": 0.028952329652572044,
      "grad_norm": 1.3655617715033017,
      "learning_rate": 9.99485423487372e-06,
      "loss": 0.46114482879638674,
      "num_input_tokens_seen": 7039040,
      "step": 430,
      "train_runtime": 830.3626,
      "train_tokens_per_second": 8477.067
    },
    {
      "epoch": 0.029288984648532183,
      "grad_norm": 1.4781980310557956,
      "learning_rate": 9.994733609155097e-06,
      "loss": 0.49649477005004883,
      "num_input_tokens_seen": 7120960,
      "step": 435,
      "train_runtime": 839.9508,
      "train_tokens_per_second": 8477.83
    },
    {
      "epoch": 0.029625639644492323,
      "grad_norm": 1.651595046842109,
      "learning_rate": 9.994611586675149e-06,
      "loss": 0.4739995956420898,
      "num_input_tokens_seen": 7202880,
      "step": 440,
      "train_runtime": 849.5673,
      "train_tokens_per_second": 8478.292
    },
    {
      "epoch": 0.029962294640452466,
      "grad_norm": 1.65626325964033,
      "learning_rate": 9.994488167468e-06,
      "loss": 0.47940568923950194,
      "num_input_tokens_seen": 7284800,
      "step": 445,
      "train_runtime": 859.0674,
      "train_tokens_per_second": 8479.893
    },
    {
      "epoch": 0.030298949636412605,
      "grad_norm": 1.3893669075117465,
      "learning_rate": 9.994363351568158e-06,
      "loss": 0.4985339164733887,
      "num_input_tokens_seen": 7366720,
      "step": 450,
      "train_runtime": 868.6665,
      "train_tokens_per_second": 8480.493
    },
    {
      "epoch": 0.030635604632372745,
      "grad_norm": 1.3893372305028011,
      "learning_rate": 9.994237139010532e-06,
      "loss": 0.4492755889892578,
      "num_input_tokens_seen": 7448640,
      "step": 455,
      "train_runtime": 878.272,
      "train_tokens_per_second": 8481.017
    },
    {
      "epoch": 0.030972259628332884,
      "grad_norm": 1.568824559101899,
      "learning_rate": 9.994109529830414e-06,
      "loss": 0.5168972969055176,
      "num_input_tokens_seen": 7530560,
      "step": 460,
      "train_runtime": 887.769,
      "train_tokens_per_second": 8482.567
    },
    {
      "epoch": 0.03130891462429303,
      "grad_norm": 1.4859793935912373,
      "learning_rate": 9.993980524063494e-06,
      "loss": 0.5170444488525391,
      "num_input_tokens_seen": 7612480,
      "step": 465,
      "train_runtime": 897.3712,
      "train_tokens_per_second": 8483.089
    },
    {
      "epoch": 0.03164556962025317,
      "grad_norm": 1.446577209090975,
      "learning_rate": 9.993850121745842e-06,
      "loss": 0.5123265743255615,
      "num_input_tokens_seen": 7694400,
      "step": 470,
      "train_runtime": 906.9927,
      "train_tokens_per_second": 8483.42
    },
    {
      "epoch": 0.031982224616213306,
      "grad_norm": 1.7276148436191094,
      "learning_rate": 9.99371832291393e-06,
      "loss": 0.4632590293884277,
      "num_input_tokens_seen": 7776320,
      "step": 475,
      "train_runtime": 916.4955,
      "train_tokens_per_second": 8484.843
    },
    {
      "epoch": 0.032318879612173446,
      "grad_norm": 1.5925527123726595,
      "learning_rate": 9.993585127604612e-06,
      "loss": 0.5060209274291992,
      "num_input_tokens_seen": 7858240,
      "step": 480,
      "train_runtime": 926.1212,
      "train_tokens_per_second": 8485.109
    },
    {
      "epoch": 0.032655534608133585,
      "grad_norm": 1.4868404160044988,
      "learning_rate": 9.993450535855136e-06,
      "loss": 0.5028887748718261,
      "num_input_tokens_seen": 7938856,
      "step": 485,
      "train_runtime": 936.1931,
      "train_tokens_per_second": 8479.935
    },
    {
      "epoch": 0.032992189604093725,
      "grad_norm": 1.6202340286813566,
      "learning_rate": 9.993314547703142e-06,
      "loss": 0.47409682273864745,
      "num_input_tokens_seen": 8020776,
      "step": 490,
      "train_runtime": 945.7384,
      "train_tokens_per_second": 8480.966
    },
    {
      "epoch": 0.033328844600053864,
      "grad_norm": 1.5180343966359493,
      "learning_rate": 9.993177163186655e-06,
      "loss": 0.484273624420166,
      "num_input_tokens_seen": 8102696,
      "step": 495,
      "train_runtime": 955.2965,
      "train_tokens_per_second": 8481.865
    },
    {
      "epoch": 0.033665499596014004,
      "grad_norm": 1.3991620953494934,
      "learning_rate": 9.993038382344097e-06,
      "loss": 0.4703634262084961,
      "num_input_tokens_seen": 8184616,
      "step": 500,
      "train_runtime": 964.8932,
      "train_tokens_per_second": 8482.406
    },
    {
      "epoch": 0.03400215459197414,
      "grad_norm": 2.0143522368846654,
      "learning_rate": 9.992898205214279e-06,
      "loss": 0.4855672836303711,
      "num_input_tokens_seen": 8266536,
      "step": 505,
      "train_runtime": 974.4484,
      "train_tokens_per_second": 8483.298
    },
    {
      "epoch": 0.03433880958793428,
      "grad_norm": 1.5083710462546542,
      "learning_rate": 9.992756631836397e-06,
      "loss": 0.4965198516845703,
      "num_input_tokens_seen": 8348456,
      "step": 510,
      "train_runtime": 984.0293,
      "train_tokens_per_second": 8483.951
    },
    {
      "epoch": 0.03467546458389442,
      "grad_norm": 1.3938883513309688,
      "learning_rate": 9.992613662250044e-06,
      "loss": 0.4315831184387207,
      "num_input_tokens_seen": 8430376,
      "step": 515,
      "train_runtime": 993.6336,
      "train_tokens_per_second": 8484.391
    },
    {
      "epoch": 0.03501211957985457,
      "grad_norm": 1.3606457338317033,
      "learning_rate": 9.992469296495201e-06,
      "loss": 0.4850588798522949,
      "num_input_tokens_seen": 8512296,
      "step": 520,
      "train_runtime": 1003.2488,
      "train_tokens_per_second": 8484.731
    },
    {
      "epoch": 0.03534877457581471,
      "grad_norm": 1.4176232779305433,
      "learning_rate": 9.992323534612238e-06,
      "loss": 0.4297802448272705,
      "num_input_tokens_seen": 8594216,
      "step": 525,
      "train_runtime": 1012.755,
      "train_tokens_per_second": 8485.977
    },
    {
      "epoch": 0.03568542957177485,
      "grad_norm": 2.2867334694716606,
      "learning_rate": 9.99217637664192e-06,
      "loss": 0.5113659381866456,
      "num_input_tokens_seen": 8674888,
      "step": 530,
      "train_runtime": 1022.3575,
      "train_tokens_per_second": 8485.181
    },
    {
      "epoch": 0.03602208456773499,
      "grad_norm": 1.5197241122394904,
      "learning_rate": 9.992027822625397e-06,
      "loss": 0.49874114990234375,
      "num_input_tokens_seen": 8756808,
      "step": 535,
      "train_runtime": 1031.9692,
      "train_tokens_per_second": 8485.532
    },
    {
      "epoch": 0.036358739563695126,
      "grad_norm": 1.6320372111755794,
      "learning_rate": 9.99187787260421e-06,
      "loss": 0.45468988418579104,
      "num_input_tokens_seen": 8838728,
      "step": 540,
      "train_runtime": 1041.4823,
      "train_tokens_per_second": 8486.681
    },
    {
      "epoch": 0.036695394559655266,
      "grad_norm": 1.4494713500476573,
      "learning_rate": 9.991726526620294e-06,
      "loss": 0.46222734451293945,
      "num_input_tokens_seen": 8920648,
      "step": 545,
      "train_runtime": 1051.1069,
      "train_tokens_per_second": 8486.909
    },
    {
      "epoch": 0.037032049555615405,
      "grad_norm": 1.647148438778568,
      "learning_rate": 9.991573784715973e-06,
      "loss": 0.5329260349273681,
      "num_input_tokens_seen": 9002568,
      "step": 550,
      "train_runtime": 1060.7282,
      "train_tokens_per_second": 8487.158
    },
    {
      "epoch": 0.037368704551575545,
      "grad_norm": 1.8811313110801269,
      "learning_rate": 9.99141964693396e-06,
      "loss": 0.5072747230529785,
      "num_input_tokens_seen": 9084488,
      "step": 555,
      "train_runtime": 1070.2356,
      "train_tokens_per_second": 8488.307
    },
    {
      "epoch": 0.037705359547535684,
      "grad_norm": 1.3155188584292783,
      "learning_rate": 9.99126411331736e-06,
      "loss": 0.509545087814331,
      "num_input_tokens_seen": 9166408,
      "step": 560,
      "train_runtime": 1079.8374,
      "train_tokens_per_second": 8488.693
    },
    {
      "epoch": 0.038042014543495824,
      "grad_norm": 1.4523388426673158,
      "learning_rate": 9.991107183909665e-06,
      "loss": 0.5235973834991455,
      "num_input_tokens_seen": 9248328,
      "step": 565,
      "train_runtime": 1089.442,
      "train_tokens_per_second": 8489.051
    },
    {
      "epoch": 0.03837866953945596,
      "grad_norm": 1.480953692651925,
      "learning_rate": 9.990948858754763e-06,
      "loss": 0.46100525856018065,
      "num_input_tokens_seen": 9330248,
      "step": 570,
      "train_runtime": 1098.9814,
      "train_tokens_per_second": 8489.905
    },
    {
      "epoch": 0.0387153245354161,
      "grad_norm": 1.624469909998911,
      "learning_rate": 9.990789137896927e-06,
      "loss": 0.4867257118225098,
      "num_input_tokens_seen": 9412168,
      "step": 575,
      "train_runtime": 1108.5578,
      "train_tokens_per_second": 8490.462
    },
    {
      "epoch": 0.03905197953137624,
      "grad_norm": 1.4248197336228023,
      "learning_rate": 9.990628021380823e-06,
      "loss": 0.5077487468719483,
      "num_input_tokens_seen": 9494088,
      "step": 580,
      "train_runtime": 1118.177,
      "train_tokens_per_second": 8490.684
    },
    {
      "epoch": 0.03938863452733639,
      "grad_norm": 1.4387564556020163,
      "learning_rate": 9.990465509251508e-06,
      "loss": 0.5649009704589844,
      "num_input_tokens_seen": 9576008,
      "step": 585,
      "train_runtime": 1127.7148,
      "train_tokens_per_second": 8491.515
    },
    {
      "epoch": 0.03972528952329653,
      "grad_norm": 1.429009901599511,
      "learning_rate": 9.990301601554425e-06,
      "loss": 0.47269377708435056,
      "num_input_tokens_seen": 9657928,
      "step": 590,
      "train_runtime": 1137.2693,
      "train_tokens_per_second": 8492.208
    },
    {
      "epoch": 0.04006194451925667,
      "grad_norm": 1.5049728708789871,
      "learning_rate": 9.990136298335414e-06,
      "loss": 0.4184258460998535,
      "num_input_tokens_seen": 9739848,
      "step": 595,
      "train_runtime": 1146.8752,
      "train_tokens_per_second": 8492.509
    },
    {
      "epoch": 0.04039859951521681,
      "grad_norm": 1.5637028236833304,
      "learning_rate": 9.989969599640698e-06,
      "loss": 0.445817232131958,
      "num_input_tokens_seen": 9821768,
      "step": 600,
      "train_runtime": 1156.4198,
      "train_tokens_per_second": 8493.255
    },
    {
      "epoch": 0.040735254511176947,
      "grad_norm": 1.427047814096542,
      "learning_rate": 9.989801505516895e-06,
      "loss": 0.5265957832336425,
      "num_input_tokens_seen": 9903688,
      "step": 605,
      "train_runtime": 1165.9878,
      "train_tokens_per_second": 8493.818
    },
    {
      "epoch": 0.041071909507137086,
      "grad_norm": 1.59998610482266,
      "learning_rate": 9.989632016011012e-06,
      "loss": 0.5175544738769531,
      "num_input_tokens_seen": 9985608,
      "step": 610,
      "train_runtime": 1175.6082,
      "train_tokens_per_second": 8493.993
    },
    {
      "epoch": 0.041408564503097225,
      "grad_norm": 1.3984816846731087,
      "learning_rate": 9.989461131170448e-06,
      "loss": 0.4851548671722412,
      "num_input_tokens_seen": 10067528,
      "step": 615,
      "train_runtime": 1185.2406,
      "train_tokens_per_second": 8494.08
    },
    {
      "epoch": 0.041745219499057365,
      "grad_norm": 1.3956446954844701,
      "learning_rate": 9.98928885104299e-06,
      "loss": 0.4202116966247559,
      "num_input_tokens_seen": 10149448,
      "step": 620,
      "train_runtime": 1194.7431,
      "train_tokens_per_second": 8495.088
    },
    {
      "epoch": 0.042081874495017504,
      "grad_norm": 1.47064753444382,
      "learning_rate": 9.989115175676812e-06,
      "loss": 0.4495251655578613,
      "num_input_tokens_seen": 10231368,
      "step": 625,
      "train_runtime": 1204.3589,
      "train_tokens_per_second": 8495.282
    },
    {
      "epoch": 0.042418529490977644,
      "grad_norm": 1.4673023702926393,
      "learning_rate": 9.988940105120483e-06,
      "loss": 0.4924504280090332,
      "num_input_tokens_seen": 10313288,
      "step": 630,
      "train_runtime": 1213.9533,
      "train_tokens_per_second": 8495.622
    },
    {
      "epoch": 0.04275518448693778,
      "grad_norm": 1.5334844857623462,
      "learning_rate": 9.988763639422965e-06,
      "loss": 0.5415166854858399,
      "num_input_tokens_seen": 10395208,
      "step": 635,
      "train_runtime": 1223.5697,
      "train_tokens_per_second": 8495.804
    },
    {
      "epoch": 0.04309183948289792,
      "grad_norm": 1.4659920199065402,
      "learning_rate": 9.988585778633604e-06,
      "loss": 0.46565613746643064,
      "num_input_tokens_seen": 10477128,
      "step": 640,
      "train_runtime": 1233.1207,
      "train_tokens_per_second": 8496.433
    },
    {
      "epoch": 0.04342849447885807,
      "grad_norm": 1.4326968697885085,
      "learning_rate": 9.988406522802134e-06,
      "loss": 0.44670858383178713,
      "num_input_tokens_seen": 10558920,
      "step": 645,
      "train_runtime": 1243.1359,
      "train_tokens_per_second": 8493.778
    },
    {
      "epoch": 0.04376514947481821,
      "grad_norm": 1.3713229000796185,
      "learning_rate": 9.98822587197869e-06,
      "loss": 0.44916224479675293,
      "num_input_tokens_seen": 10640840,
      "step": 650,
      "train_runtime": 1252.7089,
      "train_tokens_per_second": 8494.264
    },
    {
      "epoch": 0.04410180447077835,
      "grad_norm": 3.9853572669978834,
      "learning_rate": 9.988043826213785e-06,
      "loss": 0.47483487129211427,
      "num_input_tokens_seen": 10722760,
      "step": 655,
      "train_runtime": 1262.334,
      "train_tokens_per_second": 8494.392
    },
    {
      "epoch": 0.04443845946673849,
      "grad_norm": 1.2473115480988812,
      "learning_rate": 9.98786038555833e-06,
      "loss": 0.4335452079772949,
      "num_input_tokens_seen": 10804680,
      "step": 660,
      "train_runtime": 1271.8355,
      "train_tokens_per_second": 8495.344
    },
    {
      "epoch": 0.04477511446269863,
      "grad_norm": 1.2708390785402737,
      "learning_rate": 9.987675550063625e-06,
      "loss": 0.42711191177368163,
      "num_input_tokens_seen": 10886600,
      "step": 665,
      "train_runtime": 1281.455,
      "train_tokens_per_second": 8495.499
    },
    {
      "epoch": 0.04511176945865877,
      "grad_norm": 1.7533068171508794,
      "learning_rate": 9.987489319781355e-06,
      "loss": 0.49003019332885744,
      "num_input_tokens_seen": 10968520,
      "step": 670,
      "train_runtime": 1291.0938,
      "train_tokens_per_second": 8495.525
    },
    {
      "epoch": 0.045448424454618906,
      "grad_norm": 1.686836839644352,
      "learning_rate": 9.987301694763602e-06,
      "loss": 0.5393509864807129,
      "num_input_tokens_seen": 11050440,
      "step": 675,
      "train_runtime": 1300.7156,
      "train_tokens_per_second": 8495.662
    },
    {
      "epoch": 0.045785079450579046,
      "grad_norm": 1.5840654960135818,
      "learning_rate": 9.987112675062834e-06,
      "loss": 0.43062119483947753,
      "num_input_tokens_seen": 11132360,
      "step": 680,
      "train_runtime": 1310.2244,
      "train_tokens_per_second": 8496.53
    },
    {
      "epoch": 0.046121734446539185,
      "grad_norm": 1.4751550675811271,
      "learning_rate": 9.986922260731908e-06,
      "loss": 0.4726102352142334,
      "num_input_tokens_seen": 11214280,
      "step": 685,
      "train_runtime": 1319.8391,
      "train_tokens_per_second": 8496.702
    },
    {
      "epoch": 0.046458389442499325,
      "grad_norm": 1.3510487733296197,
      "learning_rate": 9.986730451824073e-06,
      "loss": 0.49521818161010744,
      "num_input_tokens_seen": 11295880,
      "step": 690,
      "train_runtime": 1329.4594,
      "train_tokens_per_second": 8496.597
    },
    {
      "epoch": 0.046795044438459464,
      "grad_norm": 1.5611190798569319,
      "learning_rate": 9.986537248392969e-06,
      "loss": 0.49445400238037107,
      "num_input_tokens_seen": 11377800,
      "step": 695,
      "train_runtime": 1339.0078,
      "train_tokens_per_second": 8497.187
    },
    {
      "epoch": 0.047131699434419604,
      "grad_norm": 1.7225307478840042,
      "learning_rate": 9.986342650492626e-06,
      "loss": 0.471039867401123,
      "num_input_tokens_seen": 11459720,
      "step": 700,
      "train_runtime": 1348.5781,
      "train_tokens_per_second": 8497.632
    },
    {
      "epoch": 0.04746835443037975,
      "grad_norm": 1.4492319751887295,
      "learning_rate": 9.98614665817746e-06,
      "loss": 0.4740910053253174,
      "num_input_tokens_seen": 11541640,
      "step": 705,
      "train_runtime": 1358.1913,
      "train_tokens_per_second": 8497.802
    },
    {
      "epoch": 0.04780500942633989,
      "grad_norm": 1.5280953580919032,
      "learning_rate": 9.98594927150228e-06,
      "loss": 0.5026487350463867,
      "num_input_tokens_seen": 11623560,
      "step": 710,
      "train_runtime": 1367.7979,
      "train_tokens_per_second": 8498.01
    },
    {
      "epoch": 0.04814166442230003,
      "grad_norm": 1.336188131043133,
      "learning_rate": 9.985750490522286e-06,
      "loss": 0.4420804500579834,
      "num_input_tokens_seen": 11705480,
      "step": 715,
      "train_runtime": 1377.353,
      "train_tokens_per_second": 8498.533
    },
    {
      "epoch": 0.04847831941826017,
      "grad_norm": 1.481359561373116,
      "learning_rate": 9.985550315293066e-06,
      "loss": 0.4678789138793945,
      "num_input_tokens_seen": 11787400,
      "step": 720,
      "train_runtime": 1386.9123,
      "train_tokens_per_second": 8499.023
    },
    {
      "epoch": 0.04881497441422031,
      "grad_norm": 1.660388097595603,
      "learning_rate": 9.985348745870598e-06,
      "loss": 0.4955454349517822,
      "num_input_tokens_seen": 11869320,
      "step": 725,
      "train_runtime": 1396.4788,
      "train_tokens_per_second": 8499.463
    },
    {
      "epoch": 0.04915162941018045,
      "grad_norm": 1.3386213465874939,
      "learning_rate": 9.98514578231125e-06,
      "loss": 0.41736741065979005,
      "num_input_tokens_seen": 11951240,
      "step": 730,
      "train_runtime": 1406.0972,
      "train_tokens_per_second": 8499.583
    },
    {
      "epoch": 0.04948828440614059,
      "grad_norm": 1.2988685321662323,
      "learning_rate": 9.984941424671781e-06,
      "loss": 0.4819840431213379,
      "num_input_tokens_seen": 12033160,
      "step": 735,
      "train_runtime": 1415.7238,
      "train_tokens_per_second": 8499.652
    },
    {
      "epoch": 0.049824939402100726,
      "grad_norm": 1.343188658018391,
      "learning_rate": 9.984735673009337e-06,
      "loss": 0.46618499755859377,
      "num_input_tokens_seen": 12115080,
      "step": 740,
      "train_runtime": 1425.2314,
      "train_tokens_per_second": 8500.43
    },
    {
      "epoch": 0.050161594398060866,
      "grad_norm": 1.308282664548342,
      "learning_rate": 9.984528527381458e-06,
      "loss": 0.4717452049255371,
      "num_input_tokens_seen": 12197000,
      "step": 745,
      "train_runtime": 1434.8597,
      "train_tokens_per_second": 8500.483
    },
    {
      "epoch": 0.050498249394021005,
      "grad_norm": 1.486730547944254,
      "learning_rate": 9.984319987846072e-06,
      "loss": 0.498822021484375,
      "num_input_tokens_seen": 12278920,
      "step": 750,
      "train_runtime": 1444.4644,
      "train_tokens_per_second": 8500.674
    },
    {
      "epoch": 0.050834904389981145,
      "grad_norm": 1.4665710740257287,
      "learning_rate": 9.984110054461495e-06,
      "loss": 0.4719644546508789,
      "num_input_tokens_seen": 12360840,
      "step": 755,
      "train_runtime": 1454.0829,
      "train_tokens_per_second": 8500.781
    },
    {
      "epoch": 0.051171559385941284,
      "grad_norm": 1.3015546771879685,
      "learning_rate": 9.983898727286435e-06,
      "loss": 0.47872190475463866,
      "num_input_tokens_seen": 12442760,
      "step": 760,
      "train_runtime": 1463.5896,
      "train_tokens_per_second": 8501.536
    },
    {
      "epoch": 0.05150821438190143,
      "grad_norm": 1.3138646184478815,
      "learning_rate": 9.983686006379985e-06,
      "loss": 0.45392751693725586,
      "num_input_tokens_seen": 12524680,
      "step": 765,
      "train_runtime": 1473.2005,
      "train_tokens_per_second": 8501.681
    },
    {
      "epoch": 0.05184486937786157,
      "grad_norm": 1.3601159260937807,
      "learning_rate": 9.983471891801638e-06,
      "loss": 0.44568710327148436,
      "num_input_tokens_seen": 12606600,
      "step": 770,
      "train_runtime": 1482.8222,
      "train_tokens_per_second": 8501.761
    },
    {
      "epoch": 0.05218152437382171,
      "grad_norm": 1.3956598930168813,
      "learning_rate": 9.983256383611267e-06,
      "loss": 0.4925371170043945,
      "num_input_tokens_seen": 12688376,
      "step": 775,
      "train_runtime": 1492.8543,
      "train_tokens_per_second": 8499.407
    },
    {
      "epoch": 0.05251817936978185,
      "grad_norm": 1.48612874988105,
      "learning_rate": 9.983039481869138e-06,
      "loss": 0.48917016983032224,
      "num_input_tokens_seen": 12770296,
      "step": 780,
      "train_runtime": 1502.4138,
      "train_tokens_per_second": 8499.853
    },
    {
      "epoch": 0.05285483436574199,
      "grad_norm": 1.421832580822415,
      "learning_rate": 9.98282118663591e-06,
      "loss": 0.4675449371337891,
      "num_input_tokens_seen": 12852216,
      "step": 785,
      "train_runtime": 1512.0403,
      "train_tokens_per_second": 8499.916
    },
    {
      "epoch": 0.05319148936170213,
      "grad_norm": 1.4176312638471198,
      "learning_rate": 9.982601497972624e-06,
      "loss": 0.4409438133239746,
      "num_input_tokens_seen": 12934136,
      "step": 790,
      "train_runtime": 1521.6706,
      "train_tokens_per_second": 8499.958
    },
    {
      "epoch": 0.05352814435766227,
      "grad_norm": 1.4146611229310089,
      "learning_rate": 9.982380415940717e-06,
      "loss": 0.43826751708984374,
      "num_input_tokens_seen": 13016056,
      "step": 795,
      "train_runtime": 1531.3096,
      "train_tokens_per_second": 8499.951
    },
    {
      "epoch": 0.05386479935362241,
      "grad_norm": 1.325556951354761,
      "learning_rate": 9.982157940602015e-06,
      "loss": 0.5199309349060058,
      "num_input_tokens_seen": 13097976,
      "step": 800,
      "train_runtime": 1541.0496,
      "train_tokens_per_second": 8499.386
    },
    {
      "epoch": 0.05420145434958255,
      "grad_norm": 1.5674701057301266,
      "learning_rate": 9.981934072018732e-06,
      "loss": 0.5106181621551513,
      "num_input_tokens_seen": 13179896,
      "step": 805,
      "train_runtime": 1550.7095,
      "train_tokens_per_second": 8499.268
    },
    {
      "epoch": 0.054538109345542686,
      "grad_norm": 1.3679811719098505,
      "learning_rate": 9.981708810253472e-06,
      "loss": 0.501543664932251,
      "num_input_tokens_seen": 13261816,
      "step": 810,
      "train_runtime": 1560.3264,
      "train_tokens_per_second": 8499.386
    },
    {
      "epoch": 0.054874764341502826,
      "grad_norm": 1.4713290620773554,
      "learning_rate": 9.981482155369229e-06,
      "loss": 0.5054832935333252,
      "num_input_tokens_seen": 13343736,
      "step": 815,
      "train_runtime": 1569.9727,
      "train_tokens_per_second": 8499.343
    },
    {
      "epoch": 0.055211419337462965,
      "grad_norm": 1.8562483978200643,
      "learning_rate": 9.981254107429385e-06,
      "loss": 0.4534252166748047,
      "num_input_tokens_seen": 13425656,
      "step": 820,
      "train_runtime": 1579.6378,
      "train_tokens_per_second": 8499.199
    },
    {
      "epoch": 0.05554807433342311,
      "grad_norm": 1.34054104701682,
      "learning_rate": 9.981024666497716e-06,
      "loss": 0.47512040138244627,
      "num_input_tokens_seen": 13507576,
      "step": 825,
      "train_runtime": 1589.1974,
      "train_tokens_per_second": 8499.621
    },
    {
      "epoch": 0.05588472932938325,
      "grad_norm": 1.5519913083818826,
      "learning_rate": 9.98079383263838e-06,
      "loss": 0.47691545486450193,
      "num_input_tokens_seen": 13589112,
      "step": 830,
      "train_runtime": 1599.1728,
      "train_tokens_per_second": 8497.588
    },
    {
      "epoch": 0.05622138432534339,
      "grad_norm": 1.305083568129242,
      "learning_rate": 9.980561605915932e-06,
      "loss": 0.48986005783081055,
      "num_input_tokens_seen": 13671032,
      "step": 835,
      "train_runtime": 1608.8761,
      "train_tokens_per_second": 8497.256
    },
    {
      "epoch": 0.05655803932130353,
      "grad_norm": 1.6221756885349827,
      "learning_rate": 9.980327986395316e-06,
      "loss": 0.5014022350311279,
      "num_input_tokens_seen": 13752528,
      "step": 840,
      "train_runtime": 1618.8797,
      "train_tokens_per_second": 8495.089
    },
    {
      "epoch": 0.05689469431726367,
      "grad_norm": 1.461668616813517,
      "learning_rate": 9.980092974141855e-06,
      "loss": 0.44611349105834963,
      "num_input_tokens_seen": 13834448,
      "step": 845,
      "train_runtime": 1628.5004,
      "train_tokens_per_second": 8495.207
    },
    {
      "epoch": 0.05723134931322381,
      "grad_norm": 1.5406594804144744,
      "learning_rate": 9.979856569221276e-06,
      "loss": 0.48315958976745604,
      "num_input_tokens_seen": 13916368,
      "step": 850,
      "train_runtime": 1638.1726,
      "train_tokens_per_second": 8495.056
    },
    {
      "epoch": 0.05756800430918395,
      "grad_norm": 1.316745701991176,
      "learning_rate": 9.979618771699688e-06,
      "loss": 0.4663818359375,
      "num_input_tokens_seen": 13998288,
      "step": 855,
      "train_runtime": 1647.7971,
      "train_tokens_per_second": 8495.153
    },
    {
      "epoch": 0.05790465930514409,
      "grad_norm": 1.4568046792800382,
      "learning_rate": 9.979379581643588e-06,
      "loss": 0.4606689453125,
      "num_input_tokens_seen": 14080208,
      "step": 860,
      "train_runtime": 1657.4437,
      "train_tokens_per_second": 8495.135
    },
    {
      "epoch": 0.05824131430110423,
      "grad_norm": 1.380830051664812,
      "learning_rate": 9.979138999119866e-06,
      "loss": 0.5057209968566895,
      "num_input_tokens_seen": 14162128,
      "step": 865,
      "train_runtime": 1667.2561,
      "train_tokens_per_second": 8494.273
    },
    {
      "epoch": 0.05857796929706437,
      "grad_norm": 1.2465375561124492,
      "learning_rate": 9.978897024195801e-06,
      "loss": 0.48077707290649413,
      "num_input_tokens_seen": 14244048,
      "step": 870,
      "train_runtime": 1676.7608,
      "train_tokens_per_second": 8494.979
    },
    {
      "epoch": 0.058914624293024506,
      "grad_norm": 1.3655631416316563,
      "learning_rate": 9.978653656939058e-06,
      "loss": 0.46390571594238283,
      "num_input_tokens_seen": 14325968,
      "step": 875,
      "train_runtime": 1686.3407,
      "train_tokens_per_second": 8495.299
    },
    {
      "epoch": 0.059251279288984646,
      "grad_norm": 1.8723735288829817,
      "learning_rate": 9.978408897417699e-06,
      "loss": 0.49671382904052735,
      "num_input_tokens_seen": 14407888,
      "step": 880,
      "train_runtime": 1695.9411,
      "train_tokens_per_second": 8495.512
    },
    {
      "epoch": 0.059587934284944785,
      "grad_norm": 1.1870634278579553,
      "learning_rate": 9.978162745700165e-06,
      "loss": 0.49252119064331057,
      "num_input_tokens_seen": 14489808,
      "step": 885,
      "train_runtime": 1705.5395,
      "train_tokens_per_second": 8495.733
    },
    {
      "epoch": 0.05992458928090493,
      "grad_norm": 1.8929724532187515,
      "learning_rate": 9.977915201855293e-06,
      "loss": 0.4274531364440918,
      "num_input_tokens_seen": 14571728,
      "step": 890,
      "train_runtime": 1715.1417,
      "train_tokens_per_second": 8495.932
    },
    {
      "epoch": 0.06026124427686507,
      "grad_norm": 1.1194657847047664,
      "learning_rate": 9.977666265952309e-06,
      "loss": 0.4456298828125,
      "num_input_tokens_seen": 14653648,
      "step": 895,
      "train_runtime": 1724.7037,
      "train_tokens_per_second": 8496.328
    },
    {
      "epoch": 0.06059789927282521,
      "grad_norm": 1.400574053424781,
      "learning_rate": 9.977415938060823e-06,
      "loss": 0.4879148483276367,
      "num_input_tokens_seen": 14735568,
      "step": 900,
      "train_runtime": 1734.2681,
      "train_tokens_per_second": 8496.707
    },
    {
      "epoch": 0.06093455426878535,
      "grad_norm": 1.4813741535476959,
      "learning_rate": 9.977164218250846e-06,
      "loss": 0.44144539833068847,
      "num_input_tokens_seen": 14817488,
      "step": 905,
      "train_runtime": 1743.8791,
      "train_tokens_per_second": 8496.855
    },
    {
      "epoch": 0.06127120926474549,
      "grad_norm": 1.41646029586394,
      "learning_rate": 9.976911106592765e-06,
      "loss": 0.45835204124450685,
      "num_input_tokens_seen": 14899408,
      "step": 910,
      "train_runtime": 1753.5102,
      "train_tokens_per_second": 8496.904
    },
    {
      "epoch": 0.06160786426070563,
      "grad_norm": 1.4176940271267793,
      "learning_rate": 9.976656603157363e-06,
      "loss": 0.47731742858886717,
      "num_input_tokens_seen": 14981328,
      "step": 915,
      "train_runtime": 1763.1448,
      "train_tokens_per_second": 8496.936
    },
    {
      "epoch": 0.06194451925666577,
      "grad_norm": 1.5738173654509675,
      "learning_rate": 9.976400708015811e-06,
      "loss": 0.48295793533325193,
      "num_input_tokens_seen": 15063248,
      "step": 920,
      "train_runtime": 1772.7546,
      "train_tokens_per_second": 8497.086
    },
    {
      "epoch": 0.06228117425262591,
      "grad_norm": 1.4255018924681924,
      "learning_rate": 9.97614342123967e-06,
      "loss": 0.489838171005249,
      "num_input_tokens_seen": 15145168,
      "step": 925,
      "train_runtime": 1782.2642,
      "train_tokens_per_second": 8497.712
    },
    {
      "epoch": 0.06261782924858605,
      "grad_norm": 1.976978463555357,
      "learning_rate": 9.97588474290089e-06,
      "loss": 0.44320383071899416,
      "num_input_tokens_seen": 15227088,
      "step": 930,
      "train_runtime": 1791.7687,
      "train_tokens_per_second": 8498.356
    },
    {
      "epoch": 0.0629544842445462,
      "grad_norm": 1.448457650053793,
      "learning_rate": 9.975624673071806e-06,
      "loss": 0.40511503219604494,
      "num_input_tokens_seen": 15309008,
      "step": 935,
      "train_runtime": 1801.3718,
      "train_tokens_per_second": 8498.528
    },
    {
      "epoch": 0.06329113924050633,
      "grad_norm": 1.6031249422528817,
      "learning_rate": 9.97536321182515e-06,
      "loss": 0.4405013084411621,
      "num_input_tokens_seen": 15390472,
      "step": 940,
      "train_runtime": 1811.3221,
      "train_tokens_per_second": 8496.817
    },
    {
      "epoch": 0.06362779423646647,
      "grad_norm": 1.3058752513422678,
      "learning_rate": 9.975100359234037e-06,
      "loss": 0.4939435958862305,
      "num_input_tokens_seen": 15472392,
      "step": 945,
      "train_runtime": 1820.8685,
      "train_tokens_per_second": 8497.259
    },
    {
      "epoch": 0.06396444923242661,
      "grad_norm": 1.3129061251023932,
      "learning_rate": 9.974836115371972e-06,
      "loss": 0.48113350868225097,
      "num_input_tokens_seen": 15554312,
      "step": 950,
      "train_runtime": 1830.4318,
      "train_tokens_per_second": 8497.619
    },
    {
      "epoch": 0.06430110422838675,
      "grad_norm": 1.2807357575791518,
      "learning_rate": 9.974570480312852e-06,
      "loss": 0.42735538482666013,
      "num_input_tokens_seen": 15636232,
      "step": 955,
      "train_runtime": 1840.0624,
      "train_tokens_per_second": 8497.664
    },
    {
      "epoch": 0.06463775922434689,
      "grad_norm": 1.2766022129851744,
      "learning_rate": 9.974303454130962e-06,
      "loss": 0.4824409484863281,
      "num_input_tokens_seen": 15718152,
      "step": 960,
      "train_runtime": 1849.6701,
      "train_tokens_per_second": 8497.814
    },
    {
      "epoch": 0.06497441422030703,
      "grad_norm": 1.3867485303648484,
      "learning_rate": 9.97403503690097e-06,
      "loss": 0.4429784774780273,
      "num_input_tokens_seen": 15800072,
      "step": 965,
      "train_runtime": 1859.2234,
      "train_tokens_per_second": 8498.21
    },
    {
      "epoch": 0.06531106921626717,
      "grad_norm": 1.2643764301919538,
      "learning_rate": 9.973765228697942e-06,
      "loss": 0.42247982025146485,
      "num_input_tokens_seen": 15881992,
      "step": 970,
      "train_runtime": 1868.789,
      "train_tokens_per_second": 8498.547
    },
    {
      "epoch": 0.06564772421222731,
      "grad_norm": 1.4799341990851371,
      "learning_rate": 9.973494029597327e-06,
      "loss": 0.4619112014770508,
      "num_input_tokens_seen": 15963248,
      "step": 975,
      "train_runtime": 1878.6318,
      "train_tokens_per_second": 8497.274
    },
    {
      "epoch": 0.06598437920818745,
      "grad_norm": 1.3987710984498254,
      "learning_rate": 9.973221439674968e-06,
      "loss": 0.424973201751709,
      "num_input_tokens_seen": 16045168,
      "step": 980,
      "train_runtime": 1888.2425,
      "train_tokens_per_second": 8497.409
    },
    {
      "epoch": 0.06632103420414759,
      "grad_norm": 1.2538981822958506,
      "learning_rate": 9.972947459007089e-06,
      "loss": 0.4931222915649414,
      "num_input_tokens_seen": 16127088,
      "step": 985,
      "train_runtime": 1897.7914,
      "train_tokens_per_second": 8497.819
    },
    {
      "epoch": 0.06665768920010773,
      "grad_norm": 1.391542105421352,
      "learning_rate": 9.972672087670313e-06,
      "loss": 0.47448458671569826,
      "num_input_tokens_seen": 16209008,
      "step": 990,
      "train_runtime": 1907.3469,
      "train_tokens_per_second": 8498.196
    },
    {
      "epoch": 0.06699434419606787,
      "grad_norm": 1.3632636578629882,
      "learning_rate": 9.972395325741644e-06,
      "loss": 0.40822691917419435,
      "num_input_tokens_seen": 16290928,
      "step": 995,
      "train_runtime": 1916.972,
      "train_tokens_per_second": 8498.261
    },
    {
      "epoch": 0.06733099919202801,
      "grad_norm": 1.3023402784341078,
      "learning_rate": 9.972117173298476e-06,
      "loss": 0.4392500400543213,
      "num_input_tokens_seen": 16372848,
      "step": 1000,
      "train_runtime": 1926.5749,
      "train_tokens_per_second": 8498.423
    },
    {
      "epoch": 0.06766765418798815,
      "grad_norm": 1.3001202763237532,
      "learning_rate": 9.971837630418597e-06,
      "loss": 0.4953733444213867,
      "num_input_tokens_seen": 16454768,
      "step": 1005,
      "train_runtime": 1936.624,
      "train_tokens_per_second": 8496.625
    },
    {
      "epoch": 0.06800430918394829,
      "grad_norm": 1.3643817151995299,
      "learning_rate": 9.971556697180179e-06,
      "loss": 0.48952360153198243,
      "num_input_tokens_seen": 16536688,
      "step": 1010,
      "train_runtime": 1946.3411,
      "train_tokens_per_second": 8496.295
    },
    {
      "epoch": 0.06834096417990843,
      "grad_norm": 1.4197670313596151,
      "learning_rate": 9.971274373661784e-06,
      "loss": 0.4681891918182373,
      "num_input_tokens_seen": 16618608,
      "step": 1015,
      "train_runtime": 1955.9954,
      "train_tokens_per_second": 8496.241
    },
    {
      "epoch": 0.06867761917586857,
      "grad_norm": 1.2435294945659279,
      "learning_rate": 9.97099065994236e-06,
      "loss": 0.4071362972259521,
      "num_input_tokens_seen": 16700528,
      "step": 1020,
      "train_runtime": 1965.6431,
      "train_tokens_per_second": 8496.216
    },
    {
      "epoch": 0.0690142741718287,
      "grad_norm": 1.3929208642157378,
      "learning_rate": 9.970705556101253e-06,
      "loss": 0.4581589221954346,
      "num_input_tokens_seen": 16781480,
      "step": 1025,
      "train_runtime": 1975.3233,
      "train_tokens_per_second": 8495.561
    },
    {
      "epoch": 0.06935092916778884,
      "grad_norm": 1.3818944535665907,
      "learning_rate": 9.970419062218186e-06,
      "loss": 0.4729026794433594,
      "num_input_tokens_seen": 16863400,
      "step": 1030,
      "train_runtime": 1984.9491,
      "train_tokens_per_second": 8495.634
    },
    {
      "epoch": 0.06968758416374898,
      "grad_norm": 1.560208184320684,
      "learning_rate": 9.970131178373276e-06,
      "loss": 0.5047868728637696,
      "num_input_tokens_seen": 16945320,
      "step": 1035,
      "train_runtime": 1994.5563,
      "train_tokens_per_second": 8495.784
    },
    {
      "epoch": 0.07002423915970914,
      "grad_norm": 1.4769629236986894,
      "learning_rate": 9.969841904647033e-06,
      "loss": 0.5048693656921387,
      "num_input_tokens_seen": 17027240,
      "step": 1040,
      "train_runtime": 2004.0696,
      "train_tokens_per_second": 8496.332
    },
    {
      "epoch": 0.07036089415566928,
      "grad_norm": 1.2618316074808307,
      "learning_rate": 9.969551241120349e-06,
      "loss": 0.43936963081359864,
      "num_input_tokens_seen": 17109160,
      "step": 1045,
      "train_runtime": 2013.6254,
      "train_tokens_per_second": 8496.694
    },
    {
      "epoch": 0.07069754915162942,
      "grad_norm": 1.256554468352807,
      "learning_rate": 9.969259187874507e-06,
      "loss": 0.46021265983581544,
      "num_input_tokens_seen": 17191080,
      "step": 1050,
      "train_runtime": 2023.2397,
      "train_tokens_per_second": 8496.808
    },
    {
      "epoch": 0.07103420414758956,
      "grad_norm": 1.4863119436514498,
      "learning_rate": 9.968965744991178e-06,
      "loss": 0.49019269943237304,
      "num_input_tokens_seen": 17273000,
      "step": 1055,
      "train_runtime": 2032.8085,
      "train_tokens_per_second": 8497.111
    },
    {
      "epoch": 0.0713708591435497,
      "grad_norm": 1.3480572746740533,
      "learning_rate": 9.968670912552422e-06,
      "loss": 0.4333824157714844,
      "num_input_tokens_seen": 17354920,
      "step": 1060,
      "train_runtime": 2042.414,
      "train_tokens_per_second": 8497.259
    },
    {
      "epoch": 0.07170751413950983,
      "grad_norm": 1.1491650301270417,
      "learning_rate": 9.968374690640692e-06,
      "loss": 0.45323843955993653,
      "num_input_tokens_seen": 17436840,
      "step": 1065,
      "train_runtime": 2052.0282,
      "train_tokens_per_second": 8497.368
    },
    {
      "epoch": 0.07204416913546997,
      "grad_norm": 1.5648521827476547,
      "learning_rate": 9.968077079338822e-06,
      "loss": 0.47956228256225586,
      "num_input_tokens_seen": 17518760,
      "step": 1070,
      "train_runtime": 2061.5284,
      "train_tokens_per_second": 8497.947
    },
    {
      "epoch": 0.07238082413143011,
      "grad_norm": 1.409246007043887,
      "learning_rate": 9.967778078730038e-06,
      "loss": 0.45871829986572266,
      "num_input_tokens_seen": 17600680,
      "step": 1075,
      "train_runtime": 2071.1178,
      "train_tokens_per_second": 8498.155
    },
    {
      "epoch": 0.07271747912739025,
      "grad_norm": 1.4372398110421984,
      "learning_rate": 9.967477688897957e-06,
      "loss": 0.5181427955627441,
      "num_input_tokens_seen": 17682600,
      "step": 1080,
      "train_runtime": 2080.6342,
      "train_tokens_per_second": 8498.659
    },
    {
      "epoch": 0.07305413412335039,
      "grad_norm": 1.193022031452731,
      "learning_rate": 9.967175909926578e-06,
      "loss": 0.5251864910125732,
      "num_input_tokens_seen": 17764520,
      "step": 1085,
      "train_runtime": 2090.2502,
      "train_tokens_per_second": 8498.753
    },
    {
      "epoch": 0.07339078911931053,
      "grad_norm": 1.5193050026803752,
      "learning_rate": 9.966872741900296e-06,
      "loss": 0.5133438110351562,
      "num_input_tokens_seen": 17846440,
      "step": 1090,
      "train_runtime": 2099.7514,
      "train_tokens_per_second": 8499.311
    },
    {
      "epoch": 0.07372744411527067,
      "grad_norm": 1.4893817767402653,
      "learning_rate": 9.96656818490389e-06,
      "loss": 0.4766097545623779,
      "num_input_tokens_seen": 17928360,
      "step": 1095,
      "train_runtime": 2109.3602,
      "train_tokens_per_second": 8499.43
    },
    {
      "epoch": 0.07406409911123081,
      "grad_norm": 1.4986034962668435,
      "learning_rate": 9.966262239022529e-06,
      "loss": 0.5057106971740722,
      "num_input_tokens_seen": 18010280,
      "step": 1100,
      "train_runtime": 2118.9646,
      "train_tokens_per_second": 8499.566
    },
    {
      "epoch": 0.07440075410719095,
      "grad_norm": 1.395248094860535,
      "learning_rate": 9.965954904341768e-06,
      "loss": 0.5058935165405274,
      "num_input_tokens_seen": 18092200,
      "step": 1105,
      "train_runtime": 2128.5941,
      "train_tokens_per_second": 8499.601
    },
    {
      "epoch": 0.07473740910315109,
      "grad_norm": 1.391383317145339,
      "learning_rate": 9.965646180947555e-06,
      "loss": 0.4792300224304199,
      "num_input_tokens_seen": 18174120,
      "step": 1110,
      "train_runtime": 2138.0946,
      "train_tokens_per_second": 8500.148
    },
    {
      "epoch": 0.07507406409911123,
      "grad_norm": 1.2263903321037921,
      "learning_rate": 9.96533606892622e-06,
      "loss": 0.4532012939453125,
      "num_input_tokens_seen": 18256040,
      "step": 1115,
      "train_runtime": 2147.7145,
      "train_tokens_per_second": 8500.217
    },
    {
      "epoch": 0.07541071909507137,
      "grad_norm": 1.2327948122731545,
      "learning_rate": 9.965024568364487e-06,
      "loss": 0.46173825263977053,
      "num_input_tokens_seen": 18337960,
      "step": 1120,
      "train_runtime": 2157.3252,
      "train_tokens_per_second": 8500.322
    },
    {
      "epoch": 0.07574737409103151,
      "grad_norm": 1.5296474343240793,
      "learning_rate": 9.964711679349467e-06,
      "loss": 0.45076284408569334,
      "num_input_tokens_seen": 18419880,
      "step": 1125,
      "train_runtime": 2166.9536,
      "train_tokens_per_second": 8500.358
    },
    {
      "epoch": 0.07608402908699165,
      "grad_norm": 1.2854049779375292,
      "learning_rate": 9.964397401968656e-06,
      "loss": 0.5333981037139892,
      "num_input_tokens_seen": 18501800,
      "step": 1130,
      "train_runtime": 2176.5382,
      "train_tokens_per_second": 8500.563
    },
    {
      "epoch": 0.07642068408295179,
      "grad_norm": 1.4087243961843778,
      "learning_rate": 9.964081736309942e-06,
      "loss": 0.4624027252197266,
      "num_input_tokens_seen": 18583720,
      "step": 1135,
      "train_runtime": 2186.0841,
      "train_tokens_per_second": 8500.917
    },
    {
      "epoch": 0.07675733907891193,
      "grad_norm": 1.2513452869665356,
      "learning_rate": 9.963764682461599e-06,
      "loss": 0.4868816375732422,
      "num_input_tokens_seen": 18665640,
      "step": 1140,
      "train_runtime": 2195.661,
      "train_tokens_per_second": 8501.148
    },
    {
      "epoch": 0.07709399407487207,
      "grad_norm": 1.1312319495641716,
      "learning_rate": 9.96344624051229e-06,
      "loss": 0.4741383075714111,
      "num_input_tokens_seen": 18747560,
      "step": 1145,
      "train_runtime": 2205.2744,
      "train_tokens_per_second": 8501.237
    },
    {
      "epoch": 0.0774306490708322,
      "grad_norm": 1.4407710847742023,
      "learning_rate": 9.96312641055107e-06,
      "loss": 0.4379236698150635,
      "num_input_tokens_seen": 18829480,
      "step": 1150,
      "train_runtime": 2215.1931,
      "train_tokens_per_second": 8500.153
    },
    {
      "epoch": 0.07776730406679234,
      "grad_norm": 1.193630916024369,
      "learning_rate": 9.962805192667375e-06,
      "loss": 0.4506418704986572,
      "num_input_tokens_seen": 18911400,
      "step": 1155,
      "train_runtime": 2225.2317,
      "train_tokens_per_second": 8498.621
    },
    {
      "epoch": 0.07810395906275248,
      "grad_norm": 1.383181156735051,
      "learning_rate": 9.962482586951033e-06,
      "loss": 0.464204216003418,
      "num_input_tokens_seen": 18993320,
      "step": 1160,
      "train_runtime": 2235.2173,
      "train_tokens_per_second": 8497.304
    },
    {
      "epoch": 0.07844061405871264,
      "grad_norm": 1.3028872133216942,
      "learning_rate": 9.962158593492258e-06,
      "loss": 0.4624335289001465,
      "num_input_tokens_seen": 19075240,
      "step": 1165,
      "train_runtime": 2245.0672,
      "train_tokens_per_second": 8496.512
    },
    {
      "epoch": 0.07877726905467278,
      "grad_norm": 1.3484685080375343,
      "learning_rate": 9.961833212381657e-06,
      "loss": 0.4532326698303223,
      "num_input_tokens_seen": 19157160,
      "step": 1170,
      "train_runtime": 2254.9113,
      "train_tokens_per_second": 8495.749
    },
    {
      "epoch": 0.07911392405063292,
      "grad_norm": 1.2127197497236062,
      "learning_rate": 9.961506443710221e-06,
      "loss": 0.4584930419921875,
      "num_input_tokens_seen": 19239080,
      "step": 1175,
      "train_runtime": 2264.4208,
      "train_tokens_per_second": 8496.248
    },
    {
      "epoch": 0.07945057904659306,
      "grad_norm": 1.4687168336709204,
      "learning_rate": 9.961178287569326e-06,
      "loss": 0.42391386032104494,
      "num_input_tokens_seen": 19321000,
      "step": 1180,
      "train_runtime": 2273.9358,
      "train_tokens_per_second": 8496.722
    },
    {
      "epoch": 0.0797872340425532,
      "grad_norm": 1.3060240818013071,
      "learning_rate": 9.960848744050747e-06,
      "loss": 0.4523880958557129,
      "num_input_tokens_seen": 19402920,
      "step": 1185,
      "train_runtime": 2283.433,
      "train_tokens_per_second": 8497.258
    },
    {
      "epoch": 0.08012388903851334,
      "grad_norm": 1.181397887279694,
      "learning_rate": 9.960517813246633e-06,
      "loss": 0.4877971649169922,
      "num_input_tokens_seen": 19484672,
      "step": 1190,
      "train_runtime": 2293.4271,
      "train_tokens_per_second": 8495.876
    },
    {
      "epoch": 0.08046054403447347,
      "grad_norm": 1.343980759946302,
      "learning_rate": 9.96018549524953e-06,
      "loss": 0.45887532234191897,
      "num_input_tokens_seen": 19566592,
      "step": 1195,
      "train_runtime": 2302.9289,
      "train_tokens_per_second": 8496.394
    },
    {
      "epoch": 0.08079719903043361,
      "grad_norm": 1.317791704905608,
      "learning_rate": 9.959851790152371e-06,
      "loss": 0.4539045810699463,
      "num_input_tokens_seen": 19648512,
      "step": 1200,
      "train_runtime": 2312.434,
      "train_tokens_per_second": 8496.896
    },
    {
      "epoch": 0.08113385402639375,
      "grad_norm": 1.2279023882664588,
      "learning_rate": 9.959516698048475e-06,
      "loss": 0.415986156463623,
      "num_input_tokens_seen": 19730432,
      "step": 1205,
      "train_runtime": 2321.9387,
      "train_tokens_per_second": 8497.396
    },
    {
      "epoch": 0.08147050902235389,
      "grad_norm": 1.1557093478899543,
      "learning_rate": 9.959180219031551e-06,
      "loss": 0.3991895914077759,
      "num_input_tokens_seen": 19812352,
      "step": 1210,
      "train_runtime": 2331.4418,
      "train_tokens_per_second": 8497.897
    },
    {
      "epoch": 0.08180716401831403,
      "grad_norm": 1.175271947664517,
      "learning_rate": 9.958842353195688e-06,
      "loss": 0.40545878410339353,
      "num_input_tokens_seen": 19894272,
      "step": 1215,
      "train_runtime": 2340.9493,
      "train_tokens_per_second": 8498.378
    },
    {
      "epoch": 0.08214381901427417,
      "grad_norm": 1.3514633436908332,
      "learning_rate": 9.958503100635377e-06,
      "loss": 0.44681634902954104,
      "num_input_tokens_seen": 19975728,
      "step": 1220,
      "train_runtime": 2350.7785,
      "train_tokens_per_second": 8497.495
    },
    {
      "epoch": 0.08248047401023431,
      "grad_norm": 1.3597481803263345,
      "learning_rate": 9.958162461445484e-06,
      "loss": 0.5013622283935547,
      "num_input_tokens_seen": 20057648,
      "step": 1225,
      "train_runtime": 2360.4051,
      "train_tokens_per_second": 8497.545
    },
    {
      "epoch": 0.08281712900619445,
      "grad_norm": 1.268090982270246,
      "learning_rate": 9.957820435721271e-06,
      "loss": 0.4621617317199707,
      "num_input_tokens_seen": 20139568,
      "step": 1230,
      "train_runtime": 2369.9075,
      "train_tokens_per_second": 8498.04
    },
    {
      "epoch": 0.08315378400215459,
      "grad_norm": 2.7646903764060005,
      "learning_rate": 9.95747702355838e-06,
      "loss": 0.49355716705322267,
      "num_input_tokens_seen": 20221488,
      "step": 1235,
      "train_runtime": 2379.4094,
      "train_tokens_per_second": 8498.532
    },
    {
      "epoch": 0.08349043899811473,
      "grad_norm": 1.6988234245013412,
      "learning_rate": 9.957132225052848e-06,
      "loss": 0.450040864944458,
      "num_input_tokens_seen": 20303408,
      "step": 1240,
      "train_runtime": 2388.9119,
      "train_tokens_per_second": 8499.019
    },
    {
      "epoch": 0.08382709399407487,
      "grad_norm": 1.2334083374303277,
      "learning_rate": 9.956786040301098e-06,
      "loss": 0.4560500144958496,
      "num_input_tokens_seen": 20385328,
      "step": 1245,
      "train_runtime": 2398.4087,
      "train_tokens_per_second": 8499.522
    },
    {
      "epoch": 0.08416374899003501,
      "grad_norm": 1.280029323236878,
      "learning_rate": 9.956438469399936e-06,
      "loss": 0.43802318572998045,
      "num_input_tokens_seen": 20467248,
      "step": 1250,
      "train_runtime": 2407.9103,
      "train_tokens_per_second": 8500.004
    },
    {
      "epoch": 0.08450040398599515,
      "grad_norm": 1.8301815743144363,
      "learning_rate": 9.956089512446562e-06,
      "loss": 0.4463841438293457,
      "num_input_tokens_seen": 20549168,
      "step": 1255,
      "train_runtime": 2417.4142,
      "train_tokens_per_second": 8500.475
    },
    {
      "epoch": 0.08483705898195529,
      "grad_norm": 1.4025445397240888,
      "learning_rate": 9.955739169538559e-06,
      "loss": 0.4787457942962646,
      "num_input_tokens_seen": 20631088,
      "step": 1260,
      "train_runtime": 2426.9108,
      "train_tokens_per_second": 8500.967
    },
    {
      "epoch": 0.08517371397791543,
      "grad_norm": 1.3020928274286314,
      "learning_rate": 9.955387440773902e-06,
      "loss": 0.44107303619384763,
      "num_input_tokens_seen": 20713008,
      "step": 1265,
      "train_runtime": 2436.5065,
      "train_tokens_per_second": 8501.109
    },
    {
      "epoch": 0.08551036897387557,
      "grad_norm": 1.5496630229902022,
      "learning_rate": 9.955034326250947e-06,
      "loss": 0.44536395072937013,
      "num_input_tokens_seen": 20794928,
      "step": 1270,
      "train_runtime": 2446.1925,
      "train_tokens_per_second": 8500.937
    },
    {
      "epoch": 0.0858470239698357,
      "grad_norm": 1.0437954069376774,
      "learning_rate": 9.954679826068444e-06,
      "loss": 0.4218449115753174,
      "num_input_tokens_seen": 20876624,
      "step": 1275,
      "train_runtime": 2456.3566,
      "train_tokens_per_second": 8499.02
    },
    {
      "epoch": 0.08618367896579585,
      "grad_norm": 1.3187301107672444,
      "learning_rate": 9.954323940325526e-06,
      "loss": 0.48881192207336427,
      "num_input_tokens_seen": 20958544,
      "step": 1280,
      "train_runtime": 2465.982,
      "train_tokens_per_second": 8499.066
    },
    {
      "epoch": 0.086520333961756,
      "grad_norm": 1.3108620318824755,
      "learning_rate": 9.953966669121715e-06,
      "loss": 0.43204050064086913,
      "num_input_tokens_seen": 21040464,
      "step": 1285,
      "train_runtime": 2475.5803,
      "train_tokens_per_second": 8499.205
    },
    {
      "epoch": 0.08685698895771614,
      "grad_norm": 1.4704914015446335,
      "learning_rate": 9.953608012556924e-06,
      "loss": 0.46093382835388186,
      "num_input_tokens_seen": 21122384,
      "step": 1290,
      "train_runtime": 2485.0842,
      "train_tokens_per_second": 8499.665
    },
    {
      "epoch": 0.08719364395367628,
      "grad_norm": 1.2175222199824807,
      "learning_rate": 9.953247970731449e-06,
      "loss": 0.4608863353729248,
      "num_input_tokens_seen": 21204304,
      "step": 1295,
      "train_runtime": 2494.7003,
      "train_tokens_per_second": 8499.74
    },
    {
      "epoch": 0.08753029894963642,
      "grad_norm": 1.2914007399096417,
      "learning_rate": 9.952886543745973e-06,
      "loss": 0.4783607482910156,
      "num_input_tokens_seen": 21286224,
      "step": 1300,
      "train_runtime": 2504.3147,
      "train_tokens_per_second": 8499.82
    },
    {
      "epoch": 0.08786695394559656,
      "grad_norm": 1.406383539270264,
      "learning_rate": 9.952523731701566e-06,
      "loss": 0.46276102066040037,
      "num_input_tokens_seen": 21368144,
      "step": 1305,
      "train_runtime": 2513.8134,
      "train_tokens_per_second": 8500.291
    },
    {
      "epoch": 0.0882036089415567,
      "grad_norm": 1.477303630459254,
      "learning_rate": 9.952159534699693e-06,
      "loss": 0.4576850891113281,
      "num_input_tokens_seen": 21449456,
      "step": 1310,
      "train_runtime": 2523.7247,
      "train_tokens_per_second": 8499.127
    },
    {
      "epoch": 0.08854026393751684,
      "grad_norm": 1.5142219316308023,
      "learning_rate": 9.951793952842197e-06,
      "loss": 0.4770622730255127,
      "num_input_tokens_seen": 21531376,
      "step": 1315,
      "train_runtime": 2533.3617,
      "train_tokens_per_second": 8499.132
    },
    {
      "epoch": 0.08887691893347698,
      "grad_norm": 1.2478899921633293,
      "learning_rate": 9.95142698623131e-06,
      "loss": 0.42185153961181643,
      "num_input_tokens_seen": 21613296,
      "step": 1320,
      "train_runtime": 2542.909,
      "train_tokens_per_second": 8499.437
    },
    {
      "epoch": 0.08921357392943711,
      "grad_norm": 1.3896071861340384,
      "learning_rate": 9.951058634969657e-06,
      "loss": 0.4937292098999023,
      "num_input_tokens_seen": 21694480,
      "step": 1325,
      "train_runtime": 2552.6954,
      "train_tokens_per_second": 8498.656
    },
    {
      "epoch": 0.08955022892539725,
      "grad_norm": 1.6283261030165295,
      "learning_rate": 9.950688899160244e-06,
      "loss": 0.46234331130981443,
      "num_input_tokens_seen": 21776400,
      "step": 1330,
      "train_runtime": 2562.3085,
      "train_tokens_per_second": 8498.742
    },
    {
      "epoch": 0.0898868839213574,
      "grad_norm": 1.7771225566983757,
      "learning_rate": 9.950317778906469e-06,
      "loss": 0.38193721771240235,
      "num_input_tokens_seen": 21858320,
      "step": 1335,
      "train_runtime": 2571.9183,
      "train_tokens_per_second": 8498.839
    },
    {
      "epoch": 0.09022353891731753,
      "grad_norm": 1.3782341807608627,
      "learning_rate": 9.949945274312109e-06,
      "loss": 0.4592720031738281,
      "num_input_tokens_seen": 21940240,
      "step": 1340,
      "train_runtime": 2581.4193,
      "train_tokens_per_second": 8499.293
    },
    {
      "epoch": 0.09056019391327767,
      "grad_norm": 1.3414518268429256,
      "learning_rate": 9.94957138548134e-06,
      "loss": 0.5193123817443848,
      "num_input_tokens_seen": 22022160,
      "step": 1345,
      "train_runtime": 2591.0425,
      "train_tokens_per_second": 8499.343
    },
    {
      "epoch": 0.09089684890923781,
      "grad_norm": 1.521332008926501,
      "learning_rate": 9.949196112518715e-06,
      "loss": 0.48576831817626953,
      "num_input_tokens_seen": 22104080,
      "step": 1350,
      "train_runtime": 2600.6648,
      "train_tokens_per_second": 8499.396
    },
    {
      "epoch": 0.09123350390519795,
      "grad_norm": 1.2495475267285732,
      "learning_rate": 9.94881945552918e-06,
      "loss": 0.4715422630310059,
      "num_input_tokens_seen": 22186000,
      "step": 1355,
      "train_runtime": 2610.1768,
      "train_tokens_per_second": 8499.807
    },
    {
      "epoch": 0.09157015890115809,
      "grad_norm": 1.1363745030958747,
      "learning_rate": 9.948441414618064e-06,
      "loss": 0.4907670974731445,
      "num_input_tokens_seen": 22267920,
      "step": 1360,
      "train_runtime": 2619.7899,
      "train_tokens_per_second": 8499.888
    },
    {
      "epoch": 0.09190681389711823,
      "grad_norm": 1.5790751677434862,
      "learning_rate": 9.948061989891085e-06,
      "loss": 0.4402290344238281,
      "num_input_tokens_seen": 22349840,
      "step": 1365,
      "train_runtime": 2629.4086,
      "train_tokens_per_second": 8499.949
    },
    {
      "epoch": 0.09224346889307837,
      "grad_norm": 1.4538486483658548,
      "learning_rate": 9.94768118145435e-06,
      "loss": 0.47271904945373533,
      "num_input_tokens_seen": 22431760,
      "step": 1370,
      "train_runtime": 2638.919,
      "train_tokens_per_second": 8500.36
    },
    {
      "epoch": 0.09258012388903851,
      "grad_norm": 1.4701034167432965,
      "learning_rate": 9.947298989414349e-06,
      "loss": 0.4599665641784668,
      "num_input_tokens_seen": 22513680,
      "step": 1375,
      "train_runtime": 2648.5371,
      "train_tokens_per_second": 8500.421
    },
    {
      "epoch": 0.09291677888499865,
      "grad_norm": 1.2937738539778552,
      "learning_rate": 9.946915413877963e-06,
      "loss": 0.4393216609954834,
      "num_input_tokens_seen": 22595600,
      "step": 1380,
      "train_runtime": 2658.1504,
      "train_tokens_per_second": 8500.497
    },
    {
      "epoch": 0.09325343388095879,
      "grad_norm": 1.4408304854197043,
      "learning_rate": 9.946530454952454e-06,
      "loss": 0.4748242378234863,
      "num_input_tokens_seen": 22677520,
      "step": 1385,
      "train_runtime": 2667.7004,
      "train_tokens_per_second": 8500.775
    },
    {
      "epoch": 0.09359008887691893,
      "grad_norm": 1.352042145787845,
      "learning_rate": 9.946144112745477e-06,
      "loss": 0.48343682289123535,
      "num_input_tokens_seen": 22759440,
      "step": 1390,
      "train_runtime": 2677.2708,
      "train_tokens_per_second": 8500.985
    },
    {
      "epoch": 0.09392674387287907,
      "grad_norm": 1.421870059849259,
      "learning_rate": 9.94575638736507e-06,
      "loss": 0.44847497940063474,
      "num_input_tokens_seen": 22841360,
      "step": 1395,
      "train_runtime": 2686.889,
      "train_tokens_per_second": 8501.044
    },
    {
      "epoch": 0.09426339886883921,
      "grad_norm": 1.4406865122623544,
      "learning_rate": 9.945367278919662e-06,
      "loss": 0.5123852729797364,
      "num_input_tokens_seen": 22923280,
      "step": 1400,
      "train_runtime": 2696.4489,
      "train_tokens_per_second": 8501.285
    },
    {
      "epoch": 0.09460005386479936,
      "grad_norm": 1.2733146238139843,
      "learning_rate": 9.944976787518064e-06,
      "loss": 0.4725001335144043,
      "num_input_tokens_seen": 23005200,
      "step": 1405,
      "train_runtime": 2706.0134,
      "train_tokens_per_second": 8501.51
    },
    {
      "epoch": 0.0949367088607595,
      "grad_norm": 1.6040747155171895,
      "learning_rate": 9.944584913269474e-06,
      "loss": 0.49306402206420896,
      "num_input_tokens_seen": 23087120,
      "step": 1410,
      "train_runtime": 2715.6206,
      "train_tokens_per_second": 8501.6
    },
    {
      "epoch": 0.09527336385671964,
      "grad_norm": 1.4020165336518258,
      "learning_rate": 9.944191656283481e-06,
      "loss": 0.4558537006378174,
      "num_input_tokens_seen": 23169040,
      "step": 1415,
      "train_runtime": 2725.2328,
      "train_tokens_per_second": 8501.674
    },
    {
      "epoch": 0.09561001885267978,
      "grad_norm": 1.528629253217098,
      "learning_rate": 9.943797016670059e-06,
      "loss": 0.4396226406097412,
      "num_input_tokens_seen": 23250960,
      "step": 1420,
      "train_runtime": 2734.7441,
      "train_tokens_per_second": 8502.061
    },
    {
      "epoch": 0.09594667384863992,
      "grad_norm": 1.309873734808832,
      "learning_rate": 9.943400994539565e-06,
      "loss": 0.44457759857177737,
      "num_input_tokens_seen": 23332880,
      "step": 1425,
      "train_runtime": 2744.3688,
      "train_tokens_per_second": 8502.093
    },
    {
      "epoch": 0.09628332884460006,
      "grad_norm": 1.189313501786972,
      "learning_rate": 9.943003590002744e-06,
      "loss": 0.43305983543396,
      "num_input_tokens_seen": 23414800,
      "step": 1430,
      "train_runtime": 2753.9755,
      "train_tokens_per_second": 8502.182
    },
    {
      "epoch": 0.0966199838405602,
      "grad_norm": 1.2589408208421022,
      "learning_rate": 9.942604803170733e-06,
      "loss": 0.4704472541809082,
      "num_input_tokens_seen": 23496608,
      "step": 1435,
      "train_runtime": 2764.0357,
      "train_tokens_per_second": 8500.834
    },
    {
      "epoch": 0.09695663883652034,
      "grad_norm": 1.196641163343287,
      "learning_rate": 9.94220463415505e-06,
      "loss": 0.4407841205596924,
      "num_input_tokens_seen": 23577728,
      "step": 1440,
      "train_runtime": 2773.8315,
      "train_tokens_per_second": 8500.058
    },
    {
      "epoch": 0.09729329383248048,
      "grad_norm": 1.2382998117581916,
      "learning_rate": 9.9418030830676e-06,
      "loss": 0.48865904808044436,
      "num_input_tokens_seen": 23659648,
      "step": 1445,
      "train_runtime": 2783.4358,
      "train_tokens_per_second": 8500.159
    },
    {
      "epoch": 0.09762994882844062,
      "grad_norm": 1.288846464388524,
      "learning_rate": 9.941400150020676e-06,
      "loss": 0.4640057563781738,
      "num_input_tokens_seen": 23741568,
      "step": 1450,
      "train_runtime": 2792.9947,
      "train_tokens_per_second": 8500.399
    },
    {
      "epoch": 0.09796660382440076,
      "grad_norm": 1.2882912184943514,
      "learning_rate": 9.940995835126957e-06,
      "loss": 0.5090675830841065,
      "num_input_tokens_seen": 23823488,
      "step": 1455,
      "train_runtime": 2802.573,
      "train_tokens_per_second": 8500.577
    },
    {
      "epoch": 0.0983032588203609,
      "grad_norm": 1.1417407973451787,
      "learning_rate": 9.940590138499508e-06,
      "loss": 0.46141514778137205,
      "num_input_tokens_seen": 23905408,
      "step": 1460,
      "train_runtime": 2812.1223,
      "train_tokens_per_second": 8500.842
    },
    {
      "epoch": 0.09863991381632103,
      "grad_norm": 1.124171680103979,
      "learning_rate": 9.940183060251783e-06,
      "loss": 0.41905922889709474,
      "num_input_tokens_seen": 23987328,
      "step": 1465,
      "train_runtime": 2821.6825,
      "train_tokens_per_second": 8501.073
    },
    {
      "epoch": 0.09897656881228117,
      "grad_norm": 1.1772240125861628,
      "learning_rate": 9.939774600497616e-06,
      "loss": 0.40096302032470704,
      "num_input_tokens_seen": 24069248,
      "step": 1470,
      "train_runtime": 2831.2455,
      "train_tokens_per_second": 8501.293
    },
    {
      "epoch": 0.09931322380824131,
      "grad_norm": 1.582528930717682,
      "learning_rate": 9.939364759351236e-06,
      "loss": 0.4646428108215332,
      "num_input_tokens_seen": 24151168,
      "step": 1475,
      "train_runtime": 2840.8369,
      "train_tokens_per_second": 8501.427
    },
    {
      "epoch": 0.09964987880420145,
      "grad_norm": 1.2087423580866878,
      "learning_rate": 9.938953536927249e-06,
      "loss": 0.43613157272338865,
      "num_input_tokens_seen": 24233088,
      "step": 1480,
      "train_runtime": 2850.3902,
      "train_tokens_per_second": 8501.674
    },
    {
      "epoch": 0.09998653380016159,
      "grad_norm": 1.3074045497023292,
      "learning_rate": 9.938540933340657e-06,
      "loss": 0.49275712966918944,
      "num_input_tokens_seen": 24315008,
      "step": 1485,
      "train_runtime": 2859.965,
      "train_tokens_per_second": 8501.855
    },
    {
      "epoch": 0.10032318879612173,
      "grad_norm": 1.1503553685202381,
      "learning_rate": 9.938126948706839e-06,
      "loss": 0.441726016998291,
      "num_input_tokens_seen": 24396928,
      "step": 1490,
      "train_runtime": 2869.4715,
      "train_tokens_per_second": 8502.237
    },
    {
      "epoch": 0.10065984379208187,
      "grad_norm": 1.2948639002286264,
      "learning_rate": 9.937711583141567e-06,
      "loss": 0.4858980178833008,
      "num_input_tokens_seen": 24478848,
      "step": 1495,
      "train_runtime": 2879.0813,
      "train_tokens_per_second": 8502.312
    },
    {
      "epoch": 0.10099649878804201,
      "grad_norm": 1.2513963534174573,
      "learning_rate": 9.937294836760995e-06,
      "loss": 0.4592639923095703,
      "num_input_tokens_seen": 24560768,
      "step": 1500,
      "train_runtime": 2888.684,
      "train_tokens_per_second": 8502.407
    },
    {
      "epoch": 0.10133315378400215,
      "grad_norm": 1.548188271594729,
      "learning_rate": 9.936876709681668e-06,
      "loss": 0.4954355239868164,
      "num_input_tokens_seen": 24642688,
      "step": 1505,
      "train_runtime": 2898.2304,
      "train_tokens_per_second": 8502.667
    },
    {
      "epoch": 0.10166980877996229,
      "grad_norm": 1.1468191984386327,
      "learning_rate": 9.93645720202051e-06,
      "loss": 0.434722375869751,
      "num_input_tokens_seen": 24724608,
      "step": 1510,
      "train_runtime": 2907.7976,
      "train_tokens_per_second": 8502.864
    },
    {
      "epoch": 0.10200646377592243,
      "grad_norm": 1.3027331929283412,
      "learning_rate": 9.936036313894836e-06,
      "loss": 0.5021354198455811,
      "num_input_tokens_seen": 24806360,
      "step": 1515,
      "train_runtime": 2917.917,
      "train_tokens_per_second": 8501.393
    },
    {
      "epoch": 0.10234311877188257,
      "grad_norm": 1.249145035924313,
      "learning_rate": 9.935614045422349e-06,
      "loss": 0.47060489654541016,
      "num_input_tokens_seen": 24888280,
      "step": 1520,
      "train_runtime": 2927.561,
      "train_tokens_per_second": 8501.37
    },
    {
      "epoch": 0.10267977376784271,
      "grad_norm": 1.2263890594354285,
      "learning_rate": 9.935190396721132e-06,
      "loss": 0.4087702751159668,
      "num_input_tokens_seen": 24970200,
      "step": 1525,
      "train_runtime": 2937.1145,
      "train_tokens_per_second": 8501.609
    },
    {
      "epoch": 0.10301642876380286,
      "grad_norm": 1.5355559172633964,
      "learning_rate": 9.934765367909658e-06,
      "loss": 0.4779378414154053,
      "num_input_tokens_seen": 25052120,
      "step": 1530,
      "train_runtime": 2946.6715,
      "train_tokens_per_second": 8501.837
    },
    {
      "epoch": 0.103353083759763,
      "grad_norm": 1.3966890499243032,
      "learning_rate": 9.934338959106783e-06,
      "loss": 0.4314533233642578,
      "num_input_tokens_seen": 25134040,
      "step": 1535,
      "train_runtime": 2956.2802,
      "train_tokens_per_second": 8501.914
    },
    {
      "epoch": 0.10368973875572314,
      "grad_norm": 1.603262865200472,
      "learning_rate": 9.933911170431753e-06,
      "loss": 0.49365696907043455,
      "num_input_tokens_seen": 25215232,
      "step": 1540,
      "train_runtime": 2966.0917,
      "train_tokens_per_second": 8501.164
    },
    {
      "epoch": 0.10402639375168328,
      "grad_norm": 1.2865954202144114,
      "learning_rate": 9.933482002004198e-06,
      "loss": 0.4462307929992676,
      "num_input_tokens_seen": 25297152,
      "step": 1545,
      "train_runtime": 2975.6007,
      "train_tokens_per_second": 8501.528
    },
    {
      "epoch": 0.10436304874764342,
      "grad_norm": 1.5650910387272232,
      "learning_rate": 9.933051453944134e-06,
      "loss": 0.4848374366760254,
      "num_input_tokens_seen": 25379072,
      "step": 1550,
      "train_runtime": 2985.2321,
      "train_tokens_per_second": 8501.541
    },
    {
      "epoch": 0.10469970374360356,
      "grad_norm": 1.249016028564887,
      "learning_rate": 9.932619526371958e-06,
      "loss": 0.4457110404968262,
      "num_input_tokens_seen": 25460992,
      "step": 1555,
      "train_runtime": 2994.8124,
      "train_tokens_per_second": 8501.698
    },
    {
      "epoch": 0.1050363587395637,
      "grad_norm": 1.3379116705335679,
      "learning_rate": 9.932186219408463e-06,
      "loss": 0.4413485527038574,
      "num_input_tokens_seen": 25542912,
      "step": 1560,
      "train_runtime": 3004.4142,
      "train_tokens_per_second": 8501.794
    },
    {
      "epoch": 0.10537301373552384,
      "grad_norm": 1.3343979854707015,
      "learning_rate": 9.931751533174819e-06,
      "loss": 0.46096324920654297,
      "num_input_tokens_seen": 25624832,
      "step": 1565,
      "train_runtime": 3013.9154,
      "train_tokens_per_second": 8502.174
    },
    {
      "epoch": 0.10570966873148398,
      "grad_norm": 1.3160825716834073,
      "learning_rate": 9.931315467792584e-06,
      "loss": 0.4740478038787842,
      "num_input_tokens_seen": 25706752,
      "step": 1570,
      "train_runtime": 3023.5239,
      "train_tokens_per_second": 8502.249
    },
    {
      "epoch": 0.10604632372744412,
      "grad_norm": 1.33409653798544,
      "learning_rate": 9.930878023383705e-06,
      "loss": 0.44043984413146975,
      "num_input_tokens_seen": 25788672,
      "step": 1575,
      "train_runtime": 3033.1448,
      "train_tokens_per_second": 8502.288
    },
    {
      "epoch": 0.10638297872340426,
      "grad_norm": 1.1516082442817897,
      "learning_rate": 9.93043920007051e-06,
      "loss": 0.4731790542602539,
      "num_input_tokens_seen": 25870592,
      "step": 1580,
      "train_runtime": 3042.698,
      "train_tokens_per_second": 8502.517
    },
    {
      "epoch": 0.1067196337193644,
      "grad_norm": 1.2876067139154308,
      "learning_rate": 9.929998997975712e-06,
      "loss": 0.4295042037963867,
      "num_input_tokens_seen": 25952512,
      "step": 1585,
      "train_runtime": 3052.2778,
      "train_tokens_per_second": 8502.67
    },
    {
      "epoch": 0.10705628871532454,
      "grad_norm": 1.526235024337487,
      "learning_rate": 9.929557417222417e-06,
      "loss": 0.4590339660644531,
      "num_input_tokens_seen": 26034432,
      "step": 1590,
      "train_runtime": 3061.8819,
      "train_tokens_per_second": 8502.755
    },
    {
      "epoch": 0.10739294371128467,
      "grad_norm": 1.3209187802452418,
      "learning_rate": 9.929114457934109e-06,
      "loss": 0.4657303333282471,
      "num_input_tokens_seen": 26115176,
      "step": 1595,
      "train_runtime": 3071.5994,
      "train_tokens_per_second": 8502.142
    },
    {
      "epoch": 0.10772959870724481,
      "grad_norm": 1.2825614997073713,
      "learning_rate": 9.92867012023466e-06,
      "loss": 0.4645052909851074,
      "num_input_tokens_seen": 26197096,
      "step": 1600,
      "train_runtime": 3081.1425,
      "train_tokens_per_second": 8502.397
    },
    {
      "epoch": 0.10806625370320495,
      "grad_norm": 1.2310152452303005,
      "learning_rate": 9.92822440424833e-06,
      "loss": 0.47003836631774903,
      "num_input_tokens_seen": 26279016,
      "step": 1605,
      "train_runtime": 3090.6979,
      "train_tokens_per_second": 8502.615
    },
    {
      "epoch": 0.1084029086991651,
      "grad_norm": 1.4151348456817163,
      "learning_rate": 9.927777310099759e-06,
      "loss": 0.5417779922485352,
      "num_input_tokens_seen": 26360936,
      "step": 1610,
      "train_runtime": 3100.3288,
      "train_tokens_per_second": 8502.626
    },
    {
      "epoch": 0.10873956369512523,
      "grad_norm": 1.5523483725630076,
      "learning_rate": 9.927328837913976e-06,
      "loss": 0.45824384689331055,
      "num_input_tokens_seen": 26442856,
      "step": 1615,
      "train_runtime": 3109.9302,
      "train_tokens_per_second": 8502.717
    },
    {
      "epoch": 0.10907621869108537,
      "grad_norm": 1.3096915053365048,
      "learning_rate": 9.926878987816397e-06,
      "loss": 0.504150676727295,
      "num_input_tokens_seen": 26524776,
      "step": 1620,
      "train_runtime": 3119.475,
      "train_tokens_per_second": 8502.962
    },
    {
      "epoch": 0.10941287368704551,
      "grad_norm": 1.2606351715291328,
      "learning_rate": 9.92642775993282e-06,
      "loss": 0.48359050750732424,
      "num_input_tokens_seen": 26606696,
      "step": 1625,
      "train_runtime": 3129.0296,
      "train_tokens_per_second": 8503.178
    },
    {
      "epoch": 0.10974952868300565,
      "grad_norm": 1.1538081576688077,
      "learning_rate": 9.925975154389428e-06,
      "loss": 0.4344356536865234,
      "num_input_tokens_seen": 26688616,
      "step": 1630,
      "train_runtime": 3138.6343,
      "train_tokens_per_second": 8503.258
    },
    {
      "epoch": 0.11008618367896579,
      "grad_norm": 1.0928611880930525,
      "learning_rate": 9.925521171312792e-06,
      "loss": 0.41940717697143554,
      "num_input_tokens_seen": 26770536,
      "step": 1635,
      "train_runtime": 3148.2696,
      "train_tokens_per_second": 8503.254
    },
    {
      "epoch": 0.11042283867492593,
      "grad_norm": 1.392673349993868,
      "learning_rate": 9.925065810829868e-06,
      "loss": 0.4688426494598389,
      "num_input_tokens_seen": 26852456,
      "step": 1640,
      "train_runtime": 3157.8124,
      "train_tokens_per_second": 8503.5
    },
    {
      "epoch": 0.11075949367088607,
      "grad_norm": 1.4726545333578833,
      "learning_rate": 9.924609073067994e-06,
      "loss": 0.47595720291137694,
      "num_input_tokens_seen": 26934376,
      "step": 1645,
      "train_runtime": 3167.3604,
      "train_tokens_per_second": 8503.73
    },
    {
      "epoch": 0.11109614866684622,
      "grad_norm": 1.2878695318323756,
      "learning_rate": 9.924150958154897e-06,
      "loss": 0.4883429050445557,
      "num_input_tokens_seen": 27016296,
      "step": 1650,
      "train_runtime": 3176.9579,
      "train_tokens_per_second": 8503.826
    },
    {
      "epoch": 0.11143280366280636,
      "grad_norm": 1.2884345098708248,
      "learning_rate": 9.923691466218686e-06,
      "loss": 0.4735406398773193,
      "num_input_tokens_seen": 27098216,
      "step": 1655,
      "train_runtime": 3186.5062,
      "train_tokens_per_second": 8504.052
    },
    {
      "epoch": 0.1117694586587665,
      "grad_norm": 1.3482976431847593,
      "learning_rate": 9.923230597387856e-06,
      "loss": 0.4414034843444824,
      "num_input_tokens_seen": 27180136,
      "step": 1660,
      "train_runtime": 3196.0871,
      "train_tokens_per_second": 8504.191
    },
    {
      "epoch": 0.11210611365472664,
      "grad_norm": 1.3999314963277785,
      "learning_rate": 9.922768351791289e-06,
      "loss": 0.47736682891845705,
      "num_input_tokens_seen": 27262056,
      "step": 1665,
      "train_runtime": 3205.7202,
      "train_tokens_per_second": 8504.191
    },
    {
      "epoch": 0.11244276865068678,
      "grad_norm": 1.415957385522472,
      "learning_rate": 9.92230472955825e-06,
      "loss": 0.4359127044677734,
      "num_input_tokens_seen": 27343976,
      "step": 1670,
      "train_runtime": 3215.3139,
      "train_tokens_per_second": 8504.294
    },
    {
      "epoch": 0.11277942364664692,
      "grad_norm": 1.1413297831301632,
      "learning_rate": 9.92183973081839e-06,
      "loss": 0.4285294532775879,
      "num_input_tokens_seen": 27425896,
      "step": 1675,
      "train_runtime": 3224.8183,
      "train_tokens_per_second": 8504.633
    },
    {
      "epoch": 0.11311607864260706,
      "grad_norm": 1.2221291740710125,
      "learning_rate": 9.921373355701743e-06,
      "loss": 0.5141834259033203,
      "num_input_tokens_seen": 27507816,
      "step": 1680,
      "train_runtime": 3234.4292,
      "train_tokens_per_second": 8504.689
    },
    {
      "epoch": 0.1134527336385672,
      "grad_norm": 1.4121719942361535,
      "learning_rate": 9.920905604338732e-06,
      "loss": 0.5054280281066894,
      "num_input_tokens_seen": 27589736,
      "step": 1685,
      "train_runtime": 3244.0421,
      "train_tokens_per_second": 8504.74
    },
    {
      "epoch": 0.11378938863452734,
      "grad_norm": 1.2802528881717004,
      "learning_rate": 9.920436476860158e-06,
      "loss": 0.41542892456054686,
      "num_input_tokens_seen": 27671656,
      "step": 1690,
      "train_runtime": 3253.5945,
      "train_tokens_per_second": 8504.949
    },
    {
      "epoch": 0.11412604363048748,
      "grad_norm": 1.380047761684455,
      "learning_rate": 9.919965973397214e-06,
      "loss": 0.43852930068969725,
      "num_input_tokens_seen": 27753576,
      "step": 1695,
      "train_runtime": 3263.1671,
      "train_tokens_per_second": 8505.104
    },
    {
      "epoch": 0.11446269862644762,
      "grad_norm": 1.5913579537853728,
      "learning_rate": 9.919494094081475e-06,
      "loss": 0.5007463932037354,
      "num_input_tokens_seen": 27835496,
      "step": 1700,
      "train_runtime": 3272.7914,
      "train_tokens_per_second": 8505.124
    },
    {
      "epoch": 0.11479935362240776,
      "grad_norm": 1.3629072607124249,
      "learning_rate": 9.919020839044899e-06,
      "loss": 0.4702188491821289,
      "num_input_tokens_seen": 27917416,
      "step": 1705,
      "train_runtime": 3282.3374,
      "train_tokens_per_second": 8505.346
    },
    {
      "epoch": 0.1151360086183679,
      "grad_norm": 1.3250572205412514,
      "learning_rate": 9.918546208419832e-06,
      "loss": 0.451420259475708,
      "num_input_tokens_seen": 27999336,
      "step": 1710,
      "train_runtime": 3291.9022,
      "train_tokens_per_second": 8505.519
    },
    {
      "epoch": 0.11547266361432804,
      "grad_norm": 1.3415822223331002,
      "learning_rate": 9.918070202339e-06,
      "loss": 0.44277048110961914,
      "num_input_tokens_seen": 28081256,
      "step": 1715,
      "train_runtime": 3301.5129,
      "train_tokens_per_second": 8505.572
    },
    {
      "epoch": 0.11580931861028818,
      "grad_norm": 1.370861619168358,
      "learning_rate": 9.91759282093552e-06,
      "loss": 0.42481560707092286,
      "num_input_tokens_seen": 28163176,
      "step": 1720,
      "train_runtime": 3311.019,
      "train_tokens_per_second": 8505.894
    },
    {
      "epoch": 0.11614597360624832,
      "grad_norm": 1.1622582743079934,
      "learning_rate": 9.917114064342888e-06,
      "loss": 0.421583890914917,
      "num_input_tokens_seen": 28244368,
      "step": 1725,
      "train_runtime": 3320.7523,
      "train_tokens_per_second": 8505.413
    },
    {
      "epoch": 0.11648262860220845,
      "grad_norm": 1.355215744139873,
      "learning_rate": 9.916633932694988e-06,
      "loss": 0.45598583221435546,
      "num_input_tokens_seen": 28326288,
      "step": 1730,
      "train_runtime": 3330.3816,
      "train_tokens_per_second": 8505.418
    },
    {
      "epoch": 0.1168192835981686,
      "grad_norm": 1.2544397138207652,
      "learning_rate": 9.916152426126086e-06,
      "loss": 0.3725733757019043,
      "num_input_tokens_seen": 28408208,
      "step": 1735,
      "train_runtime": 3339.9903,
      "train_tokens_per_second": 8505.476
    },
    {
      "epoch": 0.11715593859412873,
      "grad_norm": 1.4535686444575866,
      "learning_rate": 9.915669544770837e-06,
      "loss": 0.4538601875305176,
      "num_input_tokens_seen": 28490128,
      "step": 1740,
      "train_runtime": 3349.4965,
      "train_tokens_per_second": 8505.794
    },
    {
      "epoch": 0.11749259359008887,
      "grad_norm": 1.1426629062552305,
      "learning_rate": 9.915185288764272e-06,
      "loss": 0.46796045303344724,
      "num_input_tokens_seen": 28572048,
      "step": 1745,
      "train_runtime": 3359.1047,
      "train_tokens_per_second": 8505.852
    },
    {
      "epoch": 0.11782924858604901,
      "grad_norm": 1.2780829529376905,
      "learning_rate": 9.914699658241815e-06,
      "loss": 0.4281333923339844,
      "num_input_tokens_seen": 28653968,
      "step": 1750,
      "train_runtime": 3368.7161,
      "train_tokens_per_second": 8505.902
    },
    {
      "epoch": 0.11816590358200915,
      "grad_norm": 1.2933145952581953,
      "learning_rate": 9.91421265333927e-06,
      "loss": 0.4601719379425049,
      "num_input_tokens_seen": 28735888,
      "step": 1755,
      "train_runtime": 3378.2251,
      "train_tokens_per_second": 8506.209
    },
    {
      "epoch": 0.11850255857796929,
      "grad_norm": 1.1701030250025612,
      "learning_rate": 9.913724274192829e-06,
      "loss": 0.46593704223632815,
      "num_input_tokens_seen": 28817808,
      "step": 1760,
      "train_runtime": 3387.863,
      "train_tokens_per_second": 8506.19
    },
    {
      "epoch": 0.11883921357392943,
      "grad_norm": 1.2808063009300412,
      "learning_rate": 9.91323452093906e-06,
      "loss": 0.41030025482177734,
      "num_input_tokens_seen": 28899728,
      "step": 1765,
      "train_runtime": 3397.4735,
      "train_tokens_per_second": 8506.241
    },
    {
      "epoch": 0.11917586856988957,
      "grad_norm": 1.3441538415587657,
      "learning_rate": 9.912743393714927e-06,
      "loss": 0.42085628509521483,
      "num_input_tokens_seen": 28981648,
      "step": 1770,
      "train_runtime": 3407.0334,
      "train_tokens_per_second": 8506.417
    },
    {
      "epoch": 0.11951252356584972,
      "grad_norm": 1.6039056147071364,
      "learning_rate": 9.912250892657769e-06,
      "loss": 0.43581132888793944,
      "num_input_tokens_seen": 29063568,
      "step": 1775,
      "train_runtime": 3416.6099,
      "train_tokens_per_second": 8506.552
    },
    {
      "epoch": 0.11984917856180986,
      "grad_norm": 1.3812884555720064,
      "learning_rate": 9.911757017905312e-06,
      "loss": 0.45807590484619143,
      "num_input_tokens_seen": 29145488,
      "step": 1780,
      "train_runtime": 3426.231,
      "train_tokens_per_second": 8506.574
    },
    {
      "epoch": 0.12018583355777,
      "grad_norm": 1.509993712404879,
      "learning_rate": 9.911261769595668e-06,
      "loss": 0.4546051979064941,
      "num_input_tokens_seen": 29227408,
      "step": 1785,
      "train_runtime": 3435.7817,
      "train_tokens_per_second": 8506.771
    },
    {
      "epoch": 0.12052248855373014,
      "grad_norm": 1.1429389030179467,
      "learning_rate": 9.910765147867332e-06,
      "loss": 0.4275208950042725,
      "num_input_tokens_seen": 29309328,
      "step": 1790,
      "train_runtime": 3445.3424,
      "train_tokens_per_second": 8506.942
    },
    {
      "epoch": 0.12085914354969028,
      "grad_norm": 1.3076042093103168,
      "learning_rate": 9.91026715285918e-06,
      "loss": 0.45406379699707033,
      "num_input_tokens_seen": 29391248,
      "step": 1795,
      "train_runtime": 3454.9611,
      "train_tokens_per_second": 8506.969
    },
    {
      "epoch": 0.12119579854565042,
      "grad_norm": 1.4356376137474554,
      "learning_rate": 9.909767784710476e-06,
      "loss": 0.4969953536987305,
      "num_input_tokens_seen": 29473168,
      "step": 1800,
      "train_runtime": 3464.5819,
      "train_tokens_per_second": 8506.991
    },
    {
      "epoch": 0.12153245354161056,
      "grad_norm": 1.1967510312621676,
      "learning_rate": 9.909267043560868e-06,
      "loss": 0.4781789779663086,
      "num_input_tokens_seen": 29554736,
      "step": 1805,
      "train_runtime": 3474.0971,
      "train_tokens_per_second": 8507.17
    },
    {
      "epoch": 0.1218691085375707,
      "grad_norm": 1.342691711352661,
      "learning_rate": 9.908764929550388e-06,
      "loss": 0.46741695404052735,
      "num_input_tokens_seen": 29636656,
      "step": 1810,
      "train_runtime": 3483.6715,
      "train_tokens_per_second": 8507.305
    },
    {
      "epoch": 0.12220576353353084,
      "grad_norm": 1.1433442879668085,
      "learning_rate": 9.908261442819444e-06,
      "loss": 0.44350738525390626,
      "num_input_tokens_seen": 29718576,
      "step": 1815,
      "train_runtime": 3493.2795,
      "train_tokens_per_second": 8507.357
    },
    {
      "epoch": 0.12254241852949098,
      "grad_norm": 1.4746773763863592,
      "learning_rate": 9.907756583508842e-06,
      "loss": 0.4932860374450684,
      "num_input_tokens_seen": 29800496,
      "step": 1820,
      "train_runtime": 3502.7867,
      "train_tokens_per_second": 8507.654
    },
    {
      "epoch": 0.12287907352545112,
      "grad_norm": 1.2310568124233834,
      "learning_rate": 9.907250351759761e-06,
      "loss": 0.43315463066101073,
      "num_input_tokens_seen": 29882416,
      "step": 1825,
      "train_runtime": 3512.4154,
      "train_tokens_per_second": 8507.654
    },
    {
      "epoch": 0.12321572852141126,
      "grad_norm": 1.233654937122128,
      "learning_rate": 9.906742747713766e-06,
      "loss": 0.43630223274230956,
      "num_input_tokens_seen": 29964336,
      "step": 1830,
      "train_runtime": 3522.0446,
      "train_tokens_per_second": 8507.654
    },
    {
      "epoch": 0.1235523835173714,
      "grad_norm": 1.212084502003912,
      "learning_rate": 9.906233771512808e-06,
      "loss": 0.46129393577575684,
      "num_input_tokens_seen": 30046256,
      "step": 1835,
      "train_runtime": 3531.5515,
      "train_tokens_per_second": 8507.948
    },
    {
      "epoch": 0.12388903851333154,
      "grad_norm": 1.6930864511870543,
      "learning_rate": 9.905723423299222e-06,
      "loss": 0.5001296520233154,
      "num_input_tokens_seen": 30128176,
      "step": 1840,
      "train_runtime": 3541.1908,
      "train_tokens_per_second": 8507.922
    },
    {
      "epoch": 0.12422569350929168,
      "grad_norm": 1.2565612547716776,
      "learning_rate": 9.905211703215723e-06,
      "loss": 0.437269401550293,
      "num_input_tokens_seen": 30209600,
      "step": 1845,
      "train_runtime": 3551.1096,
      "train_tokens_per_second": 8507.087
    },
    {
      "epoch": 0.12456234850525182,
      "grad_norm": 0.9969295408127448,
      "learning_rate": 9.904698611405412e-06,
      "loss": 0.41403942108154296,
      "num_input_tokens_seen": 30291520,
      "step": 1850,
      "train_runtime": 3560.6662,
      "train_tokens_per_second": 8507.262
    },
    {
      "epoch": 0.12489900350121196,
      "grad_norm": 1.3147630106782393,
      "learning_rate": 9.904184148011774e-06,
      "loss": 0.4702615737915039,
      "num_input_tokens_seen": 30373440,
      "step": 1855,
      "train_runtime": 3570.2281,
      "train_tokens_per_second": 8507.423
    },
    {
      "epoch": 0.1252356584971721,
      "grad_norm": 1.423156385710832,
      "learning_rate": 9.903668313178681e-06,
      "loss": 0.4794650077819824,
      "num_input_tokens_seen": 30455360,
      "step": 1860,
      "train_runtime": 3579.843,
      "train_tokens_per_second": 8507.457
    },
    {
      "epoch": 0.12557231349313225,
      "grad_norm": 1.2275055629929208,
      "learning_rate": 9.903151107050377e-06,
      "loss": 0.41808629035949707,
      "num_input_tokens_seen": 30537216,
      "step": 1865,
      "train_runtime": 3590.0287,
      "train_tokens_per_second": 8506.12
    },
    {
      "epoch": 0.1259089684890924,
      "grad_norm": 1.3319527315349933,
      "learning_rate": 9.902632529771502e-06,
      "loss": 0.4679993152618408,
      "num_input_tokens_seen": 30619136,
      "step": 1870,
      "train_runtime": 3599.5357,
      "train_tokens_per_second": 8506.413
    },
    {
      "epoch": 0.12624562348505253,
      "grad_norm": 1.3459552512110742,
      "learning_rate": 9.902112581487074e-06,
      "loss": 0.44898433685302735,
      "num_input_tokens_seen": 30701056,
      "step": 1875,
      "train_runtime": 3609.0936,
      "train_tokens_per_second": 8506.583
    },
    {
      "epoch": 0.12658227848101267,
      "grad_norm": 1.3624115772069854,
      "learning_rate": 9.901591262342494e-06,
      "loss": 0.4598982810974121,
      "num_input_tokens_seen": 30782976,
      "step": 1880,
      "train_runtime": 3618.7324,
      "train_tokens_per_second": 8506.563
    },
    {
      "epoch": 0.1269189334769728,
      "grad_norm": 1.5102833797005606,
      "learning_rate": 9.901068572483548e-06,
      "loss": 0.43053579330444336,
      "num_input_tokens_seen": 30864896,
      "step": 1885,
      "train_runtime": 3628.2371,
      "train_tokens_per_second": 8506.857
    },
    {
      "epoch": 0.12725558847293295,
      "grad_norm": 1.3507635321906766,
      "learning_rate": 9.900544512056402e-06,
      "loss": 0.5108934879302979,
      "num_input_tokens_seen": 30946816,
      "step": 1890,
      "train_runtime": 3637.8617,
      "train_tokens_per_second": 8506.87
    },
    {
      "epoch": 0.12759224346889309,
      "grad_norm": 1.3150847146816664,
      "learning_rate": 9.900019081207612e-06,
      "loss": 0.4436016082763672,
      "num_input_tokens_seen": 31028736,
      "step": 1895,
      "train_runtime": 3647.461,
      "train_tokens_per_second": 8506.941
    },
    {
      "epoch": 0.12792889846485322,
      "grad_norm": 1.4502431438833698,
      "learning_rate": 9.89949228008411e-06,
      "loss": 0.4544196128845215,
      "num_input_tokens_seen": 31110656,
      "step": 1900,
      "train_runtime": 3656.954,
      "train_tokens_per_second": 8507.259
    },
    {
      "epoch": 0.12826555346081336,
      "grad_norm": 1.3094893088837114,
      "learning_rate": 9.898964108833216e-06,
      "loss": 0.47542829513549806,
      "num_input_tokens_seen": 31192576,
      "step": 1905,
      "train_runtime": 3666.5678,
      "train_tokens_per_second": 8507.296
    },
    {
      "epoch": 0.1286022084567735,
      "grad_norm": 1.5274493729192724,
      "learning_rate": 9.898434567602629e-06,
      "loss": 0.47165703773498535,
      "num_input_tokens_seen": 31274496,
      "step": 1910,
      "train_runtime": 3676.1867,
      "train_tokens_per_second": 8507.32
    },
    {
      "epoch": 0.12893886345273364,
      "grad_norm": 1.2520125250279648,
      "learning_rate": 9.897903656540437e-06,
      "loss": 0.47268152236938477,
      "num_input_tokens_seen": 31356416,
      "step": 1915,
      "train_runtime": 3685.7373,
      "train_tokens_per_second": 8507.502
    },
    {
      "epoch": 0.12927551844869378,
      "grad_norm": 1.360241714671591,
      "learning_rate": 9.897371375795103e-06,
      "loss": 0.45409770011901857,
      "num_input_tokens_seen": 31438336,
      "step": 1920,
      "train_runtime": 3695.303,
      "train_tokens_per_second": 8507.648
    },
    {
      "epoch": 0.12961217344465392,
      "grad_norm": 1.2309756704286872,
      "learning_rate": 9.896837725515484e-06,
      "loss": 0.4227640151977539,
      "num_input_tokens_seen": 31520256,
      "step": 1925,
      "train_runtime": 3704.9391,
      "train_tokens_per_second": 8507.631
    },
    {
      "epoch": 0.12994882844061406,
      "grad_norm": 1.1716880476339786,
      "learning_rate": 9.89630270585081e-06,
      "loss": 0.4504727840423584,
      "num_input_tokens_seen": 31602176,
      "step": 1930,
      "train_runtime": 3714.4936,
      "train_tokens_per_second": 8507.802
    },
    {
      "epoch": 0.1302854834365742,
      "grad_norm": 1.3161468984305853,
      "learning_rate": 9.895766316950694e-06,
      "loss": 0.5064333915710449,
      "num_input_tokens_seen": 31684096,
      "step": 1935,
      "train_runtime": 3724.0539,
      "train_tokens_per_second": 8507.958
    },
    {
      "epoch": 0.13062213843253434,
      "grad_norm": 1.2895778803786522,
      "learning_rate": 9.89522855896514e-06,
      "loss": 0.4501958847045898,
      "num_input_tokens_seen": 31766016,
      "step": 1940,
      "train_runtime": 3733.6676,
      "train_tokens_per_second": 8507.992
    },
    {
      "epoch": 0.13095879342849448,
      "grad_norm": 1.3110348816022837,
      "learning_rate": 9.89468943204453e-06,
      "loss": 0.4589832305908203,
      "num_input_tokens_seen": 31847936,
      "step": 1945,
      "train_runtime": 3743.2788,
      "train_tokens_per_second": 8508.032
    },
    {
      "epoch": 0.13129544842445462,
      "grad_norm": 1.3173356225839254,
      "learning_rate": 9.89414893633963e-06,
      "loss": 0.44786930084228516,
      "num_input_tokens_seen": 31929648,
      "step": 1950,
      "train_runtime": 3753.2609,
      "train_tokens_per_second": 8507.175
    },
    {
      "epoch": 0.13163210342041476,
      "grad_norm": 1.2140630111488038,
      "learning_rate": 9.893607072001584e-06,
      "loss": 0.4498766899108887,
      "num_input_tokens_seen": 32011568,
      "step": 1955,
      "train_runtime": 3762.8766,
      "train_tokens_per_second": 8507.206
    },
    {
      "epoch": 0.1319687584163749,
      "grad_norm": 1.3134919905093254,
      "learning_rate": 9.893063839181925e-06,
      "loss": 0.4392902374267578,
      "num_input_tokens_seen": 32093488,
      "step": 1960,
      "train_runtime": 3772.515,
      "train_tokens_per_second": 8507.186
    },
    {
      "epoch": 0.13230541341233504,
      "grad_norm": 1.2982121116635907,
      "learning_rate": 9.892519238032566e-06,
      "loss": 0.4505149364471436,
      "num_input_tokens_seen": 32175408,
      "step": 1965,
      "train_runtime": 3782.0212,
      "train_tokens_per_second": 8507.464
    },
    {
      "epoch": 0.13264206840829518,
      "grad_norm": 1.3301442779191317,
      "learning_rate": 9.891973268705804e-06,
      "loss": 0.4733837604522705,
      "num_input_tokens_seen": 32257328,
      "step": 1970,
      "train_runtime": 3791.6436,
      "train_tokens_per_second": 8507.479
    },
    {
      "epoch": 0.13297872340425532,
      "grad_norm": 1.2400593569195915,
      "learning_rate": 9.891425931354316e-06,
      "loss": 0.4344335556030273,
      "num_input_tokens_seen": 32339248,
      "step": 1975,
      "train_runtime": 3801.2563,
      "train_tokens_per_second": 8507.516
    },
    {
      "epoch": 0.13331537840021546,
      "grad_norm": 1.3037163016303945,
      "learning_rate": 9.890877226131165e-06,
      "loss": 0.41803750991821287,
      "num_input_tokens_seen": 32421168,
      "step": 1980,
      "train_runtime": 3810.8117,
      "train_tokens_per_second": 8507.68
    },
    {
      "epoch": 0.1336520333961756,
      "grad_norm": 1.2322397333029596,
      "learning_rate": 9.890327153189793e-06,
      "loss": 0.40838894844055174,
      "num_input_tokens_seen": 32503088,
      "step": 1985,
      "train_runtime": 3820.6203,
      "train_tokens_per_second": 8507.28
    },
    {
      "epoch": 0.13398868839213574,
      "grad_norm": 1.3375766856996845,
      "learning_rate": 9.889775712684026e-06,
      "loss": 0.451218318939209,
      "num_input_tokens_seen": 32585008,
      "step": 1990,
      "train_runtime": 3830.144,
      "train_tokens_per_second": 8507.515
    },
    {
      "epoch": 0.13432534338809587,
      "grad_norm": 1.140858232977991,
      "learning_rate": 9.889222904768074e-06,
      "loss": 0.4191764831542969,
      "num_input_tokens_seen": 32666928,
      "step": 1995,
      "train_runtime": 3839.7145,
      "train_tokens_per_second": 8507.645
    },
    {
      "epoch": 0.13466199838405601,
      "grad_norm": 1.2690304660408125,
      "learning_rate": 9.888668729596528e-06,
      "loss": 0.4338832378387451,
      "num_input_tokens_seen": 32748848,
      "step": 2000,
      "train_runtime": 3849.3264,
      "train_tokens_per_second": 8507.683
    },
    {
      "epoch": 0.13499865338001615,
      "grad_norm": 1.3668072432058989,
      "learning_rate": 9.88811318732436e-06,
      "loss": 0.46068515777587893,
      "num_input_tokens_seen": 32830664,
      "step": 2005,
      "train_runtime": 3859.7007,
      "train_tokens_per_second": 8506.013
    },
    {
      "epoch": 0.1353353083759763,
      "grad_norm": 1.5788024472009023,
      "learning_rate": 9.887556278106927e-06,
      "loss": 0.447916316986084,
      "num_input_tokens_seen": 32912584,
      "step": 2010,
      "train_runtime": 3869.554,
      "train_tokens_per_second": 8505.524
    },
    {
      "epoch": 0.13567196337193643,
      "grad_norm": 1.3047379184749133,
      "learning_rate": 9.886998002099964e-06,
      "loss": 0.47930493354797366,
      "num_input_tokens_seen": 32994504,
      "step": 2015,
      "train_runtime": 3879.4802,
      "train_tokens_per_second": 8504.878
    },
    {
      "epoch": 0.13600861836789657,
      "grad_norm": 1.2344984828742016,
      "learning_rate": 9.886438359459595e-06,
      "loss": 0.4891185760498047,
      "num_input_tokens_seen": 33076424,
      "step": 2020,
      "train_runtime": 3889.0888,
      "train_tokens_per_second": 8504.929
    },
    {
      "epoch": 0.1363452733638567,
      "grad_norm": 1.3001543737767831,
      "learning_rate": 9.885877350342319e-06,
      "loss": 0.4543811798095703,
      "num_input_tokens_seen": 33158344,
      "step": 2025,
      "train_runtime": 3898.6366,
      "train_tokens_per_second": 8505.113
    },
    {
      "epoch": 0.13668192835981685,
      "grad_norm": 1.2410802637882201,
      "learning_rate": 9.885314974905024e-06,
      "loss": 0.4188577175140381,
      "num_input_tokens_seen": 33240264,
      "step": 2030,
      "train_runtime": 3908.1977,
      "train_tokens_per_second": 8505.267
    },
    {
      "epoch": 0.137018583355777,
      "grad_norm": 1.3955222361745654,
      "learning_rate": 9.884751233304973e-06,
      "loss": 0.46915860176086427,
      "num_input_tokens_seen": 33322184,
      "step": 2035,
      "train_runtime": 3917.8073,
      "train_tokens_per_second": 8505.315
    },
    {
      "epoch": 0.13735523835173713,
      "grad_norm": 1.4110475047225328,
      "learning_rate": 9.884186125699816e-06,
      "loss": 0.5105852603912353,
      "num_input_tokens_seen": 33404104,
      "step": 2040,
      "train_runtime": 3927.4246,
      "train_tokens_per_second": 8505.346
    },
    {
      "epoch": 0.13769189334769727,
      "grad_norm": 1.2964214350215517,
      "learning_rate": 9.883619652247584e-06,
      "loss": 0.4454387664794922,
      "num_input_tokens_seen": 33486024,
      "step": 2045,
      "train_runtime": 3936.9802,
      "train_tokens_per_second": 8505.51
    },
    {
      "epoch": 0.1380285483436574,
      "grad_norm": 1.2028162070163437,
      "learning_rate": 9.883051813106687e-06,
      "loss": 0.3921892404556274,
      "num_input_tokens_seen": 33567944,
      "step": 2050,
      "train_runtime": 3946.53,
      "train_tokens_per_second": 8505.686
    },
    {
      "epoch": 0.13836520333961755,
      "grad_norm": 1.5077776225113901,
      "learning_rate": 9.882482608435924e-06,
      "loss": 0.49668121337890625,
      "num_input_tokens_seen": 33649864,
      "step": 2055,
      "train_runtime": 3956.1321,
      "train_tokens_per_second": 8505.748
    },
    {
      "epoch": 0.1387018583355777,
      "grad_norm": 1.2255204261090815,
      "learning_rate": 9.881912038394468e-06,
      "loss": 0.44399189949035645,
      "num_input_tokens_seen": 33731784,
      "step": 2060,
      "train_runtime": 3965.6922,
      "train_tokens_per_second": 8505.901
    },
    {
      "epoch": 0.13903851333153783,
      "grad_norm": 1.203118599734307,
      "learning_rate": 9.881340103141878e-06,
      "loss": 0.4436651229858398,
      "num_input_tokens_seen": 33813704,
      "step": 2065,
      "train_runtime": 3975.1904,
      "train_tokens_per_second": 8506.185
    },
    {
      "epoch": 0.13937516832749797,
      "grad_norm": 1.0044045186722828,
      "learning_rate": 9.880766802838092e-06,
      "loss": 0.42821512222290037,
      "num_input_tokens_seen": 33895624,
      "step": 2070,
      "train_runtime": 3984.8165,
      "train_tokens_per_second": 8506.194
    },
    {
      "epoch": 0.1397118233234581,
      "grad_norm": 1.1550539843419323,
      "learning_rate": 9.880192137643434e-06,
      "loss": 0.4165791988372803,
      "num_input_tokens_seen": 33977544,
      "step": 2075,
      "train_runtime": 3994.4246,
      "train_tokens_per_second": 8506.242
    },
    {
      "epoch": 0.14004847831941827,
      "grad_norm": 1.7301546425045358,
      "learning_rate": 9.879616107718608e-06,
      "loss": 0.4200141429901123,
      "num_input_tokens_seen": 34059464,
      "step": 2080,
      "train_runtime": 4004.0271,
      "train_tokens_per_second": 8506.302
    },
    {
      "epoch": 0.1403851333153784,
      "grad_norm": 1.3653932648008262,
      "learning_rate": 9.879038713224695e-06,
      "loss": 0.4574295997619629,
      "num_input_tokens_seen": 34141384,
      "step": 2085,
      "train_runtime": 4013.5334,
      "train_tokens_per_second": 8506.565
    },
    {
      "epoch": 0.14072178831133855,
      "grad_norm": 1.2932451100104374,
      "learning_rate": 9.878459954323165e-06,
      "loss": 0.45157876014709475,
      "num_input_tokens_seen": 34223304,
      "step": 2090,
      "train_runtime": 4023.1576,
      "train_tokens_per_second": 8506.578
    },
    {
      "epoch": 0.1410584433072987,
      "grad_norm": 1.2500261400957167,
      "learning_rate": 9.877879831175865e-06,
      "loss": 0.4439502716064453,
      "num_input_tokens_seen": 34305224,
      "step": 2095,
      "train_runtime": 4032.7845,
      "train_tokens_per_second": 8506.585
    },
    {
      "epoch": 0.14139509830325883,
      "grad_norm": 1.2270671673520623,
      "learning_rate": 9.877298343945025e-06,
      "loss": 0.44665069580078126,
      "num_input_tokens_seen": 34387144,
      "step": 2100,
      "train_runtime": 4042.3885,
      "train_tokens_per_second": 8506.64
    },
    {
      "epoch": 0.14173175329921897,
      "grad_norm": 1.0753642510592378,
      "learning_rate": 9.876715492793256e-06,
      "loss": 0.4314713001251221,
      "num_input_tokens_seen": 34469064,
      "step": 2105,
      "train_runtime": 4051.8843,
      "train_tokens_per_second": 8506.922
    },
    {
      "epoch": 0.1420684082951791,
      "grad_norm": 1.2244455197772863,
      "learning_rate": 9.87613127788355e-06,
      "loss": 0.39420418739318847,
      "num_input_tokens_seen": 34550984,
      "step": 2110,
      "train_runtime": 4061.4837,
      "train_tokens_per_second": 8506.986
    },
    {
      "epoch": 0.14240506329113925,
      "grad_norm": 1.445949628694106,
      "learning_rate": 9.87554569937928e-06,
      "loss": 0.4866490364074707,
      "num_input_tokens_seen": 34632904,
      "step": 2115,
      "train_runtime": 4071.0981,
      "train_tokens_per_second": 8507.018
    },
    {
      "epoch": 0.1427417182870994,
      "grad_norm": 1.3217265679982484,
      "learning_rate": 9.874958757444203e-06,
      "loss": 0.48250637054443357,
      "num_input_tokens_seen": 34714824,
      "step": 2120,
      "train_runtime": 4080.6482,
      "train_tokens_per_second": 8507.184
    },
    {
      "epoch": 0.14307837328305953,
      "grad_norm": 1.2261599444818778,
      "learning_rate": 9.874370452242454e-06,
      "loss": 0.47803549766540526,
      "num_input_tokens_seen": 34796744,
      "step": 2125,
      "train_runtime": 4090.1526,
      "train_tokens_per_second": 8507.444
    },
    {
      "epoch": 0.14341502827901967,
      "grad_norm": 1.2358462023274313,
      "learning_rate": 9.873780783938553e-06,
      "loss": 0.5271174430847168,
      "num_input_tokens_seen": 34878664,
      "step": 2130,
      "train_runtime": 4099.7526,
      "train_tokens_per_second": 8507.505
    },
    {
      "epoch": 0.1437516832749798,
      "grad_norm": 1.1014696685631542,
      "learning_rate": 9.873189752697396e-06,
      "loss": 0.42470674514770507,
      "num_input_tokens_seen": 34960584,
      "step": 2135,
      "train_runtime": 4109.3495,
      "train_tokens_per_second": 8507.571
    },
    {
      "epoch": 0.14408833827093995,
      "grad_norm": 1.2827146857063543,
      "learning_rate": 9.872597358684265e-06,
      "loss": 0.46950674057006836,
      "num_input_tokens_seen": 35042504,
      "step": 2140,
      "train_runtime": 4118.8919,
      "train_tokens_per_second": 8507.75
    },
    {
      "epoch": 0.1444249932669001,
      "grad_norm": 1.2027634432414949,
      "learning_rate": 9.872003602064816e-06,
      "loss": 0.46025867462158204,
      "num_input_tokens_seen": 35124424,
      "step": 2145,
      "train_runtime": 4128.4462,
      "train_tokens_per_second": 8507.904
    },
    {
      "epoch": 0.14476164826286023,
      "grad_norm": 1.2293249942558635,
      "learning_rate": 9.8714084830051e-06,
      "loss": 0.45737490653991697,
      "num_input_tokens_seen": 35206344,
      "step": 2150,
      "train_runtime": 4138.0576,
      "train_tokens_per_second": 8507.94
    },
    {
      "epoch": 0.14509830325882037,
      "grad_norm": 1.028420529529435,
      "learning_rate": 9.870812001671533e-06,
      "loss": 0.47080326080322266,
      "num_input_tokens_seen": 35288264,
      "step": 2155,
      "train_runtime": 4147.663,
      "train_tokens_per_second": 8507.987
    },
    {
      "epoch": 0.1454349582547805,
      "grad_norm": 1.4866773893606018,
      "learning_rate": 9.870214158230922e-06,
      "loss": 0.4912544250488281,
      "num_input_tokens_seen": 35370184,
      "step": 2160,
      "train_runtime": 4157.1682,
      "train_tokens_per_second": 8508.24
    },
    {
      "epoch": 0.14577161325074064,
      "grad_norm": 1.1974716254929318,
      "learning_rate": 9.869614952850454e-06,
      "loss": 0.39612202644348143,
      "num_input_tokens_seen": 35452104,
      "step": 2165,
      "train_runtime": 4166.8117,
      "train_tokens_per_second": 8508.209
    },
    {
      "epoch": 0.14610826824670078,
      "grad_norm": 1.2728768108752517,
      "learning_rate": 9.86901438569769e-06,
      "loss": 0.46419663429260255,
      "num_input_tokens_seen": 35534024,
      "step": 2170,
      "train_runtime": 4176.388,
      "train_tokens_per_second": 8508.315
    },
    {
      "epoch": 0.14644492324266092,
      "grad_norm": 1.2945357113700868,
      "learning_rate": 9.86841245694058e-06,
      "loss": 0.40476813316345217,
      "num_input_tokens_seen": 35615944,
      "step": 2175,
      "train_runtime": 4185.9389,
      "train_tokens_per_second": 8508.472
    },
    {
      "epoch": 0.14678157823862106,
      "grad_norm": 1.325310686649326,
      "learning_rate": 9.86780916674745e-06,
      "loss": 0.47794480323791505,
      "num_input_tokens_seen": 35697864,
      "step": 2180,
      "train_runtime": 4195.5197,
      "train_tokens_per_second": 8508.568
    },
    {
      "epoch": 0.1471182332345812,
      "grad_norm": 1.2459990278395265,
      "learning_rate": 9.86720451528701e-06,
      "loss": 0.447496223449707,
      "num_input_tokens_seen": 35778944,
      "step": 2185,
      "train_runtime": 4205.2499,
      "train_tokens_per_second": 8508.161
    },
    {
      "epoch": 0.14745488823054134,
      "grad_norm": 1.3680223798301994,
      "learning_rate": 9.866598502728348e-06,
      "loss": 0.45435943603515627,
      "num_input_tokens_seen": 35860864,
      "step": 2190,
      "train_runtime": 4214.8227,
      "train_tokens_per_second": 8508.273
    },
    {
      "epoch": 0.14779154322650148,
      "grad_norm": 1.2692527585652014,
      "learning_rate": 9.865991129240931e-06,
      "loss": 0.4777120590209961,
      "num_input_tokens_seen": 35942784,
      "step": 2195,
      "train_runtime": 4224.3747,
      "train_tokens_per_second": 8508.427
    },
    {
      "epoch": 0.14812819822246162,
      "grad_norm": 1.0842206094937676,
      "learning_rate": 9.865382394994614e-06,
      "loss": 0.3857145309448242,
      "num_input_tokens_seen": 36024704,
      "step": 2200,
      "train_runtime": 4233.929,
      "train_tokens_per_second": 8508.575
    },
    {
      "epoch": 0.14846485321842176,
      "grad_norm": 1.2777998889728877,
      "learning_rate": 9.864772300159622e-06,
      "loss": 0.4355044364929199,
      "num_input_tokens_seen": 36106624,
      "step": 2205,
      "train_runtime": 4243.539,
      "train_tokens_per_second": 8508.611
    },
    {
      "epoch": 0.1488015082143819,
      "grad_norm": 1.4300545685784418,
      "learning_rate": 9.864160844906571e-06,
      "loss": 0.4512486934661865,
      "num_input_tokens_seen": 36188544,
      "step": 2210,
      "train_runtime": 4253.1728,
      "train_tokens_per_second": 8508.6
    },
    {
      "epoch": 0.14913816321034204,
      "grad_norm": 1.2284695559930918,
      "learning_rate": 9.863548029406449e-06,
      "loss": 0.4123145580291748,
      "num_input_tokens_seen": 36270464,
      "step": 2215,
      "train_runtime": 4262.6774,
      "train_tokens_per_second": 8508.846
    },
    {
      "epoch": 0.14947481820630218,
      "grad_norm": 1.1946671936551052,
      "learning_rate": 9.86293385383063e-06,
      "loss": 0.42649059295654296,
      "num_input_tokens_seen": 36352384,
      "step": 2220,
      "train_runtime": 4272.3156,
      "train_tokens_per_second": 8508.825
    },
    {
      "epoch": 0.14981147320226232,
      "grad_norm": 1.282144665605806,
      "learning_rate": 9.862318318350865e-06,
      "loss": 0.46100783348083496,
      "num_input_tokens_seen": 36434304,
      "step": 2225,
      "train_runtime": 4281.9308,
      "train_tokens_per_second": 8508.849
    },
    {
      "epoch": 0.15014812819822246,
      "grad_norm": 1.203127417116651,
      "learning_rate": 9.861701423139287e-06,
      "loss": 0.4041304588317871,
      "num_input_tokens_seen": 36516224,
      "step": 2230,
      "train_runtime": 4291.5372,
      "train_tokens_per_second": 8508.892
    },
    {
      "epoch": 0.1504847831941826,
      "grad_norm": 1.1840261332231918,
      "learning_rate": 9.861083168368411e-06,
      "loss": 0.447785758972168,
      "num_input_tokens_seen": 36598144,
      "step": 2235,
      "train_runtime": 4301.0458,
      "train_tokens_per_second": 8509.127
    },
    {
      "epoch": 0.15082143819014274,
      "grad_norm": 1.357231171413884,
      "learning_rate": 9.860463554211125e-06,
      "loss": 0.4030327796936035,
      "num_input_tokens_seen": 36680064,
      "step": 2240,
      "train_runtime": 4310.6785,
      "train_tokens_per_second": 8509.116
    },
    {
      "epoch": 0.15115809318610288,
      "grad_norm": 1.3181136958853408,
      "learning_rate": 9.859842580840706e-06,
      "loss": 0.4936062335968018,
      "num_input_tokens_seen": 36761984,
      "step": 2245,
      "train_runtime": 4320.2836,
      "train_tokens_per_second": 8509.16
    },
    {
      "epoch": 0.15149474818206302,
      "grad_norm": 1.2177469436646586,
      "learning_rate": 9.859220248430806e-06,
      "loss": 0.447477912902832,
      "num_input_tokens_seen": 36843904,
      "step": 2250,
      "train_runtime": 4329.9042,
      "train_tokens_per_second": 8509.173
    },
    {
      "epoch": 0.15183140317802316,
      "grad_norm": 1.0683272355717086,
      "learning_rate": 9.85859655715546e-06,
      "loss": 0.4246512889862061,
      "num_input_tokens_seen": 36925824,
      "step": 2255,
      "train_runtime": 4339.4078,
      "train_tokens_per_second": 8509.416
    },
    {
      "epoch": 0.1521680581739833,
      "grad_norm": 1.4187399020532894,
      "learning_rate": 9.857971507189077e-06,
      "loss": 0.45186443328857423,
      "num_input_tokens_seen": 37007744,
      "step": 2260,
      "train_runtime": 4349.0165,
      "train_tokens_per_second": 8509.451
    },
    {
      "epoch": 0.15250471316994343,
      "grad_norm": 1.298548579443837,
      "learning_rate": 9.857345098706455e-06,
      "loss": 0.42166843414306643,
      "num_input_tokens_seen": 37089664,
      "step": 2265,
      "train_runtime": 4358.6239,
      "train_tokens_per_second": 8509.489
    },
    {
      "epoch": 0.15284136816590357,
      "grad_norm": 1.2301650171734946,
      "learning_rate": 9.856717331882765e-06,
      "loss": 0.45794153213500977,
      "num_input_tokens_seen": 37171584,
      "step": 2270,
      "train_runtime": 4368.1443,
      "train_tokens_per_second": 8509.697
    },
    {
      "epoch": 0.1531780231618637,
      "grad_norm": 1.1395525414733962,
      "learning_rate": 9.85608820689356e-06,
      "loss": 0.4052178382873535,
      "num_input_tokens_seen": 37253504,
      "step": 2275,
      "train_runtime": 4377.7223,
      "train_tokens_per_second": 8509.791
    },
    {
      "epoch": 0.15351467815782385,
      "grad_norm": 1.2271956628445695,
      "learning_rate": 9.855457723914772e-06,
      "loss": 0.45613574981689453,
      "num_input_tokens_seen": 37335120,
      "step": 2280,
      "train_runtime": 4387.7707,
      "train_tokens_per_second": 8508.904
    },
    {
      "epoch": 0.153851333153784,
      "grad_norm": 1.344979872074416,
      "learning_rate": 9.854825883122713e-06,
      "loss": 0.4232954978942871,
      "num_input_tokens_seen": 37417040,
      "step": 2285,
      "train_runtime": 4397.375,
      "train_tokens_per_second": 8508.949
    },
    {
      "epoch": 0.15418798814974413,
      "grad_norm": 1.0787049269740172,
      "learning_rate": 9.85419268469408e-06,
      "loss": 0.38846962451934813,
      "num_input_tokens_seen": 37498960,
      "step": 2290,
      "train_runtime": 4406.9358,
      "train_tokens_per_second": 8509.078
    },
    {
      "epoch": 0.15452464314570427,
      "grad_norm": 1.133241598349482,
      "learning_rate": 9.85355812880594e-06,
      "loss": 0.4081723690032959,
      "num_input_tokens_seen": 37580880,
      "step": 2295,
      "train_runtime": 4416.5034,
      "train_tokens_per_second": 8509.193
    },
    {
      "epoch": 0.1548612981416644,
      "grad_norm": 1.134407965706332,
      "learning_rate": 9.852922215635747e-06,
      "loss": 0.4050453186035156,
      "num_input_tokens_seen": 37662800,
      "step": 2300,
      "train_runtime": 4426.1453,
      "train_tokens_per_second": 8509.165
    },
    {
      "epoch": 0.15519795313762455,
      "grad_norm": 1.2460374710129467,
      "learning_rate": 9.85228494536133e-06,
      "loss": 0.4566253662109375,
      "num_input_tokens_seen": 37744720,
      "step": 2305,
      "train_runtime": 4435.7586,
      "train_tokens_per_second": 8509.192
    },
    {
      "epoch": 0.1555346081335847,
      "grad_norm": 1.2227658752507506,
      "learning_rate": 9.851646318160901e-06,
      "loss": 0.4358521938323975,
      "num_input_tokens_seen": 37826640,
      "step": 2310,
      "train_runtime": 4445.3108,
      "train_tokens_per_second": 8509.335
    },
    {
      "epoch": 0.15587126312954483,
      "grad_norm": 1.1492273099269594,
      "learning_rate": 9.85100633421305e-06,
      "loss": 0.39329819679260253,
      "num_input_tokens_seen": 37908560,
      "step": 2315,
      "train_runtime": 4454.8909,
      "train_tokens_per_second": 8509.425
    },
    {
      "epoch": 0.15620791812550497,
      "grad_norm": 1.3277530143367378,
      "learning_rate": 9.850364993696749e-06,
      "loss": 0.4558074951171875,
      "num_input_tokens_seen": 37990480,
      "step": 2320,
      "train_runtime": 4464.4483,
      "train_tokens_per_second": 8509.558
    },
    {
      "epoch": 0.15654457312146514,
      "grad_norm": 1.11492482955661,
      "learning_rate": 9.849722296791339e-06,
      "loss": 0.46094822883605957,
      "num_input_tokens_seen": 38072400,
      "step": 2325,
      "train_runtime": 4474.0628,
      "train_tokens_per_second": 8509.581
    },
    {
      "epoch": 0.15688122811742528,
      "grad_norm": 1.3283514356665052,
      "learning_rate": 9.849078243676558e-06,
      "loss": 0.47058820724487305,
      "num_input_tokens_seen": 38154320,
      "step": 2330,
      "train_runtime": 4483.5799,
      "train_tokens_per_second": 8509.789
    },
    {
      "epoch": 0.15721788311338541,
      "grad_norm": 1.2897901782650403,
      "learning_rate": 9.848432834532506e-06,
      "loss": 0.39372668266296384,
      "num_input_tokens_seen": 38236240,
      "step": 2335,
      "train_runtime": 4493.2195,
      "train_tokens_per_second": 8509.765
    },
    {
      "epoch": 0.15755453810934555,
      "grad_norm": 1.1782805110513668,
      "learning_rate": 9.847786069539673e-06,
      "loss": 0.463491153717041,
      "num_input_tokens_seen": 38318160,
      "step": 2340,
      "train_runtime": 4502.8036,
      "train_tokens_per_second": 8509.845
    },
    {
      "epoch": 0.1578911931053057,
      "grad_norm": 1.3508970882526572,
      "learning_rate": 9.847137948878926e-06,
      "loss": 0.45805559158325193,
      "num_input_tokens_seen": 38400080,
      "step": 2345,
      "train_runtime": 4512.3064,
      "train_tokens_per_second": 8510.078
    },
    {
      "epoch": 0.15822784810126583,
      "grad_norm": 1.2039576754238066,
      "learning_rate": 9.846488472731505e-06,
      "loss": 0.4938191413879395,
      "num_input_tokens_seen": 38482000,
      "step": 2350,
      "train_runtime": 4521.9237,
      "train_tokens_per_second": 8510.095
    },
    {
      "epoch": 0.15856450309722597,
      "grad_norm": 1.1706257640328175,
      "learning_rate": 9.845837641279038e-06,
      "loss": 0.4420405387878418,
      "num_input_tokens_seen": 38563920,
      "step": 2355,
      "train_runtime": 4531.5511,
      "train_tokens_per_second": 8510.093
    },
    {
      "epoch": 0.1589011580931861,
      "grad_norm": 1.3404048837397826,
      "learning_rate": 9.845185454703528e-06,
      "loss": 0.4587400436401367,
      "num_input_tokens_seen": 38645840,
      "step": 2360,
      "train_runtime": 4541.0577,
      "train_tokens_per_second": 8510.317
    },
    {
      "epoch": 0.15923781308914625,
      "grad_norm": 1.100776973782913,
      "learning_rate": 9.844531913187356e-06,
      "loss": 0.4192973136901855,
      "num_input_tokens_seen": 38727760,
      "step": 2365,
      "train_runtime": 4550.6084,
      "train_tokens_per_second": 8510.458
    },
    {
      "epoch": 0.1595744680851064,
      "grad_norm": 1.4651285697806824,
      "learning_rate": 9.84387701691328e-06,
      "loss": 0.4674382209777832,
      "num_input_tokens_seen": 38809680,
      "step": 2370,
      "train_runtime": 4560.2451,
      "train_tokens_per_second": 8510.437
    },
    {
      "epoch": 0.15991112308106653,
      "grad_norm": 1.2587283867286558,
      "learning_rate": 9.843220766064443e-06,
      "loss": 0.4511577606201172,
      "num_input_tokens_seen": 38891600,
      "step": 2375,
      "train_runtime": 4569.8017,
      "train_tokens_per_second": 8510.566
    },
    {
      "epoch": 0.16024777807702667,
      "grad_norm": 1.2292667999700344,
      "learning_rate": 9.842563160824364e-06,
      "loss": 0.44477477073669436,
      "num_input_tokens_seen": 38973520,
      "step": 2380,
      "train_runtime": 4579.3801,
      "train_tokens_per_second": 8510.654
    },
    {
      "epoch": 0.1605844330729868,
      "grad_norm": 1.3210101641206013,
      "learning_rate": 9.841904201376939e-06,
      "loss": 0.5092373371124268,
      "num_input_tokens_seen": 39055440,
      "step": 2385,
      "train_runtime": 4588.9826,
      "train_tokens_per_second": 8510.697
    },
    {
      "epoch": 0.16092108806894695,
      "grad_norm": 1.3216635809625359,
      "learning_rate": 9.841243887906443e-06,
      "loss": 0.4082357406616211,
      "num_input_tokens_seen": 39137360,
      "step": 2390,
      "train_runtime": 4598.6054,
      "train_tokens_per_second": 8510.702
    },
    {
      "epoch": 0.1612577430649071,
      "grad_norm": 1.177625385024611,
      "learning_rate": 9.840582220597532e-06,
      "loss": 0.4212993621826172,
      "num_input_tokens_seen": 39219280,
      "step": 2395,
      "train_runtime": 4608.1149,
      "train_tokens_per_second": 8510.916
    },
    {
      "epoch": 0.16159439806086723,
      "grad_norm": 1.2829066107509453,
      "learning_rate": 9.839919199635238e-06,
      "loss": 0.42782058715820315,
      "num_input_tokens_seen": 39301200,
      "step": 2400,
      "train_runtime": 4617.7227,
      "train_tokens_per_second": 8510.949
    },
    {
      "epoch": 0.16193105305682737,
      "grad_norm": 1.2273785418229555,
      "learning_rate": 9.839254825204973e-06,
      "loss": 0.44123020172119143,
      "num_input_tokens_seen": 39383120,
      "step": 2405,
      "train_runtime": 4627.334,
      "train_tokens_per_second": 8510.974
    },
    {
      "epoch": 0.1622677080527875,
      "grad_norm": 1.1590321288650407,
      "learning_rate": 9.838589097492527e-06,
      "loss": 0.4352745056152344,
      "num_input_tokens_seen": 39465040,
      "step": 2410,
      "train_runtime": 4636.8367,
      "train_tokens_per_second": 8511.199
    },
    {
      "epoch": 0.16260436304874765,
      "grad_norm": 1.3708962597282897,
      "learning_rate": 9.83792201668407e-06,
      "loss": 0.4656620979309082,
      "num_input_tokens_seen": 39546960,
      "step": 2415,
      "train_runtime": 4646.4613,
      "train_tokens_per_second": 8511.2
    },
    {
      "epoch": 0.16294101804470779,
      "grad_norm": 1.481342370384807,
      "learning_rate": 9.837253582966146e-06,
      "loss": 0.4566462993621826,
      "num_input_tokens_seen": 39628880,
      "step": 2420,
      "train_runtime": 4656.0884,
      "train_tokens_per_second": 8511.196
    },
    {
      "epoch": 0.16327767304066793,
      "grad_norm": 1.1506910088581723,
      "learning_rate": 9.836583796525686e-06,
      "loss": 0.37556896209716795,
      "num_input_tokens_seen": 39710800,
      "step": 2425,
      "train_runtime": 4665.6007,
      "train_tokens_per_second": 8511.401
    },
    {
      "epoch": 0.16361432803662807,
      "grad_norm": 1.2530599821116357,
      "learning_rate": 9.835912657549986e-06,
      "loss": 0.4485467433929443,
      "num_input_tokens_seen": 39792720,
      "step": 2430,
      "train_runtime": 4675.2427,
      "train_tokens_per_second": 8511.37
    },
    {
      "epoch": 0.1639509830325882,
      "grad_norm": 1.266080977654893,
      "learning_rate": 9.835240166226734e-06,
      "loss": 0.434102725982666,
      "num_input_tokens_seen": 39874640,
      "step": 2435,
      "train_runtime": 4684.85,
      "train_tokens_per_second": 8511.402
    },
    {
      "epoch": 0.16428763802854834,
      "grad_norm": 1.2149921089876596,
      "learning_rate": 9.834566322743987e-06,
      "loss": 0.42055397033691405,
      "num_input_tokens_seen": 39956560,
      "step": 2440,
      "train_runtime": 4694.4029,
      "train_tokens_per_second": 8511.532
    },
    {
      "epoch": 0.16462429302450848,
      "grad_norm": 1.227432048776126,
      "learning_rate": 9.833891127290186e-06,
      "loss": 0.45360794067382815,
      "num_input_tokens_seen": 40038480,
      "step": 2445,
      "train_runtime": 4703.9733,
      "train_tokens_per_second": 8511.63
    },
    {
      "epoch": 0.16496094802046862,
      "grad_norm": 1.1206783155916056,
      "learning_rate": 9.833214580054145e-06,
      "loss": 0.44536361694335935,
      "num_input_tokens_seen": 40120400,
      "step": 2450,
      "train_runtime": 4713.5813,
      "train_tokens_per_second": 8511.66
    },
    {
      "epoch": 0.16529760301642876,
      "grad_norm": 1.3610280497207947,
      "learning_rate": 9.832536681225058e-06,
      "loss": 0.4762770175933838,
      "num_input_tokens_seen": 40202320,
      "step": 2455,
      "train_runtime": 4723.1438,
      "train_tokens_per_second": 8511.771
    },
    {
      "epoch": 0.1656342580123889,
      "grad_norm": 1.382405950806897,
      "learning_rate": 9.831857430992497e-06,
      "loss": 0.4204257488250732,
      "num_input_tokens_seen": 40284240,
      "step": 2460,
      "train_runtime": 4732.7067,
      "train_tokens_per_second": 8511.882
    },
    {
      "epoch": 0.16597091300834904,
      "grad_norm": 1.316822353934357,
      "learning_rate": 9.831176829546416e-06,
      "loss": 0.4499103546142578,
      "num_input_tokens_seen": 40366160,
      "step": 2465,
      "train_runtime": 4742.3173,
      "train_tokens_per_second": 8511.906
    },
    {
      "epoch": 0.16630756800430918,
      "grad_norm": 1.5435626173022292,
      "learning_rate": 9.830494877077137e-06,
      "loss": 0.5296729564666748,
      "num_input_tokens_seen": 40448080,
      "step": 2470,
      "train_runtime": 4751.9201,
      "train_tokens_per_second": 8511.945
    },
    {
      "epoch": 0.16664422300026932,
      "grad_norm": 1.2286228590831079,
      "learning_rate": 9.82981157377537e-06,
      "loss": 0.4002420425415039,
      "num_input_tokens_seen": 40530000,
      "step": 2475,
      "train_runtime": 4761.4284,
      "train_tokens_per_second": 8512.152
    },
    {
      "epoch": 0.16698087799622946,
      "grad_norm": 1.2125873830283265,
      "learning_rate": 9.829126919832198e-06,
      "loss": 0.3801423072814941,
      "num_input_tokens_seen": 40611920,
      "step": 2480,
      "train_runtime": 4771.0555,
      "train_tokens_per_second": 8512.146
    },
    {
      "epoch": 0.1673175329921896,
      "grad_norm": 1.1743035942171176,
      "learning_rate": 9.82844091543908e-06,
      "loss": 0.4198304176330566,
      "num_input_tokens_seen": 40693840,
      "step": 2485,
      "train_runtime": 4780.6664,
      "train_tokens_per_second": 8512.169
    },
    {
      "epoch": 0.16765418798814974,
      "grad_norm": 1.2573285682446638,
      "learning_rate": 9.82775356078786e-06,
      "loss": 0.4261636257171631,
      "num_input_tokens_seen": 40775760,
      "step": 2490,
      "train_runtime": 4790.1726,
      "train_tokens_per_second": 8512.378
    },
    {
      "epoch": 0.16799084298410988,
      "grad_norm": 1.3401507933805883,
      "learning_rate": 9.82706485607075e-06,
      "loss": 0.41817531585693357,
      "num_input_tokens_seen": 40857680,
      "step": 2495,
      "train_runtime": 4799.6926,
      "train_tokens_per_second": 8512.562
    },
    {
      "epoch": 0.16832749798007002,
      "grad_norm": 1.2373187152035974,
      "learning_rate": 9.826374801480346e-06,
      "loss": 0.43210697174072266,
      "num_input_tokens_seen": 40939600,
      "step": 2500,
      "train_runtime": 4809.3035,
      "train_tokens_per_second": 8512.584
    },
    {
      "epoch": 0.16866415297603016,
      "grad_norm": 1.1770824234997086,
      "learning_rate": 9.825683397209617e-06,
      "loss": 0.45387907028198243,
      "num_input_tokens_seen": 41021520,
      "step": 2505,
      "train_runtime": 4818.8035,
      "train_tokens_per_second": 8512.802
    },
    {
      "epoch": 0.1690008079719903,
      "grad_norm": 1.1926698331522856,
      "learning_rate": 9.824990643451915e-06,
      "loss": 0.4275675773620605,
      "num_input_tokens_seen": 41103440,
      "step": 2510,
      "train_runtime": 4828.4109,
      "train_tokens_per_second": 8512.83
    },
    {
      "epoch": 0.16933746296795044,
      "grad_norm": 1.2192437883255562,
      "learning_rate": 9.824296540400965e-06,
      "loss": 0.42984352111816404,
      "num_input_tokens_seen": 41185360,
      "step": 2515,
      "train_runtime": 4838.0202,
      "train_tokens_per_second": 8512.854
    },
    {
      "epoch": 0.16967411796391058,
      "grad_norm": 1.470482550999968,
      "learning_rate": 9.823601088250872e-06,
      "loss": 0.472247314453125,
      "num_input_tokens_seen": 41267280,
      "step": 2520,
      "train_runtime": 4847.518,
      "train_tokens_per_second": 8513.074
    },
    {
      "epoch": 0.17001077295987072,
      "grad_norm": 1.231048002965743,
      "learning_rate": 9.822904287196114e-06,
      "loss": 0.4486112594604492,
      "num_input_tokens_seen": 41349200,
      "step": 2525,
      "train_runtime": 4857.1377,
      "train_tokens_per_second": 8513.08
    },
    {
      "epoch": 0.17034742795583085,
      "grad_norm": 0.9857143459707178,
      "learning_rate": 9.822206137431553e-06,
      "loss": 0.4124735355377197,
      "num_input_tokens_seen": 41431120,
      "step": 2530,
      "train_runtime": 4866.7415,
      "train_tokens_per_second": 8513.113
    },
    {
      "epoch": 0.170684082951791,
      "grad_norm": 1.095964999078951,
      "learning_rate": 9.821506639152421e-06,
      "loss": 0.4178347587585449,
      "num_input_tokens_seen": 41513040,
      "step": 2535,
      "train_runtime": 4876.2949,
      "train_tokens_per_second": 8513.234
    },
    {
      "epoch": 0.17102073794775113,
      "grad_norm": 1.1817378696460301,
      "learning_rate": 9.820805792554334e-06,
      "loss": 0.4763012886047363,
      "num_input_tokens_seen": 41594960,
      "step": 2540,
      "train_runtime": 4885.8613,
      "train_tokens_per_second": 8513.332
    },
    {
      "epoch": 0.17135739294371127,
      "grad_norm": 1.381496930351716,
      "learning_rate": 9.820103597833278e-06,
      "loss": 0.4774531364440918,
      "num_input_tokens_seen": 41676448,
      "step": 2545,
      "train_runtime": 4895.8333,
      "train_tokens_per_second": 8512.636
    },
    {
      "epoch": 0.1716940479396714,
      "grad_norm": 1.1577894625636282,
      "learning_rate": 9.819400055185622e-06,
      "loss": 0.41477465629577637,
      "num_input_tokens_seen": 41758368,
      "step": 2550,
      "train_runtime": 4905.3992,
      "train_tokens_per_second": 8512.736
    },
    {
      "epoch": 0.17203070293563155,
      "grad_norm": 1.3519638410001682,
      "learning_rate": 9.818695164808108e-06,
      "loss": 0.45023221969604493,
      "num_input_tokens_seen": 41840288,
      "step": 2555,
      "train_runtime": 4914.8987,
      "train_tokens_per_second": 8512.95
    },
    {
      "epoch": 0.1723673579315917,
      "grad_norm": 1.0304196277911055,
      "learning_rate": 9.817988926897856e-06,
      "loss": 0.39937362670898435,
      "num_input_tokens_seen": 41922208,
      "step": 2560,
      "train_runtime": 4924.5051,
      "train_tokens_per_second": 8512.979
    },
    {
      "epoch": 0.17270401292755186,
      "grad_norm": 1.2075711823356583,
      "learning_rate": 9.817281341652367e-06,
      "loss": 0.43855600357055663,
      "num_input_tokens_seen": 42004128,
      "step": 2565,
      "train_runtime": 4934.1217,
      "train_tokens_per_second": 8512.99
    },
    {
      "epoch": 0.173040667923512,
      "grad_norm": 1.183183814999042,
      "learning_rate": 9.816572409269508e-06,
      "loss": 0.3899224042892456,
      "num_input_tokens_seen": 42086048,
      "step": 2570,
      "train_runtime": 4943.6311,
      "train_tokens_per_second": 8513.185
    },
    {
      "epoch": 0.17337732291947214,
      "grad_norm": 1.2008647587815244,
      "learning_rate": 9.815862129947537e-06,
      "loss": 0.4666709899902344,
      "num_input_tokens_seen": 42167968,
      "step": 2575,
      "train_runtime": 4953.2527,
      "train_tokens_per_second": 8513.187
    },
    {
      "epoch": 0.17371397791543228,
      "grad_norm": 1.1794307573753917,
      "learning_rate": 9.815150503885078e-06,
      "loss": 0.37673723697662354,
      "num_input_tokens_seen": 42249888,
      "step": 2580,
      "train_runtime": 4962.861,
      "train_tokens_per_second": 8513.212
    },
    {
      "epoch": 0.17405063291139242,
      "grad_norm": 1.2637215142629505,
      "learning_rate": 9.814437531281133e-06,
      "loss": 0.4403403282165527,
      "num_input_tokens_seen": 42331808,
      "step": 2585,
      "train_runtime": 4972.3603,
      "train_tokens_per_second": 8513.423
    },
    {
      "epoch": 0.17438728790735256,
      "grad_norm": 1.5058663028493484,
      "learning_rate": 9.813723212335085e-06,
      "loss": 0.47238998413085936,
      "num_input_tokens_seen": 42413728,
      "step": 2590,
      "train_runtime": 4981.9807,
      "train_tokens_per_second": 8513.427
    },
    {
      "epoch": 0.1747239429033127,
      "grad_norm": 1.1165399248496928,
      "learning_rate": 9.813007547246691e-06,
      "loss": 0.39557251930236814,
      "num_input_tokens_seen": 42495648,
      "step": 2595,
      "train_runtime": 4991.4976,
      "train_tokens_per_second": 8513.607
    },
    {
      "epoch": 0.17506059789927284,
      "grad_norm": 1.162821354291643,
      "learning_rate": 9.812290536216083e-06,
      "loss": 0.48587923049926757,
      "num_input_tokens_seen": 42576864,
      "step": 2600,
      "train_runtime": 5001.1299,
      "train_tokens_per_second": 8513.449
    },
    {
      "epoch": 0.17539725289523297,
      "grad_norm": 1.1897803987452376,
      "learning_rate": 9.811572179443773e-06,
      "loss": 0.46026015281677246,
      "num_input_tokens_seen": 42658784,
      "step": 2605,
      "train_runtime": 5010.6341,
      "train_tokens_per_second": 8513.65
    },
    {
      "epoch": 0.17573390789119311,
      "grad_norm": 1.1594893049041723,
      "learning_rate": 9.810852477130645e-06,
      "loss": 0.42764968872070314,
      "num_input_tokens_seen": 42740704,
      "step": 2610,
      "train_runtime": 5020.216,
      "train_tokens_per_second": 8513.718
    },
    {
      "epoch": 0.17607056288715325,
      "grad_norm": 1.2764230552064737,
      "learning_rate": 9.810131429477964e-06,
      "loss": 0.42919187545776366,
      "num_input_tokens_seen": 42822624,
      "step": 2615,
      "train_runtime": 5029.8172,
      "train_tokens_per_second": 8513.754
    },
    {
      "epoch": 0.1764072178831134,
      "grad_norm": 1.1209071610096877,
      "learning_rate": 9.809409036687364e-06,
      "loss": 0.4136063575744629,
      "num_input_tokens_seen": 42904544,
      "step": 2620,
      "train_runtime": 5039.4146,
      "train_tokens_per_second": 8513.795
    },
    {
      "epoch": 0.17674387287907353,
      "grad_norm": 1.556636242637848,
      "learning_rate": 9.808685298960867e-06,
      "loss": 0.4246169090270996,
      "num_input_tokens_seen": 42986464,
      "step": 2625,
      "train_runtime": 5048.909,
      "train_tokens_per_second": 8514.01
    },
    {
      "epoch": 0.17708052787503367,
      "grad_norm": 1.131356111253497,
      "learning_rate": 9.807960216500858e-06,
      "loss": 0.4424586296081543,
      "num_input_tokens_seen": 43068384,
      "step": 2630,
      "train_runtime": 5058.5247,
      "train_tokens_per_second": 8514.021
    },
    {
      "epoch": 0.1774171828709938,
      "grad_norm": 1.1809109210838071,
      "learning_rate": 9.807233789510106e-06,
      "loss": 0.3941640377044678,
      "num_input_tokens_seen": 43150304,
      "step": 2635,
      "train_runtime": 5068.1472,
      "train_tokens_per_second": 8514.02
    },
    {
      "epoch": 0.17775383786695395,
      "grad_norm": 1.3792348999702801,
      "learning_rate": 9.806506018191754e-06,
      "loss": 0.4119282245635986,
      "num_input_tokens_seen": 43232224,
      "step": 2640,
      "train_runtime": 5077.7232,
      "train_tokens_per_second": 8514.096
    },
    {
      "epoch": 0.1780904928629141,
      "grad_norm": 1.4335741666152604,
      "learning_rate": 9.80577690274932e-06,
      "loss": 0.44257707595825196,
      "num_input_tokens_seen": 43314144,
      "step": 2645,
      "train_runtime": 5087.2385,
      "train_tokens_per_second": 8514.274
    },
    {
      "epoch": 0.17842714785887423,
      "grad_norm": 1.0664596515258016,
      "learning_rate": 9.8050464433867e-06,
      "loss": 0.3888126850128174,
      "num_input_tokens_seen": 43396064,
      "step": 2650,
      "train_runtime": 5096.8516,
      "train_tokens_per_second": 8514.288
    },
    {
      "epoch": 0.17876380285483437,
      "grad_norm": 1.227696386869333,
      "learning_rate": 9.804314640308165e-06,
      "loss": 0.4386421203613281,
      "num_input_tokens_seen": 43477984,
      "step": 2655,
      "train_runtime": 5106.4961,
      "train_tokens_per_second": 8514.25
    },
    {
      "epoch": 0.1791004578507945,
      "grad_norm": 1.3773857197295982,
      "learning_rate": 9.803581493718361e-06,
      "loss": 0.45714893341064455,
      "num_input_tokens_seen": 43559904,
      "step": 2660,
      "train_runtime": 5115.9995,
      "train_tokens_per_second": 8514.446
    },
    {
      "epoch": 0.17943711284675465,
      "grad_norm": 1.1319883945706533,
      "learning_rate": 9.80284700382231e-06,
      "loss": 0.4239362716674805,
      "num_input_tokens_seen": 43641824,
      "step": 2665,
      "train_runtime": 5125.6136,
      "train_tokens_per_second": 8514.458
    },
    {
      "epoch": 0.1797737678427148,
      "grad_norm": 1.206446465874487,
      "learning_rate": 9.802111170825408e-06,
      "loss": 0.4650732517242432,
      "num_input_tokens_seen": 43723744,
      "step": 2670,
      "train_runtime": 5135.2412,
      "train_tokens_per_second": 8514.448
    },
    {
      "epoch": 0.18011042283867493,
      "grad_norm": 1.2706599872018713,
      "learning_rate": 9.80137399493343e-06,
      "loss": 0.43891334533691406,
      "num_input_tokens_seen": 43805664,
      "step": 2675,
      "train_runtime": 5144.7434,
      "train_tokens_per_second": 8514.645
    },
    {
      "epoch": 0.18044707783463507,
      "grad_norm": 1.2364553750783385,
      "learning_rate": 9.800635476352525e-06,
      "loss": 0.40817766189575194,
      "num_input_tokens_seen": 43887584,
      "step": 2680,
      "train_runtime": 5154.344,
      "train_tokens_per_second": 8514.679
    },
    {
      "epoch": 0.1807837328305952,
      "grad_norm": 1.2416119352962565,
      "learning_rate": 9.799895615289218e-06,
      "loss": 0.38972060680389403,
      "num_input_tokens_seen": 43969504,
      "step": 2685,
      "train_runtime": 5163.9648,
      "train_tokens_per_second": 8514.679
    },
    {
      "epoch": 0.18112038782655535,
      "grad_norm": 1.2991849077197728,
      "learning_rate": 9.799154411950407e-06,
      "loss": 0.4873291015625,
      "num_input_tokens_seen": 44051424,
      "step": 2690,
      "train_runtime": 5173.4739,
      "train_tokens_per_second": 8514.863
    },
    {
      "epoch": 0.18145704282251549,
      "grad_norm": 1.0754506218776412,
      "learning_rate": 9.79841186654337e-06,
      "loss": 0.4408556938171387,
      "num_input_tokens_seen": 44133344,
      "step": 2695,
      "train_runtime": 5183.0904,
      "train_tokens_per_second": 8514.871
    },
    {
      "epoch": 0.18179369781847562,
      "grad_norm": 1.3003360527788634,
      "learning_rate": 9.797667979275752e-06,
      "loss": 0.47945423126220704,
      "num_input_tokens_seen": 44214936,
      "step": 2700,
      "train_runtime": 5193.1054,
      "train_tokens_per_second": 8514.161
    },
    {
      "epoch": 0.18213035281443576,
      "grad_norm": 1.1101403706018098,
      "learning_rate": 9.796922750355584e-06,
      "loss": 0.40613536834716796,
      "num_input_tokens_seen": 44296856,
      "step": 2705,
      "train_runtime": 5202.6101,
      "train_tokens_per_second": 8514.352
    },
    {
      "epoch": 0.1824670078103959,
      "grad_norm": 1.435647587897159,
      "learning_rate": 9.796176179991266e-06,
      "loss": 0.4162907123565674,
      "num_input_tokens_seen": 44378776,
      "step": 2710,
      "train_runtime": 5212.1439,
      "train_tokens_per_second": 8514.496
    },
    {
      "epoch": 0.18280366280635604,
      "grad_norm": 1.1736465963743166,
      "learning_rate": 9.795428268391572e-06,
      "loss": 0.42853264808654784,
      "num_input_tokens_seen": 44460696,
      "step": 2715,
      "train_runtime": 5221.7527,
      "train_tokens_per_second": 8514.516
    },
    {
      "epoch": 0.18314031780231618,
      "grad_norm": 1.0639811364002232,
      "learning_rate": 9.794679015765655e-06,
      "loss": 0.38796334266662597,
      "num_input_tokens_seen": 44542616,
      "step": 2720,
      "train_runtime": 5231.2555,
      "train_tokens_per_second": 8514.709
    },
    {
      "epoch": 0.18347697279827632,
      "grad_norm": 1.342352070801443,
      "learning_rate": 9.793928422323041e-06,
      "loss": 0.43848066329956054,
      "num_input_tokens_seen": 44624536,
      "step": 2725,
      "train_runtime": 5240.8614,
      "train_tokens_per_second": 8514.733
    },
    {
      "epoch": 0.18381362779423646,
      "grad_norm": 1.14444632118257,
      "learning_rate": 9.793176488273629e-06,
      "loss": 0.44410295486450196,
      "num_input_tokens_seen": 44706080,
      "step": 2730,
      "train_runtime": 5250.8556,
      "train_tokens_per_second": 8514.056
    },
    {
      "epoch": 0.1841502827901966,
      "grad_norm": 1.1527956415797906,
      "learning_rate": 9.792423213827697e-06,
      "loss": 0.4207311630249023,
      "num_input_tokens_seen": 44788000,
      "step": 2735,
      "train_runtime": 5260.4989,
      "train_tokens_per_second": 8514.021
    },
    {
      "epoch": 0.18448693778615674,
      "grad_norm": 1.1530829672154872,
      "learning_rate": 9.791668599195897e-06,
      "loss": 0.4224213123321533,
      "num_input_tokens_seen": 44869920,
      "step": 2740,
      "train_runtime": 5270.0341,
      "train_tokens_per_second": 8514.161
    },
    {
      "epoch": 0.18482359278211688,
      "grad_norm": 1.1904905762632665,
      "learning_rate": 9.790912644589248e-06,
      "loss": 0.4668020248413086,
      "num_input_tokens_seen": 44951840,
      "step": 2745,
      "train_runtime": 5279.7202,
      "train_tokens_per_second": 8514.057
    },
    {
      "epoch": 0.18516024777807702,
      "grad_norm": 1.389487246494353,
      "learning_rate": 9.790155350219159e-06,
      "loss": 0.4541049003601074,
      "num_input_tokens_seen": 45033760,
      "step": 2750,
      "train_runtime": 5289.2331,
      "train_tokens_per_second": 8514.232
    },
    {
      "epoch": 0.18549690277403716,
      "grad_norm": 1.2275741357610868,
      "learning_rate": 9.789396716297399e-06,
      "loss": 0.4630864143371582,
      "num_input_tokens_seen": 45115680,
      "step": 2755,
      "train_runtime": 5298.8442,
      "train_tokens_per_second": 8514.249
    },
    {
      "epoch": 0.1858335577699973,
      "grad_norm": 1.0985062889927955,
      "learning_rate": 9.788636743036122e-06,
      "loss": 0.40701141357421877,
      "num_input_tokens_seen": 45197600,
      "step": 2760,
      "train_runtime": 5308.3764,
      "train_tokens_per_second": 8514.392
    },
    {
      "epoch": 0.18617021276595744,
      "grad_norm": 1.1220031791564755,
      "learning_rate": 9.787875430647847e-06,
      "loss": 0.43755168914794923,
      "num_input_tokens_seen": 45279520,
      "step": 2765,
      "train_runtime": 5318.0886,
      "train_tokens_per_second": 8514.247
    },
    {
      "epoch": 0.18650686776191758,
      "grad_norm": 0.9987728641058313,
      "learning_rate": 9.787112779345477e-06,
      "loss": 0.4304105281829834,
      "num_input_tokens_seen": 45361440,
      "step": 2770,
      "train_runtime": 5327.5993,
      "train_tokens_per_second": 8514.424
    },
    {
      "epoch": 0.18684352275787772,
      "grad_norm": 1.2128853361740342,
      "learning_rate": 9.786348789342283e-06,
      "loss": 0.48485450744628905,
      "num_input_tokens_seen": 45443360,
      "step": 2775,
      "train_runtime": 5337.1016,
      "train_tokens_per_second": 8514.614
    },
    {
      "epoch": 0.18718017775383786,
      "grad_norm": 1.312953862615601,
      "learning_rate": 9.785583460851912e-06,
      "loss": 0.43227510452270507,
      "num_input_tokens_seen": 45525280,
      "step": 2780,
      "train_runtime": 5346.6137,
      "train_tokens_per_second": 8514.788
    },
    {
      "epoch": 0.187516832749798,
      "grad_norm": 1.1370113647779505,
      "learning_rate": 9.784816794088386e-06,
      "loss": 0.3997438669204712,
      "num_input_tokens_seen": 45607200,
      "step": 2785,
      "train_runtime": 5356.1252,
      "train_tokens_per_second": 8514.962
    },
    {
      "epoch": 0.18785348774575814,
      "grad_norm": 1.2844316798374993,
      "learning_rate": 9.784048789266101e-06,
      "loss": 0.42080488204956057,
      "num_input_tokens_seen": 45689120,
      "step": 2790,
      "train_runtime": 5365.7732,
      "train_tokens_per_second": 8514.918
    },
    {
      "epoch": 0.18819014274171827,
      "grad_norm": 1.2670038056896182,
      "learning_rate": 9.78327944659983e-06,
      "loss": 0.4452354907989502,
      "num_input_tokens_seen": 45771040,
      "step": 2795,
      "train_runtime": 5375.4115,
      "train_tokens_per_second": 8514.891
    },
    {
      "epoch": 0.18852679773767841,
      "grad_norm": 1.2609205618569215,
      "learning_rate": 9.78250876630471e-06,
      "loss": 0.4334071159362793,
      "num_input_tokens_seen": 45852960,
      "step": 2800,
      "train_runtime": 5385.0304,
      "train_tokens_per_second": 8514.893
    },
    {
      "epoch": 0.18886345273363855,
      "grad_norm": 1.219696439306715,
      "learning_rate": 9.781736748596267e-06,
      "loss": 0.42300782203674314,
      "num_input_tokens_seen": 45934880,
      "step": 2805,
      "train_runtime": 5394.7185,
      "train_tokens_per_second": 8514.787
    },
    {
      "epoch": 0.18920010772959872,
      "grad_norm": 1.2838540203360052,
      "learning_rate": 9.78096339369039e-06,
      "loss": 0.4051196098327637,
      "num_input_tokens_seen": 46016800,
      "step": 2810,
      "train_runtime": 5404.2225,
      "train_tokens_per_second": 8514.971
    },
    {
      "epoch": 0.18953676272555886,
      "grad_norm": 1.2306987633647313,
      "learning_rate": 9.780188701803345e-06,
      "loss": 0.4592400074005127,
      "num_input_tokens_seen": 46098720,
      "step": 2815,
      "train_runtime": 5413.8661,
      "train_tokens_per_second": 8514.935
    },
    {
      "epoch": 0.189873417721519,
      "grad_norm": 1.3404641961901895,
      "learning_rate": 9.779412673151772e-06,
      "loss": 0.4534937381744385,
      "num_input_tokens_seen": 46180640,
      "step": 2820,
      "train_runtime": 5423.485,
      "train_tokens_per_second": 8514.938
    },
    {
      "epoch": 0.19021007271747914,
      "grad_norm": 1.1923411095727712,
      "learning_rate": 9.778635307952686e-06,
      "loss": 0.45238919258117677,
      "num_input_tokens_seen": 46262560,
      "step": 2825,
      "train_runtime": 5433.1495,
      "train_tokens_per_second": 8514.87
    },
    {
      "epoch": 0.19054672771343928,
      "grad_norm": 1.1221968211378168,
      "learning_rate": 9.777856606423472e-06,
      "loss": 0.4260293960571289,
      "num_input_tokens_seen": 46344480,
      "step": 2830,
      "train_runtime": 5442.6561,
      "train_tokens_per_second": 8515.048
    },
    {
      "epoch": 0.19088338270939942,
      "grad_norm": 1.343963384350832,
      "learning_rate": 9.777076568781897e-06,
      "loss": 0.448808479309082,
      "num_input_tokens_seen": 46426400,
      "step": 2835,
      "train_runtime": 5452.2437,
      "train_tokens_per_second": 8515.1
    },
    {
      "epoch": 0.19122003770535956,
      "grad_norm": 1.1284236569013701,
      "learning_rate": 9.77629519524609e-06,
      "loss": 0.3911907196044922,
      "num_input_tokens_seen": 46508320,
      "step": 2840,
      "train_runtime": 5461.7937,
      "train_tokens_per_second": 8515.21
    },
    {
      "epoch": 0.1915566927013197,
      "grad_norm": 1.3785226904848926,
      "learning_rate": 9.775512486034564e-06,
      "loss": 0.5241172790527344,
      "num_input_tokens_seen": 46590240,
      "step": 2845,
      "train_runtime": 5471.3576,
      "train_tokens_per_second": 8515.298
    },
    {
      "epoch": 0.19189334769727984,
      "grad_norm": 1.2215128059038807,
      "learning_rate": 9.774728441366197e-06,
      "loss": 0.44091949462890623,
      "num_input_tokens_seen": 46672160,
      "step": 2850,
      "train_runtime": 5480.9552,
      "train_tokens_per_second": 8515.333
    },
    {
      "epoch": 0.19223000269323998,
      "grad_norm": 1.0692640657870893,
      "learning_rate": 9.773943061460249e-06,
      "loss": 0.4249143123626709,
      "num_input_tokens_seen": 46754080,
      "step": 2855,
      "train_runtime": 5490.5534,
      "train_tokens_per_second": 8515.367
    },
    {
      "epoch": 0.19256665768920012,
      "grad_norm": 1.108067409849159,
      "learning_rate": 9.773156346536345e-06,
      "loss": 0.4164573192596436,
      "num_input_tokens_seen": 46836000,
      "step": 2860,
      "train_runtime": 5500.0924,
      "train_tokens_per_second": 8515.493
    },
    {
      "epoch": 0.19290331268516026,
      "grad_norm": 1.1816523581502831,
      "learning_rate": 9.772368296814489e-06,
      "loss": 0.3869594097137451,
      "num_input_tokens_seen": 46917920,
      "step": 2865,
      "train_runtime": 5509.6884,
      "train_tokens_per_second": 8515.531
    },
    {
      "epoch": 0.1932399676811204,
      "grad_norm": 1.3118264205023773,
      "learning_rate": 9.771578912515058e-06,
      "loss": 0.48398818969726565,
      "num_input_tokens_seen": 46999840,
      "step": 2870,
      "train_runtime": 5519.2383,
      "train_tokens_per_second": 8515.639
    },
    {
      "epoch": 0.19357662267708053,
      "grad_norm": 1.3088282211274638,
      "learning_rate": 9.7707881938588e-06,
      "loss": 0.404049015045166,
      "num_input_tokens_seen": 47081760,
      "step": 2875,
      "train_runtime": 5528.7454,
      "train_tokens_per_second": 8515.813
    },
    {
      "epoch": 0.19391327767304067,
      "grad_norm": 1.2842423940895205,
      "learning_rate": 9.769996141066836e-06,
      "loss": 0.4373288631439209,
      "num_input_tokens_seen": 47163680,
      "step": 2880,
      "train_runtime": 5538.3446,
      "train_tokens_per_second": 8515.844
    },
    {
      "epoch": 0.1942499326690008,
      "grad_norm": 1.028429663751651,
      "learning_rate": 9.76920275436066e-06,
      "loss": 0.38994417190551756,
      "num_input_tokens_seen": 47245600,
      "step": 2885,
      "train_runtime": 5547.9455,
      "train_tokens_per_second": 8515.873
    },
    {
      "epoch": 0.19458658766496095,
      "grad_norm": 1.154411147909097,
      "learning_rate": 9.768408033962145e-06,
      "loss": 0.45884027481079104,
      "num_input_tokens_seen": 47327520,
      "step": 2890,
      "train_runtime": 5557.4464,
      "train_tokens_per_second": 8516.055
    },
    {
      "epoch": 0.1949232426609211,
      "grad_norm": 1.1467053178181488,
      "learning_rate": 9.767611980093525e-06,
      "loss": 0.4258324146270752,
      "num_input_tokens_seen": 47409440,
      "step": 2895,
      "train_runtime": 5566.9502,
      "train_tokens_per_second": 8516.232
    },
    {
      "epoch": 0.19525989765688123,
      "grad_norm": 1.4237487568614042,
      "learning_rate": 9.76681459297742e-06,
      "loss": 0.4585538864135742,
      "num_input_tokens_seen": 47491360,
      "step": 2900,
      "train_runtime": 5576.4958,
      "train_tokens_per_second": 8516.345
    },
    {
      "epoch": 0.19559655265284137,
      "grad_norm": 1.3608531563890942,
      "learning_rate": 9.766015872836814e-06,
      "loss": 0.4792777538299561,
      "num_input_tokens_seen": 47573280,
      "step": 2905,
      "train_runtime": 5586.0507,
      "train_tokens_per_second": 8516.443
    },
    {
      "epoch": 0.1959332076488015,
      "grad_norm": 1.4842814710844205,
      "learning_rate": 9.765215819895066e-06,
      "loss": 0.45654945373535155,
      "num_input_tokens_seen": 47655200,
      "step": 2910,
      "train_runtime": 5595.6377,
      "train_tokens_per_second": 8516.491
    },
    {
      "epoch": 0.19626986264476165,
      "grad_norm": 1.2681348735607467,
      "learning_rate": 9.76441443437591e-06,
      "loss": 0.42092175483703614,
      "num_input_tokens_seen": 47737120,
      "step": 2915,
      "train_runtime": 5605.263,
      "train_tokens_per_second": 8516.482
    },
    {
      "epoch": 0.1966065176407218,
      "grad_norm": 1.1685100661633314,
      "learning_rate": 9.76361171650345e-06,
      "loss": 0.4368268013000488,
      "num_input_tokens_seen": 47819040,
      "step": 2920,
      "train_runtime": 5614.8225,
      "train_tokens_per_second": 8516.572
    },
    {
      "epoch": 0.19694317263668193,
      "grad_norm": 1.1679915047104785,
      "learning_rate": 9.762807666502164e-06,
      "loss": 0.44530549049377444,
      "num_input_tokens_seen": 47900960,
      "step": 2925,
      "train_runtime": 5624.4187,
      "train_tokens_per_second": 8516.606
    },
    {
      "epoch": 0.19727982763264207,
      "grad_norm": 1.1178477012889458,
      "learning_rate": 9.762002284596898e-06,
      "loss": 0.4596980571746826,
      "num_input_tokens_seen": 47982880,
      "step": 2930,
      "train_runtime": 5634.0494,
      "train_tokens_per_second": 8516.588
    },
    {
      "epoch": 0.1976164826286022,
      "grad_norm": 1.519191806535054,
      "learning_rate": 9.761195571012882e-06,
      "loss": 0.46086597442626953,
      "num_input_tokens_seen": 48064800,
      "step": 2935,
      "train_runtime": 5643.8421,
      "train_tokens_per_second": 8516.326
    },
    {
      "epoch": 0.19795313762456235,
      "grad_norm": 1.3988561593042905,
      "learning_rate": 9.760387525975705e-06,
      "loss": 0.4410702228546143,
      "num_input_tokens_seen": 48146720,
      "step": 2940,
      "train_runtime": 5653.3699,
      "train_tokens_per_second": 8516.464
    },
    {
      "epoch": 0.1982897926205225,
      "grad_norm": 1.190064896761727,
      "learning_rate": 9.759578149711337e-06,
      "loss": 0.3947296619415283,
      "num_input_tokens_seen": 48228640,
      "step": 2945,
      "train_runtime": 5662.902,
      "train_tokens_per_second": 8516.594
    },
    {
      "epoch": 0.19862644761648263,
      "grad_norm": 1.0251312494076368,
      "learning_rate": 9.758767442446117e-06,
      "loss": 0.4515190124511719,
      "num_input_tokens_seen": 48310560,
      "step": 2950,
      "train_runtime": 5672.534,
      "train_tokens_per_second": 8516.575
    },
    {
      "epoch": 0.19896310261244277,
      "grad_norm": 1.2068656608312767,
      "learning_rate": 9.757955404406755e-06,
      "loss": 0.40610947608947756,
      "num_input_tokens_seen": 48392480,
      "step": 2955,
      "train_runtime": 5682.0959,
      "train_tokens_per_second": 8516.66
    },
    {
      "epoch": 0.1992997576084029,
      "grad_norm": 1.2831442200150383,
      "learning_rate": 9.757142035820337e-06,
      "loss": 0.4035633087158203,
      "num_input_tokens_seen": 48474400,
      "step": 2960,
      "train_runtime": 5691.6835,
      "train_tokens_per_second": 8516.707
    },
    {
      "epoch": 0.19963641260436304,
      "grad_norm": 1.3164404735915487,
      "learning_rate": 9.756327336914318e-06,
      "loss": 0.4202768325805664,
      "num_input_tokens_seen": 48556320,
      "step": 2965,
      "train_runtime": 5701.2449,
      "train_tokens_per_second": 8516.792
    },
    {
      "epoch": 0.19997306760032318,
      "grad_norm": 1.325998544051736,
      "learning_rate": 9.755511307916526e-06,
      "loss": 0.4179686546325684,
      "num_input_tokens_seen": 48638240,
      "step": 2970,
      "train_runtime": 5710.8125,
      "train_tokens_per_second": 8516.869
    },
    {
      "epoch": 0.20030972259628332,
      "grad_norm": 1.1798892375111036,
      "learning_rate": 9.754693949055161e-06,
      "loss": 0.43311190605163574,
      "num_input_tokens_seen": 48720160,
      "step": 2975,
      "train_runtime": 5720.4294,
      "train_tokens_per_second": 8516.871
    },
    {
      "epoch": 0.20064637759224346,
      "grad_norm": 1.189478922952323,
      "learning_rate": 9.753875260558795e-06,
      "loss": 0.4471610069274902,
      "num_input_tokens_seen": 48801600,
      "step": 2980,
      "train_runtime": 5730.3846,
      "train_tokens_per_second": 8516.287
    },
    {
      "epoch": 0.2009830325882036,
      "grad_norm": 1.5111103551204395,
      "learning_rate": 9.75305524265637e-06,
      "loss": 0.4618964195251465,
      "num_input_tokens_seen": 48883520,
      "step": 2985,
      "train_runtime": 5740.0019,
      "train_tokens_per_second": 8516.29
    },
    {
      "epoch": 0.20131968758416374,
      "grad_norm": 1.2059978007210213,
      "learning_rate": 9.752233895577205e-06,
      "loss": 0.44857234954833985,
      "num_input_tokens_seen": 48965440,
      "step": 2990,
      "train_runtime": 5749.4984,
      "train_tokens_per_second": 8516.472
    },
    {
      "epoch": 0.20165634258012388,
      "grad_norm": 1.0013699273862064,
      "learning_rate": 9.751411219550984e-06,
      "loss": 0.40520443916320803,
      "num_input_tokens_seen": 49047360,
      "step": 2995,
      "train_runtime": 5759.0002,
      "train_tokens_per_second": 8516.645
    },
    {
      "epoch": 0.20199299757608402,
      "grad_norm": 1.1033949180875744,
      "learning_rate": 9.750587214807766e-06,
      "loss": 0.42130393981933595,
      "num_input_tokens_seen": 49129280,
      "step": 3000,
      "train_runtime": 5768.5037,
      "train_tokens_per_second": 8516.815
    },
    {
      "epoch": 0.20232965257204416,
      "grad_norm": 1.2996026532885165,
      "learning_rate": 9.749761881577981e-06,
      "loss": 0.43917622566223147,
      "num_input_tokens_seen": 49211200,
      "step": 3005,
      "train_runtime": 5778.0069,
      "train_tokens_per_second": 8516.985
    },
    {
      "epoch": 0.2026663075680043,
      "grad_norm": 1.0264493435534972,
      "learning_rate": 9.748935220092434e-06,
      "loss": 0.4145176887512207,
      "num_input_tokens_seen": 49293120,
      "step": 3010,
      "train_runtime": 5787.5047,
      "train_tokens_per_second": 8517.163
    },
    {
      "epoch": 0.20300296256396444,
      "grad_norm": 1.0835957297473238,
      "learning_rate": 9.748107230582295e-06,
      "loss": 0.438266658782959,
      "num_input_tokens_seen": 49375040,
      "step": 3015,
      "train_runtime": 5797.0072,
      "train_tokens_per_second": 8517.333
    },
    {
      "epoch": 0.20333961755992458,
      "grad_norm": 1.0999045768915143,
      "learning_rate": 9.747277913279109e-06,
      "loss": 0.4752251625061035,
      "num_input_tokens_seen": 49456960,
      "step": 3020,
      "train_runtime": 5806.5221,
      "train_tokens_per_second": 8517.484
    },
    {
      "epoch": 0.20367627255588472,
      "grad_norm": 1.3799905144681501,
      "learning_rate": 9.746447268414794e-06,
      "loss": 0.4669317722320557,
      "num_input_tokens_seen": 49538880,
      "step": 3025,
      "train_runtime": 5816.0283,
      "train_tokens_per_second": 8517.648
    },
    {
      "epoch": 0.20401292755184486,
      "grad_norm": 1.2621820617655815,
      "learning_rate": 9.745615296221635e-06,
      "loss": 0.4699692726135254,
      "num_input_tokens_seen": 49620800,
      "step": 3030,
      "train_runtime": 5825.5326,
      "train_tokens_per_second": 8517.813
    },
    {
      "epoch": 0.204349582547805,
      "grad_norm": 1.1579855797030867,
      "learning_rate": 9.74478199693229e-06,
      "loss": 0.46188921928405763,
      "num_input_tokens_seen": 49702720,
      "step": 3035,
      "train_runtime": 5835.0339,
      "train_tokens_per_second": 8517.983
    },
    {
      "epoch": 0.20468623754376514,
      "grad_norm": 1.2754663242017612,
      "learning_rate": 9.743947370779791e-06,
      "loss": 0.48488855361938477,
      "num_input_tokens_seen": 49784640,
      "step": 3040,
      "train_runtime": 5844.5353,
      "train_tokens_per_second": 8518.152
    },
    {
      "epoch": 0.20502289253972528,
      "grad_norm": 1.2695200990234408,
      "learning_rate": 9.743111417997536e-06,
      "loss": 0.457216739654541,
      "num_input_tokens_seen": 49866560,
      "step": 3045,
      "train_runtime": 5854.0394,
      "train_tokens_per_second": 8518.316
    },
    {
      "epoch": 0.20535954753568542,
      "grad_norm": 1.1605620496409133,
      "learning_rate": 9.742274138819298e-06,
      "loss": 0.44875645637512207,
      "num_input_tokens_seen": 49948000,
      "step": 3050,
      "train_runtime": 5863.883,
      "train_tokens_per_second": 8517.905
    },
    {
      "epoch": 0.20569620253164558,
      "grad_norm": 1.4291711511811938,
      "learning_rate": 9.74143553347922e-06,
      "loss": 0.48308839797973635,
      "num_input_tokens_seen": 50028840,
      "step": 3055,
      "train_runtime": 5873.5346,
      "train_tokens_per_second": 8517.672
    },
    {
      "epoch": 0.20603285752760572,
      "grad_norm": 1.3746078068386995,
      "learning_rate": 9.740595602211814e-06,
      "loss": 0.4354602813720703,
      "num_input_tokens_seen": 50110760,
      "step": 3060,
      "train_runtime": 5883.0333,
      "train_tokens_per_second": 8517.844
    },
    {
      "epoch": 0.20636951252356586,
      "grad_norm": 1.1196479007800366,
      "learning_rate": 9.739754345251963e-06,
      "loss": 0.36180663108825684,
      "num_input_tokens_seen": 50192680,
      "step": 3065,
      "train_runtime": 5892.537,
      "train_tokens_per_second": 8518.008
    },
    {
      "epoch": 0.206706167519526,
      "grad_norm": 1.1344842173485006,
      "learning_rate": 9.738911762834924e-06,
      "loss": 0.37514986991882326,
      "num_input_tokens_seen": 50274600,
      "step": 3070,
      "train_runtime": 5902.0411,
      "train_tokens_per_second": 8518.172
    },
    {
      "epoch": 0.20704282251548614,
      "grad_norm": 1.3771747100714253,
      "learning_rate": 9.738067855196323e-06,
      "loss": 0.41122021675109866,
      "num_input_tokens_seen": 50356520,
      "step": 3075,
      "train_runtime": 5911.5383,
      "train_tokens_per_second": 8518.345
    },
    {
      "epoch": 0.20737947751144628,
      "grad_norm": 1.4926023670931972,
      "learning_rate": 9.737222622572153e-06,
      "loss": 0.4811247825622559,
      "num_input_tokens_seen": 50438440,
      "step": 3080,
      "train_runtime": 5921.0343,
      "train_tokens_per_second": 8518.518
    },
    {
      "epoch": 0.20771613250740642,
      "grad_norm": 1.3951773778595116,
      "learning_rate": 9.736376065198784e-06,
      "loss": 0.4801309585571289,
      "num_input_tokens_seen": 50520360,
      "step": 3085,
      "train_runtime": 5930.5518,
      "train_tokens_per_second": 8518.661
    },
    {
      "epoch": 0.20805278750336656,
      "grad_norm": 1.0982131112521742,
      "learning_rate": 9.735528183312951e-06,
      "loss": 0.4344925880432129,
      "num_input_tokens_seen": 50602280,
      "step": 3090,
      "train_runtime": 5940.0542,
      "train_tokens_per_second": 8518.825
    },
    {
      "epoch": 0.2083894424993267,
      "grad_norm": 1.4085818746605245,
      "learning_rate": 9.734678977151762e-06,
      "loss": 0.4561446189880371,
      "num_input_tokens_seen": 50682736,
      "step": 3095,
      "train_runtime": 5949.6558,
      "train_tokens_per_second": 8518.6
    },
    {
      "epoch": 0.20872609749528684,
      "grad_norm": 1.3736396954089891,
      "learning_rate": 9.733828446952696e-06,
      "loss": 0.4277318000793457,
      "num_input_tokens_seen": 50764656,
      "step": 3100,
      "train_runtime": 5959.3091,
      "train_tokens_per_second": 8518.547
    },
    {
      "epoch": 0.20906275249124698,
      "grad_norm": 1.1962987606024325,
      "learning_rate": 9.732976592953598e-06,
      "loss": 0.46923346519470216,
      "num_input_tokens_seen": 50846576,
      "step": 3105,
      "train_runtime": 5968.9623,
      "train_tokens_per_second": 8518.495
    },
    {
      "epoch": 0.20939940748720712,
      "grad_norm": 1.1616555834107682,
      "learning_rate": 9.73212341539269e-06,
      "loss": 0.4209616184234619,
      "num_input_tokens_seen": 50928496,
      "step": 3110,
      "train_runtime": 5978.6518,
      "train_tokens_per_second": 8518.391
    },
    {
      "epoch": 0.20973606248316726,
      "grad_norm": 1.129446612149715,
      "learning_rate": 9.731268914508556e-06,
      "loss": 0.38808703422546387,
      "num_input_tokens_seen": 51010416,
      "step": 3115,
      "train_runtime": 5988.1564,
      "train_tokens_per_second": 8518.551
    },
    {
      "epoch": 0.2100727174791274,
      "grad_norm": 1.1990920061414203,
      "learning_rate": 9.730413090540158e-06,
      "loss": 0.4152238845825195,
      "num_input_tokens_seen": 51092336,
      "step": 3120,
      "train_runtime": 5997.6496,
      "train_tokens_per_second": 8518.726
    },
    {
      "epoch": 0.21040937247508754,
      "grad_norm": 1.3329725263890209,
      "learning_rate": 9.729555943726825e-06,
      "loss": 0.42784595489501953,
      "num_input_tokens_seen": 51174256,
      "step": 3125,
      "train_runtime": 6007.1433,
      "train_tokens_per_second": 8518.9
    },
    {
      "epoch": 0.21074602747104768,
      "grad_norm": 1.2322329989480878,
      "learning_rate": 9.728697474308253e-06,
      "loss": 0.4476964473724365,
      "num_input_tokens_seen": 51256136,
      "step": 3130,
      "train_runtime": 6017.2239,
      "train_tokens_per_second": 8518.236
    },
    {
      "epoch": 0.21108268246700782,
      "grad_norm": 1.2763641908180385,
      "learning_rate": 9.72783768252451e-06,
      "loss": 0.4463648319244385,
      "num_input_tokens_seen": 51338056,
      "step": 3135,
      "train_runtime": 6026.7259,
      "train_tokens_per_second": 8518.399
    },
    {
      "epoch": 0.21141933746296795,
      "grad_norm": 1.1310508837843452,
      "learning_rate": 9.726976568616036e-06,
      "loss": 0.3952059745788574,
      "num_input_tokens_seen": 51419976,
      "step": 3140,
      "train_runtime": 6036.3513,
      "train_tokens_per_second": 8518.387
    },
    {
      "epoch": 0.2117559924589281,
      "grad_norm": 1.1168360508619075,
      "learning_rate": 9.726114132823638e-06,
      "loss": 0.4087496757507324,
      "num_input_tokens_seen": 51501896,
      "step": 3145,
      "train_runtime": 6045.9,
      "train_tokens_per_second": 8518.483
    },
    {
      "epoch": 0.21209264745488823,
      "grad_norm": 1.095048807811214,
      "learning_rate": 9.725250375388493e-06,
      "loss": 0.4847414493560791,
      "num_input_tokens_seen": 51583816,
      "step": 3150,
      "train_runtime": 6055.493,
      "train_tokens_per_second": 8518.516
    },
    {
      "epoch": 0.21242930245084837,
      "grad_norm": 1.1285754618054913,
      "learning_rate": 9.724385296552147e-06,
      "loss": 0.4391319274902344,
      "num_input_tokens_seen": 51665736,
      "step": 3155,
      "train_runtime": 6065.1053,
      "train_tokens_per_second": 8518.523
    },
    {
      "epoch": 0.2127659574468085,
      "grad_norm": 1.2602043144142765,
      "learning_rate": 9.72351889655652e-06,
      "loss": 0.41916732788085936,
      "num_input_tokens_seen": 51747656,
      "step": 3160,
      "train_runtime": 6074.7149,
      "train_tokens_per_second": 8518.532
    },
    {
      "epoch": 0.21310261244276865,
      "grad_norm": 1.1926509758932977,
      "learning_rate": 9.722651175643895e-06,
      "loss": 0.4193576812744141,
      "num_input_tokens_seen": 51829576,
      "step": 3165,
      "train_runtime": 6084.2143,
      "train_tokens_per_second": 8518.697
    },
    {
      "epoch": 0.2134392674387288,
      "grad_norm": 1.176911097130805,
      "learning_rate": 9.721782134056928e-06,
      "loss": 0.3885956764221191,
      "num_input_tokens_seen": 51911496,
      "step": 3170,
      "train_runtime": 6093.8287,
      "train_tokens_per_second": 8518.7
    },
    {
      "epoch": 0.21377592243468893,
      "grad_norm": 1.0964320224657342,
      "learning_rate": 9.720911772038645e-06,
      "loss": 0.4231713771820068,
      "num_input_tokens_seen": 51993416,
      "step": 3175,
      "train_runtime": 6103.4412,
      "train_tokens_per_second": 8518.705
    },
    {
      "epoch": 0.21411257743064907,
      "grad_norm": 1.0857284767156679,
      "learning_rate": 9.720040089832436e-06,
      "loss": 0.42275247573852537,
      "num_input_tokens_seen": 52075336,
      "step": 3180,
      "train_runtime": 6112.9464,
      "train_tokens_per_second": 8518.86
    },
    {
      "epoch": 0.2144492324266092,
      "grad_norm": 1.2589764775773022,
      "learning_rate": 9.719167087682069e-06,
      "loss": 0.409316873550415,
      "num_input_tokens_seen": 52157256,
      "step": 3185,
      "train_runtime": 6122.5675,
      "train_tokens_per_second": 8518.854
    },
    {
      "epoch": 0.21478588742256935,
      "grad_norm": 1.2053503702325707,
      "learning_rate": 9.718292765831673e-06,
      "loss": 0.42337465286254883,
      "num_input_tokens_seen": 52239176,
      "step": 3190,
      "train_runtime": 6132.1271,
      "train_tokens_per_second": 8518.932
    },
    {
      "epoch": 0.2151225424185295,
      "grad_norm": 1.2336819147165112,
      "learning_rate": 9.71741712452575e-06,
      "loss": 0.44667606353759765,
      "num_input_tokens_seen": 52321096,
      "step": 3195,
      "train_runtime": 6141.631,
      "train_tokens_per_second": 8519.088
    },
    {
      "epoch": 0.21545919741448963,
      "grad_norm": 1.265961330322695,
      "learning_rate": 9.716540164009174e-06,
      "loss": 0.49404134750366213,
      "num_input_tokens_seen": 52402168,
      "step": 3200,
      "train_runtime": 6151.4205,
      "train_tokens_per_second": 8518.71
    },
    {
      "epoch": 0.21579585241044977,
      "grad_norm": 1.4883087365809178,
      "learning_rate": 9.715661884527178e-06,
      "loss": 0.46434707641601564,
      "num_input_tokens_seen": 52484088,
      "step": 3205,
      "train_runtime": 6161.0321,
      "train_tokens_per_second": 8518.717
    },
    {
      "epoch": 0.2161325074064099,
      "grad_norm": 1.1313862721623797,
      "learning_rate": 9.714782286325374e-06,
      "loss": 0.42748775482177737,
      "num_input_tokens_seen": 52566008,
      "step": 3210,
      "train_runtime": 6170.5851,
      "train_tokens_per_second": 8518.805
    },
    {
      "epoch": 0.21646916240237005,
      "grad_norm": 1.1270429991100632,
      "learning_rate": 9.713901369649738e-06,
      "loss": 0.40912432670593263,
      "num_input_tokens_seen": 52647928,
      "step": 3215,
      "train_runtime": 6180.1689,
      "train_tokens_per_second": 8518.849
    },
    {
      "epoch": 0.2168058173983302,
      "grad_norm": 1.0916261506451235,
      "learning_rate": 9.713019134746618e-06,
      "loss": 0.41660394668579104,
      "num_input_tokens_seen": 52729848,
      "step": 3220,
      "train_runtime": 6189.7926,
      "train_tokens_per_second": 8518.839
    },
    {
      "epoch": 0.21714247239429033,
      "grad_norm": 0.9828064764242315,
      "learning_rate": 9.712135581862722e-06,
      "loss": 0.45587987899780275,
      "num_input_tokens_seen": 52811488,
      "step": 3225,
      "train_runtime": 6199.7604,
      "train_tokens_per_second": 8518.311
    },
    {
      "epoch": 0.21747912739025047,
      "grad_norm": 0.9922111836181433,
      "learning_rate": 9.711250711245139e-06,
      "loss": 0.45021944046020507,
      "num_input_tokens_seen": 52893408,
      "step": 3230,
      "train_runtime": 6209.3359,
      "train_tokens_per_second": 8518.368
    },
    {
      "epoch": 0.2178157823862106,
      "grad_norm": 1.3488884024642898,
      "learning_rate": 9.710364523141317e-06,
      "loss": 0.4450529098510742,
      "num_input_tokens_seen": 52975328,
      "step": 3235,
      "train_runtime": 6218.9423,
      "train_tokens_per_second": 8518.382
    },
    {
      "epoch": 0.21815243738217074,
      "grad_norm": 1.3212500573348611,
      "learning_rate": 9.709477017799076e-06,
      "loss": 0.5077713012695313,
      "num_input_tokens_seen": 53055864,
      "step": 3240,
      "train_runtime": 6228.5475,
      "train_tokens_per_second": 8518.176
    },
    {
      "epoch": 0.21848909237813088,
      "grad_norm": 1.1781287022541025,
      "learning_rate": 9.708588195466604e-06,
      "loss": 0.4256113052368164,
      "num_input_tokens_seen": 53137784,
      "step": 3245,
      "train_runtime": 6238.0521,
      "train_tokens_per_second": 8518.33
    },
    {
      "epoch": 0.21882574737409102,
      "grad_norm": 1.1295377951782966,
      "learning_rate": 9.707698056392458e-06,
      "loss": 0.46156110763549807,
      "num_input_tokens_seen": 53219704,
      "step": 3250,
      "train_runtime": 6247.6918,
      "train_tokens_per_second": 8518.299
    },
    {
      "epoch": 0.21916240237005116,
      "grad_norm": 1.0754618109697198,
      "learning_rate": 9.706806600825562e-06,
      "loss": 0.43082284927368164,
      "num_input_tokens_seen": 53301624,
      "step": 3255,
      "train_runtime": 6257.3111,
      "train_tokens_per_second": 8518.295
    },
    {
      "epoch": 0.2194990573660113,
      "grad_norm": 1.2447206925249599,
      "learning_rate": 9.70591382901521e-06,
      "loss": 0.44476141929626467,
      "num_input_tokens_seen": 53383544,
      "step": 3260,
      "train_runtime": 6266.8192,
      "train_tokens_per_second": 8518.443
    },
    {
      "epoch": 0.21983571236197144,
      "grad_norm": 1.043692978127584,
      "learning_rate": 9.705019741211061e-06,
      "loss": 0.4163410186767578,
      "num_input_tokens_seen": 53465464,
      "step": 3265,
      "train_runtime": 6276.4236,
      "train_tokens_per_second": 8518.46
    },
    {
      "epoch": 0.22017236735793158,
      "grad_norm": 1.3845009607823044,
      "learning_rate": 9.704124337663145e-06,
      "loss": 0.4174496650695801,
      "num_input_tokens_seen": 53547384,
      "step": 3270,
      "train_runtime": 6285.988,
      "train_tokens_per_second": 8518.531
    },
    {
      "epoch": 0.22050902235389172,
      "grad_norm": 1.2868761916613507,
      "learning_rate": 9.703227618621859e-06,
      "loss": 0.4560702323913574,
      "num_input_tokens_seen": 53629304,
      "step": 3275,
      "train_runtime": 6295.5007,
      "train_tokens_per_second": 8518.672
    },
    {
      "epoch": 0.22084567734985186,
      "grad_norm": 1.4694417614333468,
      "learning_rate": 9.702329584337965e-06,
      "loss": 0.4209280490875244,
      "num_input_tokens_seen": 53711224,
      "step": 3280,
      "train_runtime": 6305.0668,
      "train_tokens_per_second": 8518.74
    },
    {
      "epoch": 0.221182332345812,
      "grad_norm": 1.0575490303891482,
      "learning_rate": 9.701430235062599e-06,
      "loss": 0.40921831130981445,
      "num_input_tokens_seen": 53793144,
      "step": 3285,
      "train_runtime": 6314.6741,
      "train_tokens_per_second": 8518.752
    },
    {
      "epoch": 0.22151898734177214,
      "grad_norm": 1.3510005404553216,
      "learning_rate": 9.700529571047259e-06,
      "loss": 0.45744547843933103,
      "num_input_tokens_seen": 53875064,
      "step": 3290,
      "train_runtime": 6324.2219,
      "train_tokens_per_second": 8518.845
    },
    {
      "epoch": 0.22185564233773228,
      "grad_norm": 1.2397423740234894,
      "learning_rate": 9.699627592543814e-06,
      "loss": 0.41347675323486327,
      "num_input_tokens_seen": 53956984,
      "step": 3295,
      "train_runtime": 6333.7909,
      "train_tokens_per_second": 8518.908
    },
    {
      "epoch": 0.22219229733369245,
      "grad_norm": 1.0573185498336461,
      "learning_rate": 9.698724299804498e-06,
      "loss": 0.4576864242553711,
      "num_input_tokens_seen": 54038904,
      "step": 3300,
      "train_runtime": 6343.3705,
      "train_tokens_per_second": 8518.958
    },
    {
      "epoch": 0.22252895232965259,
      "grad_norm": 1.3205890822914574,
      "learning_rate": 9.697819693081915e-06,
      "loss": 0.4102386474609375,
      "num_input_tokens_seen": 54120824,
      "step": 3305,
      "train_runtime": 6352.9254,
      "train_tokens_per_second": 8519.04
    },
    {
      "epoch": 0.22286560732561272,
      "grad_norm": 0.981040872794871,
      "learning_rate": 9.696913772629034e-06,
      "loss": 0.3816809177398682,
      "num_input_tokens_seen": 54202744,
      "step": 3310,
      "train_runtime": 6362.5084,
      "train_tokens_per_second": 8519.084
    },
    {
      "epoch": 0.22320226232157286,
      "grad_norm": 1.1052917104768127,
      "learning_rate": 9.696006538699195e-06,
      "loss": 0.40474863052368165,
      "num_input_tokens_seen": 54284664,
      "step": 3315,
      "train_runtime": 6372.1211,
      "train_tokens_per_second": 8519.089
    },
    {
      "epoch": 0.223538917317533,
      "grad_norm": 1.2458479200833716,
      "learning_rate": 9.695097991546102e-06,
      "loss": 0.4550772666931152,
      "num_input_tokens_seen": 54366584,
      "step": 3320,
      "train_runtime": 6381.6908,
      "train_tokens_per_second": 8519.15
    },
    {
      "epoch": 0.22387557231349314,
      "grad_norm": 1.156684593412203,
      "learning_rate": 9.694188131423828e-06,
      "loss": 0.44279985427856444,
      "num_input_tokens_seen": 54448504,
      "step": 3325,
      "train_runtime": 6391.2464,
      "train_tokens_per_second": 8519.231
    },
    {
      "epoch": 0.22421222730945328,
      "grad_norm": 1.1240862852204137,
      "learning_rate": 9.693276958586809e-06,
      "loss": 0.4797492980957031,
      "num_input_tokens_seen": 54530424,
      "step": 3330,
      "train_runtime": 6400.8517,
      "train_tokens_per_second": 8519.245
    },
    {
      "epoch": 0.22454888230541342,
      "grad_norm": 1.0916835000253744,
      "learning_rate": 9.692364473289856e-06,
      "loss": 0.4147963047027588,
      "num_input_tokens_seen": 54612344,
      "step": 3335,
      "train_runtime": 6410.4377,
      "train_tokens_per_second": 8519.285
    },
    {
      "epoch": 0.22488553730137356,
      "grad_norm": 1.4649660629568968,
      "learning_rate": 9.69145067578814e-06,
      "loss": 0.45444483757019044,
      "num_input_tokens_seen": 54694264,
      "step": 3340,
      "train_runtime": 6419.9554,
      "train_tokens_per_second": 8519.415
    },
    {
      "epoch": 0.2252221922973337,
      "grad_norm": 1.0505876111335704,
      "learning_rate": 9.690535566337205e-06,
      "loss": 0.48932790756225586,
      "num_input_tokens_seen": 54776184,
      "step": 3345,
      "train_runtime": 6429.5696,
      "train_tokens_per_second": 8519.417
    },
    {
      "epoch": 0.22555884729329384,
      "grad_norm": 1.277246670469249,
      "learning_rate": 9.689619145192955e-06,
      "loss": 0.4143549919128418,
      "num_input_tokens_seen": 54858104,
      "step": 3350,
      "train_runtime": 6439.1826,
      "train_tokens_per_second": 8519.42
    },
    {
      "epoch": 0.22589550228925398,
      "grad_norm": 1.113625496402231,
      "learning_rate": 9.688701412611663e-06,
      "loss": 0.3806758880615234,
      "num_input_tokens_seen": 54940024,
      "step": 3355,
      "train_runtime": 6448.6903,
      "train_tokens_per_second": 8519.563
    },
    {
      "epoch": 0.22623215728521412,
      "grad_norm": 1.2705855892070432,
      "learning_rate": 9.687782368849973e-06,
      "loss": 0.42154650688171386,
      "num_input_tokens_seen": 55021944,
      "step": 3360,
      "train_runtime": 6458.2983,
      "train_tokens_per_second": 8519.573
    },
    {
      "epoch": 0.22656881228117426,
      "grad_norm": 1.2355959200697268,
      "learning_rate": 9.686862014164893e-06,
      "loss": 0.4214019775390625,
      "num_input_tokens_seen": 55103864,
      "step": 3365,
      "train_runtime": 6467.9351,
      "train_tokens_per_second": 8519.545
    },
    {
      "epoch": 0.2269054672771344,
      "grad_norm": 1.158558623144231,
      "learning_rate": 9.685940348813795e-06,
      "loss": 0.3992945432662964,
      "num_input_tokens_seen": 55185784,
      "step": 3370,
      "train_runtime": 6477.4418,
      "train_tokens_per_second": 8519.688
    },
    {
      "epoch": 0.22724212227309454,
      "grad_norm": 1.1937925117752506,
      "learning_rate": 9.685017373054421e-06,
      "loss": 0.43157157897949217,
      "num_input_tokens_seen": 55267704,
      "step": 3375,
      "train_runtime": 6487.0434,
      "train_tokens_per_second": 8519.706
    },
    {
      "epoch": 0.22757877726905468,
      "grad_norm": 1.2176224559158038,
      "learning_rate": 9.68409308714488e-06,
      "loss": 0.4478471755981445,
      "num_input_tokens_seen": 55349280,
      "step": 3380,
      "train_runtime": 6497.0523,
      "train_tokens_per_second": 8519.137
    },
    {
      "epoch": 0.22791543226501482,
      "grad_norm": 1.1455784577753079,
      "learning_rate": 9.68316749134364e-06,
      "loss": 0.40642642974853516,
      "num_input_tokens_seen": 55431200,
      "step": 3385,
      "train_runtime": 6506.6052,
      "train_tokens_per_second": 8519.22
    },
    {
      "epoch": 0.22825208726097496,
      "grad_norm": 1.5052658581259843,
      "learning_rate": 9.682240585909545e-06,
      "loss": 0.4301703453063965,
      "num_input_tokens_seen": 55513120,
      "step": 3390,
      "train_runtime": 6516.1796,
      "train_tokens_per_second": 8519.274
    },
    {
      "epoch": 0.2285887422569351,
      "grad_norm": 1.2784554441450375,
      "learning_rate": 9.681312371101803e-06,
      "loss": 0.42469110488891604,
      "num_input_tokens_seen": 55595040,
      "step": 3395,
      "train_runtime": 6525.8036,
      "train_tokens_per_second": 8519.263
    },
    {
      "epoch": 0.22892539725289524,
      "grad_norm": 1.1926756504636167,
      "learning_rate": 9.680382847179984e-06,
      "loss": 0.4342975616455078,
      "num_input_tokens_seen": 55676960,
      "step": 3400,
      "train_runtime": 6535.3559,
      "train_tokens_per_second": 8519.346
    },
    {
      "epoch": 0.22926205224885537,
      "grad_norm": 1.2104326847491698,
      "learning_rate": 9.679452014404025e-06,
      "loss": 0.4501530647277832,
      "num_input_tokens_seen": 55758880,
      "step": 3405,
      "train_runtime": 6544.9309,
      "train_tokens_per_second": 8519.399
    },
    {
      "epoch": 0.22959870724481551,
      "grad_norm": 1.1633637915534858,
      "learning_rate": 9.678519873034234e-06,
      "loss": 0.472019624710083,
      "num_input_tokens_seen": 55840800,
      "step": 3410,
      "train_runtime": 6554.551,
      "train_tokens_per_second": 8519.394
    },
    {
      "epoch": 0.22993536224077565,
      "grad_norm": 1.2932101101201168,
      "learning_rate": 9.677586423331277e-06,
      "loss": 0.4513254165649414,
      "num_input_tokens_seen": 55922720,
      "step": 3415,
      "train_runtime": 6564.117,
      "train_tokens_per_second": 8519.458
    },
    {
      "epoch": 0.2302720172367358,
      "grad_norm": 1.0375412070248928,
      "learning_rate": 9.676651665556194e-06,
      "loss": 0.44385643005371095,
      "num_input_tokens_seen": 56004640,
      "step": 3420,
      "train_runtime": 6573.6869,
      "train_tokens_per_second": 8519.517
    },
    {
      "epoch": 0.23060867223269593,
      "grad_norm": 1.1806895164355133,
      "learning_rate": 9.675715599970383e-06,
      "loss": 0.42416744232177733,
      "num_input_tokens_seen": 56086560,
      "step": 3425,
      "train_runtime": 6583.3091,
      "train_tokens_per_second": 8519.509
    },
    {
      "epoch": 0.23094532722865607,
      "grad_norm": 1.2197190526578587,
      "learning_rate": 9.674778226835617e-06,
      "loss": 0.45301027297973634,
      "num_input_tokens_seen": 56168480,
      "step": 3430,
      "train_runtime": 6592.814,
      "train_tokens_per_second": 8519.652
    },
    {
      "epoch": 0.2312819822246162,
      "grad_norm": 1.2170335208228702,
      "learning_rate": 9.673839546414026e-06,
      "loss": 0.4443193435668945,
      "num_input_tokens_seen": 56250400,
      "step": 3435,
      "train_runtime": 6602.3521,
      "train_tokens_per_second": 8519.752
    },
    {
      "epoch": 0.23161863722057635,
      "grad_norm": 1.0579410766326198,
      "learning_rate": 9.672899558968107e-06,
      "loss": 0.40223007202148436,
      "num_input_tokens_seen": 56332320,
      "step": 3440,
      "train_runtime": 6611.9282,
      "train_tokens_per_second": 8519.802
    },
    {
      "epoch": 0.2319552922165365,
      "grad_norm": 1.7197215497676386,
      "learning_rate": 9.67195826476073e-06,
      "loss": 0.4418642044067383,
      "num_input_tokens_seen": 56414240,
      "step": 3445,
      "train_runtime": 6621.5427,
      "train_tokens_per_second": 8519.803
    },
    {
      "epoch": 0.23229194721249663,
      "grad_norm": 1.154259153852481,
      "learning_rate": 9.671015664055122e-06,
      "loss": 0.43287010192871095,
      "num_input_tokens_seen": 56496160,
      "step": 3450,
      "train_runtime": 6631.2732,
      "train_tokens_per_second": 8519.655
    },
    {
      "epoch": 0.23262860220845677,
      "grad_norm": 1.0440397963526644,
      "learning_rate": 9.670071757114877e-06,
      "loss": 0.40431766510009765,
      "num_input_tokens_seen": 56578080,
      "step": 3455,
      "train_runtime": 6640.7913,
      "train_tokens_per_second": 8519.78
    },
    {
      "epoch": 0.2329652572044169,
      "grad_norm": 1.1191681400991553,
      "learning_rate": 9.669126544203957e-06,
      "loss": 0.43839221000671386,
      "num_input_tokens_seen": 56659176,
      "step": 3460,
      "train_runtime": 6651.1645,
      "train_tokens_per_second": 8518.685
    },
    {
      "epoch": 0.23330191220037705,
      "grad_norm": 1.212589186570543,
      "learning_rate": 9.668180025586689e-06,
      "loss": 0.4573817253112793,
      "num_input_tokens_seen": 56741096,
      "step": 3465,
      "train_runtime": 6660.8542,
      "train_tokens_per_second": 8518.592
    },
    {
      "epoch": 0.2336385671963372,
      "grad_norm": 1.0362962671671163,
      "learning_rate": 9.667232201527761e-06,
      "loss": 0.39375553131103513,
      "num_input_tokens_seen": 56823016,
      "step": 3470,
      "train_runtime": 6670.4742,
      "train_tokens_per_second": 8518.587
    },
    {
      "epoch": 0.23397522219229733,
      "grad_norm": 1.1316049282008094,
      "learning_rate": 9.666283072292234e-06,
      "loss": 0.44550604820251466,
      "num_input_tokens_seen": 56904936,
      "step": 3475,
      "train_runtime": 6680.0245,
      "train_tokens_per_second": 8518.672
    },
    {
      "epoch": 0.23431187718825747,
      "grad_norm": 1.2579963031875576,
      "learning_rate": 9.665332638145522e-06,
      "loss": 0.41012210845947267,
      "num_input_tokens_seen": 56986856,
      "step": 3480,
      "train_runtime": 6689.5932,
      "train_tokens_per_second": 8518.733
    },
    {
      "epoch": 0.2346485321842176,
      "grad_norm": 1.4783063885930563,
      "learning_rate": 9.664380899353414e-06,
      "loss": 0.49453029632568357,
      "num_input_tokens_seen": 57068776,
      "step": 3485,
      "train_runtime": 6699.2128,
      "train_tokens_per_second": 8518.729
    },
    {
      "epoch": 0.23498518718017775,
      "grad_norm": 1.3014977158413765,
      "learning_rate": 9.663427856182062e-06,
      "loss": 0.46018595695495607,
      "num_input_tokens_seen": 57150696,
      "step": 3490,
      "train_runtime": 6708.7558,
      "train_tokens_per_second": 8518.822
    },
    {
      "epoch": 0.23532184217613789,
      "grad_norm": 1.0265168070522868,
      "learning_rate": 9.662473508897981e-06,
      "loss": 0.3994332551956177,
      "num_input_tokens_seen": 57232616,
      "step": 3495,
      "train_runtime": 6718.361,
      "train_tokens_per_second": 8518.836
    },
    {
      "epoch": 0.23565849717209802,
      "grad_norm": 1.2863728521515567,
      "learning_rate": 9.661517857768052e-06,
      "loss": 0.4548481464385986,
      "num_input_tokens_seen": 57314536,
      "step": 3500,
      "train_runtime": 6727.9145,
      "train_tokens_per_second": 8518.916
    },
    {
      "epoch": 0.23599515216805816,
      "grad_norm": 1.2647216840702709,
      "learning_rate": 9.660560903059517e-06,
      "loss": 0.41991291046142576,
      "num_input_tokens_seen": 57396456,
      "step": 3505,
      "train_runtime": 6737.4219,
      "train_tokens_per_second": 8519.053
    },
    {
      "epoch": 0.2363318071640183,
      "grad_norm": 1.1681596657406166,
      "learning_rate": 9.659602645039988e-06,
      "loss": 0.45469970703125,
      "num_input_tokens_seen": 57478376,
      "step": 3510,
      "train_runtime": 6746.927,
      "train_tokens_per_second": 8519.193
    },
    {
      "epoch": 0.23666846215997844,
      "grad_norm": 1.1747352255735741,
      "learning_rate": 9.658643083977435e-06,
      "loss": 0.37882540225982664,
      "num_input_tokens_seen": 57560296,
      "step": 3515,
      "train_runtime": 6756.4308,
      "train_tokens_per_second": 8519.335
    },
    {
      "epoch": 0.23700511715593858,
      "grad_norm": 1.1157151489647517,
      "learning_rate": 9.657682220140202e-06,
      "loss": 0.37262802124023436,
      "num_input_tokens_seen": 57641896,
      "step": 3520,
      "train_runtime": 6766.3452,
      "train_tokens_per_second": 8518.912
    },
    {
      "epoch": 0.23734177215189872,
      "grad_norm": 1.5094245283284085,
      "learning_rate": 9.656720053796987e-06,
      "loss": 0.4628415107727051,
      "num_input_tokens_seen": 57723816,
      "step": 3525,
      "train_runtime": 6775.8585,
      "train_tokens_per_second": 8519.041
    },
    {
      "epoch": 0.23767842714785886,
      "grad_norm": 1.1450752498496293,
      "learning_rate": 9.655756585216859e-06,
      "loss": 0.40655269622802737,
      "num_input_tokens_seen": 57805144,
      "step": 3530,
      "train_runtime": 6785.7477,
      "train_tokens_per_second": 8518.611
    },
    {
      "epoch": 0.238015082143819,
      "grad_norm": 1.4188139095573646,
      "learning_rate": 9.654791814669246e-06,
      "loss": 0.4503175735473633,
      "num_input_tokens_seen": 57887064,
      "step": 3535,
      "train_runtime": 6795.3579,
      "train_tokens_per_second": 8518.619
    },
    {
      "epoch": 0.23835173713977914,
      "grad_norm": 1.2023830024499835,
      "learning_rate": 9.653825742423949e-06,
      "loss": 0.40427389144897463,
      "num_input_tokens_seen": 57968984,
      "step": 3540,
      "train_runtime": 6804.9702,
      "train_tokens_per_second": 8518.624
    },
    {
      "epoch": 0.2386883921357393,
      "grad_norm": 1.1678389682260266,
      "learning_rate": 9.652858368751118e-06,
      "loss": 0.42859511375427245,
      "num_input_tokens_seen": 58050904,
      "step": 3545,
      "train_runtime": 6814.5815,
      "train_tokens_per_second": 8518.631
    },
    {
      "epoch": 0.23902504713169945,
      "grad_norm": 1.221228897418193,
      "learning_rate": 9.651889693921284e-06,
      "loss": 0.43917236328125,
      "num_input_tokens_seen": 58132824,
      "step": 3550,
      "train_runtime": 6824.0815,
      "train_tokens_per_second": 8518.776
    },
    {
      "epoch": 0.2393617021276596,
      "grad_norm": 1.2681833123258317,
      "learning_rate": 9.650919718205331e-06,
      "loss": 0.4780115127563477,
      "num_input_tokens_seen": 58214744,
      "step": 3555,
      "train_runtime": 6833.5861,
      "train_tokens_per_second": 8518.916
    },
    {
      "epoch": 0.23969835712361973,
      "grad_norm": 1.2218819537064174,
      "learning_rate": 9.649948441874509e-06,
      "loss": 0.4553210258483887,
      "num_input_tokens_seen": 58296664,
      "step": 3560,
      "train_runtime": 6843.0844,
      "train_tokens_per_second": 8519.062
    },
    {
      "epoch": 0.24003501211957987,
      "grad_norm": 0.9686006012636516,
      "learning_rate": 9.648975865200434e-06,
      "loss": 0.4180324554443359,
      "num_input_tokens_seen": 58378584,
      "step": 3565,
      "train_runtime": 6852.5826,
      "train_tokens_per_second": 8519.209
    },
    {
      "epoch": 0.24037166711554,
      "grad_norm": 1.1573170810436866,
      "learning_rate": 9.648001988455082e-06,
      "loss": 0.41420783996582033,
      "num_input_tokens_seen": 58460504,
      "step": 3570,
      "train_runtime": 6862.0796,
      "train_tokens_per_second": 8519.357
    },
    {
      "epoch": 0.24070832211150014,
      "grad_norm": 1.1200273607077555,
      "learning_rate": 9.647026811910794e-06,
      "loss": 0.451705265045166,
      "num_input_tokens_seen": 58542424,
      "step": 3575,
      "train_runtime": 6871.5811,
      "train_tokens_per_second": 8519.498
    },
    {
      "epoch": 0.24104497710746028,
      "grad_norm": 1.1792553553769725,
      "learning_rate": 9.646050335840279e-06,
      "loss": 0.48875813484191893,
      "num_input_tokens_seen": 58624344,
      "step": 3580,
      "train_runtime": 6881.0813,
      "train_tokens_per_second": 8519.641
    },
    {
      "epoch": 0.24138163210342042,
      "grad_norm": 1.2197286164332142,
      "learning_rate": 9.6450725605166e-06,
      "loss": 0.4556109428405762,
      "num_input_tokens_seen": 58706264,
      "step": 3585,
      "train_runtime": 6890.5738,
      "train_tokens_per_second": 8519.793
    },
    {
      "epoch": 0.24171828709938056,
      "grad_norm": 0.9470412710043186,
      "learning_rate": 9.644093486213192e-06,
      "loss": 0.4915739059448242,
      "num_input_tokens_seen": 58788184,
      "step": 3590,
      "train_runtime": 6900.1246,
      "train_tokens_per_second": 8519.873
    },
    {
      "epoch": 0.2420549420953407,
      "grad_norm": 1.3994992387878225,
      "learning_rate": 9.64311311320385e-06,
      "loss": 0.46520357131958007,
      "num_input_tokens_seen": 58870104,
      "step": 3595,
      "train_runtime": 6909.6826,
      "train_tokens_per_second": 8519.943
    },
    {
      "epoch": 0.24239159709130084,
      "grad_norm": 1.1655977339672836,
      "learning_rate": 9.642131441762732e-06,
      "loss": 0.4594570159912109,
      "num_input_tokens_seen": 58952024,
      "step": 3600,
      "train_runtime": 6919.2935,
      "train_tokens_per_second": 8519.948
    },
    {
      "epoch": 0.24272825208726098,
      "grad_norm": 1.1510312565659875,
      "learning_rate": 9.641148472164359e-06,
      "loss": 0.40817413330078123,
      "num_input_tokens_seen": 59033944,
      "step": 3605,
      "train_runtime": 6928.9077,
      "train_tokens_per_second": 8519.95
    },
    {
      "epoch": 0.24306490708322112,
      "grad_norm": 1.2369602945375762,
      "learning_rate": 9.640164204683616e-06,
      "loss": 0.4336637020111084,
      "num_input_tokens_seen": 59115864,
      "step": 3610,
      "train_runtime": 6938.5157,
      "train_tokens_per_second": 8519.958
    },
    {
      "epoch": 0.24340156207918126,
      "grad_norm": 1.3193720497671733,
      "learning_rate": 9.639178639595748e-06,
      "loss": 0.4624314308166504,
      "num_input_tokens_seen": 59197784,
      "step": 3615,
      "train_runtime": 6948.0118,
      "train_tokens_per_second": 8520.104
    },
    {
      "epoch": 0.2437382170751414,
      "grad_norm": 1.0874478211846965,
      "learning_rate": 9.638191777176367e-06,
      "loss": 0.37902278900146485,
      "num_input_tokens_seen": 59279704,
      "step": 3620,
      "train_runtime": 6957.6238,
      "train_tokens_per_second": 8520.108
    },
    {
      "epoch": 0.24407487207110154,
      "grad_norm": 1.1501951167064481,
      "learning_rate": 9.637203617701446e-06,
      "loss": 0.45565996170043943,
      "num_input_tokens_seen": 59361624,
      "step": 3625,
      "train_runtime": 6967.2464,
      "train_tokens_per_second": 8520.098
    },
    {
      "epoch": 0.24441152706706168,
      "grad_norm": 1.0197675091126799,
      "learning_rate": 9.63621416144732e-06,
      "loss": 0.4166267395019531,
      "num_input_tokens_seen": 59443544,
      "step": 3630,
      "train_runtime": 6976.7881,
      "train_tokens_per_second": 8520.188
    },
    {
      "epoch": 0.24474818206302182,
      "grad_norm": 1.2160185130255508,
      "learning_rate": 9.635223408690688e-06,
      "loss": 0.4266179084777832,
      "num_input_tokens_seen": 59525464,
      "step": 3635,
      "train_runtime": 6986.339,
      "train_tokens_per_second": 8520.266
    },
    {
      "epoch": 0.24508483705898196,
      "grad_norm": 1.2932258586168115,
      "learning_rate": 9.63423135970861e-06,
      "loss": 0.4542896270751953,
      "num_input_tokens_seen": 59607384,
      "step": 3640,
      "train_runtime": 6995.9485,
      "train_tokens_per_second": 8520.272
    },
    {
      "epoch": 0.2454214920549421,
      "grad_norm": 1.179108595761894,
      "learning_rate": 9.633238014778511e-06,
      "loss": 0.4287732601165771,
      "num_input_tokens_seen": 59689304,
      "step": 3645,
      "train_runtime": 7005.4985,
      "train_tokens_per_second": 8520.351
    },
    {
      "epoch": 0.24575814705090224,
      "grad_norm": 1.2296513577172206,
      "learning_rate": 9.632243374178173e-06,
      "loss": 0.4582559108734131,
      "num_input_tokens_seen": 59771224,
      "step": 3650,
      "train_runtime": 7015.0716,
      "train_tokens_per_second": 8520.401
    },
    {
      "epoch": 0.24609480204686238,
      "grad_norm": 1.0557856077240633,
      "learning_rate": 9.63124743818575e-06,
      "loss": 0.42542638778686526,
      "num_input_tokens_seen": 59853144,
      "step": 3655,
      "train_runtime": 7024.6729,
      "train_tokens_per_second": 8520.417
    },
    {
      "epoch": 0.24643145704282252,
      "grad_norm": 1.1266858018917982,
      "learning_rate": 9.630250207079745e-06,
      "loss": 0.43625602722167967,
      "num_input_tokens_seen": 59935064,
      "step": 3660,
      "train_runtime": 7034.2968,
      "train_tokens_per_second": 8520.406
    },
    {
      "epoch": 0.24676811203878266,
      "grad_norm": 1.118287465684548,
      "learning_rate": 9.629251681139038e-06,
      "loss": 0.45920476913452146,
      "num_input_tokens_seen": 60016984,
      "step": 3665,
      "train_runtime": 7043.8076,
      "train_tokens_per_second": 8520.531
    },
    {
      "epoch": 0.2471047670347428,
      "grad_norm": 1.1711078295905368,
      "learning_rate": 9.628251860642857e-06,
      "loss": 0.4357304573059082,
      "num_input_tokens_seen": 60098904,
      "step": 3670,
      "train_runtime": 7053.4123,
      "train_tokens_per_second": 8520.543
    },
    {
      "epoch": 0.24744142203070293,
      "grad_norm": 1.1085274742219997,
      "learning_rate": 9.627250745870805e-06,
      "loss": 0.41866245269775393,
      "num_input_tokens_seen": 60180824,
      "step": 3675,
      "train_runtime": 7063.0563,
      "train_tokens_per_second": 8520.507
    },
    {
      "epoch": 0.24777807702666307,
      "grad_norm": 1.4030586694040714,
      "learning_rate": 9.626248337102834e-06,
      "loss": 0.5033211708068848,
      "num_input_tokens_seen": 60262232,
      "step": 3680,
      "train_runtime": 7072.8203,
      "train_tokens_per_second": 8520.255
    },
    {
      "epoch": 0.2481147320226232,
      "grad_norm": 1.2038806378543894,
      "learning_rate": 9.625244634619269e-06,
      "loss": 0.49502973556518554,
      "num_input_tokens_seen": 60343760,
      "step": 3685,
      "train_runtime": 7082.7673,
      "train_tokens_per_second": 8519.8
    },
    {
      "epoch": 0.24845138701858335,
      "grad_norm": 1.1277710503805702,
      "learning_rate": 9.62423963870079e-06,
      "loss": 0.433339786529541,
      "num_input_tokens_seen": 60425632,
      "step": 3690,
      "train_runtime": 7092.9143,
      "train_tokens_per_second": 8519.154
    },
    {
      "epoch": 0.2487880420145435,
      "grad_norm": 1.0748052624146178,
      "learning_rate": 9.623233349628441e-06,
      "loss": 0.3709014892578125,
      "num_input_tokens_seen": 60507552,
      "step": 3695,
      "train_runtime": 7102.469,
      "train_tokens_per_second": 8519.228
    },
    {
      "epoch": 0.24912469701050363,
      "grad_norm": 1.1983128837685233,
      "learning_rate": 9.62222576768363e-06,
      "loss": 0.4087608814239502,
      "num_input_tokens_seen": 60589472,
      "step": 3700,
      "train_runtime": 7112.0398,
      "train_tokens_per_second": 8519.282
    },
    {
      "epoch": 0.24946135200646377,
      "grad_norm": 1.1580276483657497,
      "learning_rate": 9.62121689314812e-06,
      "loss": 0.4702014923095703,
      "num_input_tokens_seen": 60671392,
      "step": 3705,
      "train_runtime": 7121.648,
      "train_tokens_per_second": 8519.291
    },
    {
      "epoch": 0.2497980070024239,
      "grad_norm": 1.4084241399162263,
      "learning_rate": 9.620206726304045e-06,
      "loss": 0.43607134819030763,
      "num_input_tokens_seen": 60753312,
      "step": 3710,
      "train_runtime": 7131.2054,
      "train_tokens_per_second": 8519.361
    },
    {
      "epoch": 0.2501346619983841,
      "grad_norm": 1.136900607277101,
      "learning_rate": 9.61919526743389e-06,
      "loss": 0.4764397144317627,
      "num_input_tokens_seen": 60835232,
      "step": 3715,
      "train_runtime": 7140.7651,
      "train_tokens_per_second": 8519.428
    },
    {
      "epoch": 0.2504713169943442,
      "grad_norm": 1.2220900310503655,
      "learning_rate": 9.618182516820507e-06,
      "loss": 0.42827677726745605,
      "num_input_tokens_seen": 60917152,
      "step": 3720,
      "train_runtime": 7150.3817,
      "train_tokens_per_second": 8519.427
    },
    {
      "epoch": 0.25080797199030436,
      "grad_norm": 1.2009688975652735,
      "learning_rate": 9.61716847474711e-06,
      "loss": 0.42269105911254884,
      "num_input_tokens_seen": 60999072,
      "step": 3725,
      "train_runtime": 7159.9389,
      "train_tokens_per_second": 8519.496
    },
    {
      "epoch": 0.2511446269862645,
      "grad_norm": 1.1113378981153559,
      "learning_rate": 9.616153141497273e-06,
      "loss": 0.451932430267334,
      "num_input_tokens_seen": 61080992,
      "step": 3730,
      "train_runtime": 7169.5182,
      "train_tokens_per_second": 8519.539
    },
    {
      "epoch": 0.25148128198222464,
      "grad_norm": 12.195360275573549,
      "learning_rate": 9.61513651735493e-06,
      "loss": 0.3682888984680176,
      "num_input_tokens_seen": 61162912,
      "step": 3735,
      "train_runtime": 7179.1523,
      "train_tokens_per_second": 8519.517
    },
    {
      "epoch": 0.2518179369781848,
      "grad_norm": 1.1513686849592202,
      "learning_rate": 9.614118602604376e-06,
      "loss": 0.4225430965423584,
      "num_input_tokens_seen": 61244832,
      "step": 3740,
      "train_runtime": 7188.7664,
      "train_tokens_per_second": 8519.519
    },
    {
      "epoch": 0.2521545919741449,
      "grad_norm": 1.1554483345769846,
      "learning_rate": 9.613099397530267e-06,
      "loss": 0.38666424751281736,
      "num_input_tokens_seen": 61326752,
      "step": 3745,
      "train_runtime": 7198.2708,
      "train_tokens_per_second": 8519.651
    },
    {
      "epoch": 0.25249124697010505,
      "grad_norm": 1.2996418418938394,
      "learning_rate": 9.612078902417623e-06,
      "loss": 0.39128961563110354,
      "num_input_tokens_seen": 61408672,
      "step": 3750,
      "train_runtime": 7207.8736,
      "train_tokens_per_second": 8519.666
    },
    {
      "epoch": 0.2528279019660652,
      "grad_norm": 1.2612440647552077,
      "learning_rate": 9.611057117551821e-06,
      "loss": 0.43924388885498045,
      "num_input_tokens_seen": 61490592,
      "step": 3755,
      "train_runtime": 7217.5195,
      "train_tokens_per_second": 8519.629
    },
    {
      "epoch": 0.25316455696202533,
      "grad_norm": 1.1127499321455774,
      "learning_rate": 9.610034043218598e-06,
      "loss": 0.4387359619140625,
      "num_input_tokens_seen": 61572512,
      "step": 3760,
      "train_runtime": 7227.0151,
      "train_tokens_per_second": 8519.771
    },
    {
      "epoch": 0.2535012119579855,
      "grad_norm": 1.4079624971388949,
      "learning_rate": 9.609009679704057e-06,
      "loss": 0.45702323913574217,
      "num_input_tokens_seen": 61654432,
      "step": 3765,
      "train_runtime": 7236.5496,
      "train_tokens_per_second": 8519.866
    },
    {
      "epoch": 0.2538378669539456,
      "grad_norm": 1.187077639587616,
      "learning_rate": 9.607984027294654e-06,
      "loss": 0.4184455394744873,
      "num_input_tokens_seen": 61736352,
      "step": 3770,
      "train_runtime": 7246.1552,
      "train_tokens_per_second": 8519.877
    },
    {
      "epoch": 0.25417452194990575,
      "grad_norm": 1.177733123822592,
      "learning_rate": 9.606957086277213e-06,
      "loss": 0.416341495513916,
      "num_input_tokens_seen": 61818272,
      "step": 3775,
      "train_runtime": 7255.6543,
      "train_tokens_per_second": 8520.013
    },
    {
      "epoch": 0.2545111769458659,
      "grad_norm": 1.2405655567024314,
      "learning_rate": 9.605928856938913e-06,
      "loss": 0.41864738464355467,
      "num_input_tokens_seen": 61900192,
      "step": 3780,
      "train_runtime": 7265.1733,
      "train_tokens_per_second": 8520.126
    },
    {
      "epoch": 0.25484783194182603,
      "grad_norm": 0.9474925142716271,
      "learning_rate": 9.604899339567298e-06,
      "loss": 0.39653193950653076,
      "num_input_tokens_seen": 61982112,
      "step": 3785,
      "train_runtime": 7274.7774,
      "train_tokens_per_second": 8520.139
    },
    {
      "epoch": 0.25518448693778617,
      "grad_norm": 1.1613910290775862,
      "learning_rate": 9.603868534450263e-06,
      "loss": 0.3924715042114258,
      "num_input_tokens_seen": 62064032,
      "step": 3790,
      "train_runtime": 7284.3176,
      "train_tokens_per_second": 8520.226
    },
    {
      "epoch": 0.2555211419337463,
      "grad_norm": 1.0904160771476534,
      "learning_rate": 9.602836441876074e-06,
      "loss": 0.40245280265808103,
      "num_input_tokens_seen": 62145952,
      "step": 3795,
      "train_runtime": 7293.8867,
      "train_tokens_per_second": 8520.279
    },
    {
      "epoch": 0.25585779692970645,
      "grad_norm": 1.38303505103627,
      "learning_rate": 9.601803062133353e-06,
      "loss": 0.4203647136688232,
      "num_input_tokens_seen": 62227872,
      "step": 3800,
      "train_runtime": 7303.4973,
      "train_tokens_per_second": 8520.284
    },
    {
      "epoch": 0.2561944519256666,
      "grad_norm": 1.1000647969350579,
      "learning_rate": 9.600768395511078e-06,
      "loss": 0.40354318618774415,
      "num_input_tokens_seen": 62309792,
      "step": 3805,
      "train_runtime": 7313.0489,
      "train_tokens_per_second": 8520.358
    },
    {
      "epoch": 0.25653110692162673,
      "grad_norm": 1.1436161254563497,
      "learning_rate": 9.599732442298592e-06,
      "loss": 0.41320152282714845,
      "num_input_tokens_seen": 62391712,
      "step": 3810,
      "train_runtime": 7322.6397,
      "train_tokens_per_second": 8520.385
    },
    {
      "epoch": 0.25686776191758687,
      "grad_norm": 1.3425398475820527,
      "learning_rate": 9.598695202785596e-06,
      "loss": 0.47039003372192384,
      "num_input_tokens_seen": 62473632,
      "step": 3815,
      "train_runtime": 7332.2426,
      "train_tokens_per_second": 8520.399
    },
    {
      "epoch": 0.257204416913547,
      "grad_norm": 1.163648720526147,
      "learning_rate": 9.59765667726215e-06,
      "loss": 0.38682143688201903,
      "num_input_tokens_seen": 62555552,
      "step": 3820,
      "train_runtime": 7341.8569,
      "train_tokens_per_second": 8520.399
    },
    {
      "epoch": 0.25754107190950715,
      "grad_norm": 1.3115092568202549,
      "learning_rate": 9.596616866018677e-06,
      "loss": 0.48401341438293455,
      "num_input_tokens_seen": 62637472,
      "step": 3825,
      "train_runtime": 7351.3609,
      "train_tokens_per_second": 8520.527
    },
    {
      "epoch": 0.2578777269054673,
      "grad_norm": 1.1046706490904976,
      "learning_rate": 9.595575769345955e-06,
      "loss": 0.46727848052978516,
      "num_input_tokens_seen": 62719392,
      "step": 3830,
      "train_runtime": 7360.9894,
      "train_tokens_per_second": 8520.511
    },
    {
      "epoch": 0.2582143819014274,
      "grad_norm": 1.2412311092783603,
      "learning_rate": 9.594533387535122e-06,
      "loss": 0.4439404487609863,
      "num_input_tokens_seen": 62801312,
      "step": 3835,
      "train_runtime": 7370.5937,
      "train_tokens_per_second": 8520.523
    },
    {
      "epoch": 0.25855103689738757,
      "grad_norm": 1.231565851584679,
      "learning_rate": 9.593489720877678e-06,
      "loss": 0.41535162925720215,
      "num_input_tokens_seen": 62883232,
      "step": 3840,
      "train_runtime": 7380.0974,
      "train_tokens_per_second": 8520.651
    },
    {
      "epoch": 0.2588876918933477,
      "grad_norm": 1.0250654314337735,
      "learning_rate": 9.592444769665482e-06,
      "loss": 0.4233283042907715,
      "num_input_tokens_seen": 62965152,
      "step": 3845,
      "train_runtime": 7389.7083,
      "train_tokens_per_second": 8520.655
    },
    {
      "epoch": 0.25922434688930784,
      "grad_norm": 1.1014427272381033,
      "learning_rate": 9.59139853419075e-06,
      "loss": 0.40645432472229004,
      "num_input_tokens_seen": 63047072,
      "step": 3850,
      "train_runtime": 7399.3278,
      "train_tokens_per_second": 8520.649
    },
    {
      "epoch": 0.259561001885268,
      "grad_norm": 1.0213632259952379,
      "learning_rate": 9.590351014746059e-06,
      "loss": 0.44408321380615234,
      "num_input_tokens_seen": 63128992,
      "step": 3855,
      "train_runtime": 7408.83,
      "train_tokens_per_second": 8520.777
    },
    {
      "epoch": 0.2598976568812281,
      "grad_norm": 1.0175190984365425,
      "learning_rate": 9.589302211624343e-06,
      "loss": 0.40685324668884276,
      "num_input_tokens_seen": 63210912,
      "step": 3860,
      "train_runtime": 7418.4463,
      "train_tokens_per_second": 8520.775
    },
    {
      "epoch": 0.26023431187718826,
      "grad_norm": 1.2178338384634226,
      "learning_rate": 9.5882521251189e-06,
      "loss": 0.42577228546142576,
      "num_input_tokens_seen": 63292832,
      "step": 3865,
      "train_runtime": 7428.0667,
      "train_tokens_per_second": 8520.768
    },
    {
      "epoch": 0.2605709668731484,
      "grad_norm": 1.4149595808565176,
      "learning_rate": 9.58720075552338e-06,
      "loss": 0.4121072769165039,
      "num_input_tokens_seen": 63374752,
      "step": 3870,
      "train_runtime": 7437.6173,
      "train_tokens_per_second": 8520.841
    },
    {
      "epoch": 0.26090762186910854,
      "grad_norm": 1.215060772410176,
      "learning_rate": 9.586148103131797e-06,
      "loss": 0.43161749839782715,
      "num_input_tokens_seen": 63456672,
      "step": 3875,
      "train_runtime": 7447.2035,
      "train_tokens_per_second": 8520.873
    },
    {
      "epoch": 0.2612442768650687,
      "grad_norm": 1.1930541174282898,
      "learning_rate": 9.585094168238518e-06,
      "loss": 0.4709902286529541,
      "num_input_tokens_seen": 63538592,
      "step": 3880,
      "train_runtime": 7456.8368,
      "train_tokens_per_second": 8520.851
    },
    {
      "epoch": 0.2615809318610288,
      "grad_norm": 1.3195166393598576,
      "learning_rate": 9.58403895113828e-06,
      "loss": 0.4241816520690918,
      "num_input_tokens_seen": 63620512,
      "step": 3885,
      "train_runtime": 7466.3899,
      "train_tokens_per_second": 8520.92
    },
    {
      "epoch": 0.26191758685698896,
      "grad_norm": 1.0544957536439592,
      "learning_rate": 9.582982452126164e-06,
      "loss": 0.4121553421020508,
      "num_input_tokens_seen": 63702432,
      "step": 3890,
      "train_runtime": 7475.9661,
      "train_tokens_per_second": 8520.963
    },
    {
      "epoch": 0.2622542418529491,
      "grad_norm": 1.2125482173113729,
      "learning_rate": 9.58192467149762e-06,
      "loss": 0.41678447723388673,
      "num_input_tokens_seen": 63784352,
      "step": 3895,
      "train_runtime": 7485.5825,
      "train_tokens_per_second": 8520.96
    },
    {
      "epoch": 0.26259089684890924,
      "grad_norm": 1.1906033676639929,
      "learning_rate": 9.580865609548452e-06,
      "loss": 0.4092505931854248,
      "num_input_tokens_seen": 63866096,
      "step": 3900,
      "train_runtime": 7495.7169,
      "train_tokens_per_second": 8520.345
    },
    {
      "epoch": 0.2629275518448694,
      "grad_norm": 0.9692629382285417,
      "learning_rate": 9.579805266574824e-06,
      "loss": 0.4171299457550049,
      "num_input_tokens_seen": 63948016,
      "step": 3905,
      "train_runtime": 7505.2193,
      "train_tokens_per_second": 8520.473
    },
    {
      "epoch": 0.2632642068408295,
      "grad_norm": 1.1075332580135842,
      "learning_rate": 9.578743642873258e-06,
      "loss": 0.4110992431640625,
      "num_input_tokens_seen": 64029936,
      "step": 3910,
      "train_runtime": 7514.8295,
      "train_tokens_per_second": 8520.478
    },
    {
      "epoch": 0.26360086183678966,
      "grad_norm": 1.21134705402744,
      "learning_rate": 9.57768073874063e-06,
      "loss": 0.4519134521484375,
      "num_input_tokens_seen": 64111856,
      "step": 3915,
      "train_runtime": 7524.4662,
      "train_tokens_per_second": 8520.452
    },
    {
      "epoch": 0.2639375168327498,
      "grad_norm": 1.0750379500901475,
      "learning_rate": 9.576616554474182e-06,
      "loss": 0.4691943168640137,
      "num_input_tokens_seen": 64193256,
      "step": 3920,
      "train_runtime": 7533.9665,
      "train_tokens_per_second": 8520.513
    },
    {
      "epoch": 0.26427417182870994,
      "grad_norm": 0.96633377939753,
      "learning_rate": 9.575551090371512e-06,
      "loss": 0.4251199722290039,
      "num_input_tokens_seen": 64274912,
      "step": 3925,
      "train_runtime": 7543.5694,
      "train_tokens_per_second": 8520.491
    },
    {
      "epoch": 0.2646108268246701,
      "grad_norm": 1.294559880947049,
      "learning_rate": 9.574484346730564e-06,
      "loss": 0.4682746410369873,
      "num_input_tokens_seen": 64356832,
      "step": 3930,
      "train_runtime": 7553.1954,
      "train_tokens_per_second": 8520.478
    },
    {
      "epoch": 0.2649474818206302,
      "grad_norm": 1.0961734223505728,
      "learning_rate": 9.573416323849658e-06,
      "loss": 0.4373807430267334,
      "num_input_tokens_seen": 64438752,
      "step": 3935,
      "train_runtime": 7562.7663,
      "train_tokens_per_second": 8520.527
    },
    {
      "epoch": 0.26528413681659035,
      "grad_norm": 1.1813566323543354,
      "learning_rate": 9.572347022027461e-06,
      "loss": 0.4322005271911621,
      "num_input_tokens_seen": 64520672,
      "step": 3940,
      "train_runtime": 7572.354,
      "train_tokens_per_second": 8520.557
    },
    {
      "epoch": 0.2656207918125505,
      "grad_norm": 0.9471762530294454,
      "learning_rate": 9.571276441563e-06,
      "loss": 0.40859060287475585,
      "num_input_tokens_seen": 64602592,
      "step": 3945,
      "train_runtime": 7581.9797,
      "train_tokens_per_second": 8520.544
    },
    {
      "epoch": 0.26595744680851063,
      "grad_norm": 1.3049882422287635,
      "learning_rate": 9.570204582755659e-06,
      "loss": 0.43756899833679197,
      "num_input_tokens_seen": 64684512,
      "step": 3950,
      "train_runtime": 7591.5268,
      "train_tokens_per_second": 8520.62
    },
    {
      "epoch": 0.2662941018044708,
      "grad_norm": 1.2523128576859384,
      "learning_rate": 9.569131445905178e-06,
      "loss": 0.44669857025146487,
      "num_input_tokens_seen": 64766344,
      "step": 3955,
      "train_runtime": 7601.6372,
      "train_tokens_per_second": 8520.052
    },
    {
      "epoch": 0.2666307568004309,
      "grad_norm": 1.2167400622662294,
      "learning_rate": 9.568057031311659e-06,
      "loss": 0.4577774524688721,
      "num_input_tokens_seen": 64847792,
      "step": 3960,
      "train_runtime": 7611.5907,
      "train_tokens_per_second": 8519.611
    },
    {
      "epoch": 0.26696741179639105,
      "grad_norm": 1.2501420959664946,
      "learning_rate": 9.566981339275557e-06,
      "loss": 0.44389095306396487,
      "num_input_tokens_seen": 64929712,
      "step": 3965,
      "train_runtime": 7621.2132,
      "train_tokens_per_second": 8519.603
    },
    {
      "epoch": 0.2673040667923512,
      "grad_norm": 0.9767055665561757,
      "learning_rate": 9.565904370097687e-06,
      "loss": 0.40760555267333987,
      "num_input_tokens_seen": 65011632,
      "step": 3970,
      "train_runtime": 7630.7159,
      "train_tokens_per_second": 8519.729
    },
    {
      "epoch": 0.26764072178831133,
      "grad_norm": 1.1709202504856724,
      "learning_rate": 9.564826124079219e-06,
      "loss": 0.3832549571990967,
      "num_input_tokens_seen": 65093552,
      "step": 3975,
      "train_runtime": 7640.3293,
      "train_tokens_per_second": 8519.731
    },
    {
      "epoch": 0.26797737678427147,
      "grad_norm": 1.0649796218573926,
      "learning_rate": 9.563746601521681e-06,
      "loss": 0.42354450225830076,
      "num_input_tokens_seen": 65175472,
      "step": 3980,
      "train_runtime": 7649.9082,
      "train_tokens_per_second": 8519.772
    },
    {
      "epoch": 0.2683140317802316,
      "grad_norm": 1.108330362705277,
      "learning_rate": 9.562665802726961e-06,
      "loss": 0.428800630569458,
      "num_input_tokens_seen": 65257392,
      "step": 3985,
      "train_runtime": 7659.4126,
      "train_tokens_per_second": 8519.895
    },
    {
      "epoch": 0.26865068677619175,
      "grad_norm": 1.272726834712271,
      "learning_rate": 9.561583727997297e-06,
      "loss": 0.44766793251037595,
      "num_input_tokens_seen": 65339312,
      "step": 3990,
      "train_runtime": 7669.0136,
      "train_tokens_per_second": 8519.911
    },
    {
      "epoch": 0.2689873417721519,
      "grad_norm": 1.2428211532431286,
      "learning_rate": 9.560500377635288e-06,
      "loss": 0.42836828231811525,
      "num_input_tokens_seen": 65421232,
      "step": 3995,
      "train_runtime": 7678.6396,
      "train_tokens_per_second": 8519.899
    },
    {
      "epoch": 0.26932399676811203,
      "grad_norm": 1.269118633464097,
      "learning_rate": 9.559415751943893e-06,
      "loss": 0.42889766693115233,
      "num_input_tokens_seen": 65503152,
      "step": 4000,
      "train_runtime": 7688.143,
      "train_tokens_per_second": 8520.023
    },
    {
      "epoch": 0.26966065176407217,
      "grad_norm": 1.329906366254618,
      "learning_rate": 9.558329851226422e-06,
      "loss": 0.41655120849609373,
      "num_input_tokens_seen": 65585072,
      "step": 4005,
      "train_runtime": 7697.7601,
      "train_tokens_per_second": 8520.02
    },
    {
      "epoch": 0.2699973067600323,
      "grad_norm": 1.2095056717886283,
      "learning_rate": 9.557242675786542e-06,
      "loss": 0.4722236156463623,
      "num_input_tokens_seen": 65666992,
      "step": 4010,
      "train_runtime": 7707.3754,
      "train_tokens_per_second": 8520.02
    },
    {
      "epoch": 0.27033396175599245,
      "grad_norm": 1.0994827249276622,
      "learning_rate": 9.55615422592828e-06,
      "loss": 0.4683712005615234,
      "num_input_tokens_seen": 65748912,
      "step": 4015,
      "train_runtime": 7716.8818,
      "train_tokens_per_second": 8520.14
    },
    {
      "epoch": 0.2706706167519526,
      "grad_norm": 1.233692996352648,
      "learning_rate": 9.555064501956018e-06,
      "loss": 0.4237185478210449,
      "num_input_tokens_seen": 65830832,
      "step": 4020,
      "train_runtime": 7726.5026,
      "train_tokens_per_second": 8520.133
    },
    {
      "epoch": 0.2710072717479127,
      "grad_norm": 1.0980833185366423,
      "learning_rate": 9.553973504174493e-06,
      "loss": 0.43896074295043946,
      "num_input_tokens_seen": 65912752,
      "step": 4025,
      "train_runtime": 7736.1088,
      "train_tokens_per_second": 8520.143
    },
    {
      "epoch": 0.27134392674387287,
      "grad_norm": 1.218968881498653,
      "learning_rate": 9.5528812328888e-06,
      "loss": 0.4306613922119141,
      "num_input_tokens_seen": 65994672,
      "step": 4030,
      "train_runtime": 7745.6476,
      "train_tokens_per_second": 8520.227
    },
    {
      "epoch": 0.271680581739833,
      "grad_norm": 1.038422159109327,
      "learning_rate": 9.551787688404386e-06,
      "loss": 0.40329608917236326,
      "num_input_tokens_seen": 66076592,
      "step": 4035,
      "train_runtime": 7755.2155,
      "train_tokens_per_second": 8520.278
    },
    {
      "epoch": 0.27201723673579314,
      "grad_norm": 1.2082514637229116,
      "learning_rate": 9.550692871027061e-06,
      "loss": 0.42394533157348635,
      "num_input_tokens_seen": 66158512,
      "step": 4040,
      "train_runtime": 7764.8312,
      "train_tokens_per_second": 8520.277
    },
    {
      "epoch": 0.2723538917317533,
      "grad_norm": 1.1433895980437159,
      "learning_rate": 9.549596781062984e-06,
      "loss": 0.4099553108215332,
      "num_input_tokens_seen": 66240432,
      "step": 4045,
      "train_runtime": 7774.3821,
      "train_tokens_per_second": 8520.347
    },
    {
      "epoch": 0.2726905467277134,
      "grad_norm": 1.0450078000446643,
      "learning_rate": 9.548499418818677e-06,
      "loss": 0.3902336597442627,
      "num_input_tokens_seen": 66322352,
      "step": 4050,
      "train_runtime": 7783.9548,
      "train_tokens_per_second": 8520.393
    },
    {
      "epoch": 0.27302720172367356,
      "grad_norm": 1.1572904194317595,
      "learning_rate": 9.547400784601011e-06,
      "loss": 0.43136157989501955,
      "num_input_tokens_seen": 66404272,
      "step": 4055,
      "train_runtime": 7793.565,
      "train_tokens_per_second": 8520.398
    },
    {
      "epoch": 0.2733638567196337,
      "grad_norm": 1.1500949787595276,
      "learning_rate": 9.546300878717216e-06,
      "loss": 0.48661150932312014,
      "num_input_tokens_seen": 66486192,
      "step": 4060,
      "train_runtime": 7803.1716,
      "train_tokens_per_second": 8520.406
    },
    {
      "epoch": 0.27370051171559384,
      "grad_norm": 1.091195524630376,
      "learning_rate": 9.545199701474877e-06,
      "loss": 0.4279643058776855,
      "num_input_tokens_seen": 66568112,
      "step": 4065,
      "train_runtime": 7812.674,
      "train_tokens_per_second": 8520.529
    },
    {
      "epoch": 0.274037166711554,
      "grad_norm": 1.1260260781004428,
      "learning_rate": 9.544097253181935e-06,
      "loss": 0.37764673233032225,
      "num_input_tokens_seen": 66650032,
      "step": 4070,
      "train_runtime": 7822.2647,
      "train_tokens_per_second": 8520.554
    },
    {
      "epoch": 0.2743738217075141,
      "grad_norm": 1.2312372750716325,
      "learning_rate": 9.542993534146687e-06,
      "loss": 0.44744071960449217,
      "num_input_tokens_seen": 66731952,
      "step": 4075,
      "train_runtime": 7831.8628,
      "train_tokens_per_second": 8520.572
    },
    {
      "epoch": 0.27471047670347426,
      "grad_norm": 1.2665949537458696,
      "learning_rate": 9.541888544677784e-06,
      "loss": 0.4608957767486572,
      "num_input_tokens_seen": 66813872,
      "step": 4080,
      "train_runtime": 7841.3647,
      "train_tokens_per_second": 8520.694
    },
    {
      "epoch": 0.2750471316994344,
      "grad_norm": 1.0788678267650693,
      "learning_rate": 9.540782285084232e-06,
      "loss": 0.36507301330566405,
      "num_input_tokens_seen": 66895792,
      "step": 4085,
      "train_runtime": 7850.9786,
      "train_tokens_per_second": 8520.695
    },
    {
      "epoch": 0.27538378669539454,
      "grad_norm": 1.0567040899805318,
      "learning_rate": 9.539674755675392e-06,
      "loss": 0.40845499038696287,
      "num_input_tokens_seen": 66977712,
      "step": 4090,
      "train_runtime": 7860.6054,
      "train_tokens_per_second": 8520.681
    },
    {
      "epoch": 0.2757204416913547,
      "grad_norm": 1.5104152475038795,
      "learning_rate": 9.538565956760983e-06,
      "loss": 0.3775378465652466,
      "num_input_tokens_seen": 67059632,
      "step": 4095,
      "train_runtime": 7870.11,
      "train_tokens_per_second": 8520.8
    },
    {
      "epoch": 0.2760570966873148,
      "grad_norm": 1.1679814593091349,
      "learning_rate": 9.537455888651078e-06,
      "loss": 0.4161814212799072,
      "num_input_tokens_seen": 67141552,
      "step": 4100,
      "train_runtime": 7879.6932,
      "train_tokens_per_second": 8520.833
    },
    {
      "epoch": 0.27639375168327496,
      "grad_norm": 1.1180395306673372,
      "learning_rate": 9.536344551656103e-06,
      "loss": 0.4082015037536621,
      "num_input_tokens_seen": 67223472,
      "step": 4105,
      "train_runtime": 7889.2885,
      "train_tokens_per_second": 8520.854
    },
    {
      "epoch": 0.2767304066792351,
      "grad_norm": 1.1765362097850776,
      "learning_rate": 9.535231946086838e-06,
      "loss": 0.4095332145690918,
      "num_input_tokens_seen": 67305392,
      "step": 4110,
      "train_runtime": 7898.8192,
      "train_tokens_per_second": 8520.943
    },
    {
      "epoch": 0.27706706167519524,
      "grad_norm": 1.1287932166488706,
      "learning_rate": 9.534118072254421e-06,
      "loss": 0.4064137935638428,
      "num_input_tokens_seen": 67387312,
      "step": 4115,
      "train_runtime": 7908.6804,
      "train_tokens_per_second": 8520.677
    },
    {
      "epoch": 0.2774037166711554,
      "grad_norm": 0.9514579116238078,
      "learning_rate": 9.533002930470345e-06,
      "loss": 0.41384477615356446,
      "num_input_tokens_seen": 67469232,
      "step": 4120,
      "train_runtime": 7918.2821,
      "train_tokens_per_second": 8520.691
    },
    {
      "epoch": 0.2777403716671155,
      "grad_norm": 1.1491809829316142,
      "learning_rate": 9.531886521046452e-06,
      "loss": 0.389455771446228,
      "num_input_tokens_seen": 67551152,
      "step": 4125,
      "train_runtime": 7928.0447,
      "train_tokens_per_second": 8520.531
    },
    {
      "epoch": 0.27807702666307565,
      "grad_norm": 1.2224785974285397,
      "learning_rate": 9.530768844294947e-06,
      "loss": 0.4553849220275879,
      "num_input_tokens_seen": 67633072,
      "step": 4130,
      "train_runtime": 7937.8494,
      "train_tokens_per_second": 8520.327
    },
    {
      "epoch": 0.2784136816590358,
      "grad_norm": 1.1036585762955724,
      "learning_rate": 9.52964990052838e-06,
      "loss": 0.44520020484924316,
      "num_input_tokens_seen": 67714992,
      "step": 4135,
      "train_runtime": 7947.7665,
      "train_tokens_per_second": 8520.003
    },
    {
      "epoch": 0.27875033665499593,
      "grad_norm": 1.0987024910188143,
      "learning_rate": 9.528529690059663e-06,
      "loss": 0.40784530639648436,
      "num_input_tokens_seen": 67796912,
      "step": 4140,
      "train_runtime": 7957.3194,
      "train_tokens_per_second": 8520.069
    },
    {
      "epoch": 0.2790869916509561,
      "grad_norm": 1.226935831572079,
      "learning_rate": 9.52740821320206e-06,
      "loss": 0.45844221115112305,
      "num_input_tokens_seen": 67878832,
      "step": 4145,
      "train_runtime": 7966.8922,
      "train_tokens_per_second": 8520.114
    },
    {
      "epoch": 0.2794236466469162,
      "grad_norm": 1.3168622972911728,
      "learning_rate": 9.526285470269185e-06,
      "loss": 0.49367728233337405,
      "num_input_tokens_seen": 67960752,
      "step": 4150,
      "train_runtime": 7976.4942,
      "train_tokens_per_second": 8520.128
    },
    {
      "epoch": 0.2797603016428764,
      "grad_norm": 1.1945452601148263,
      "learning_rate": 9.525161461575012e-06,
      "loss": 0.4282985210418701,
      "num_input_tokens_seen": 68042360,
      "step": 4155,
      "train_runtime": 7986.4555,
      "train_tokens_per_second": 8519.719
    },
    {
      "epoch": 0.28009695663883655,
      "grad_norm": 1.101984036014093,
      "learning_rate": 9.524036187433867e-06,
      "loss": 0.3911474943161011,
      "num_input_tokens_seen": 68124280,
      "step": 4160,
      "train_runtime": 7996.019,
      "train_tokens_per_second": 8519.775
    },
    {
      "epoch": 0.2804336116347967,
      "grad_norm": 1.0438715945400499,
      "learning_rate": 9.522909648160427e-06,
      "loss": 0.38232200145721434,
      "num_input_tokens_seen": 68206200,
      "step": 4165,
      "train_runtime": 8005.6303,
      "train_tokens_per_second": 8519.779
    },
    {
      "epoch": 0.2807702666307568,
      "grad_norm": 1.110536123441758,
      "learning_rate": 9.521781844069727e-06,
      "loss": 0.4204099655151367,
      "num_input_tokens_seen": 68288120,
      "step": 4170,
      "train_runtime": 8015.2514,
      "train_tokens_per_second": 8519.773
    },
    {
      "epoch": 0.28110692162671697,
      "grad_norm": 1.2515993806026682,
      "learning_rate": 9.520652775477154e-06,
      "loss": 0.454621696472168,
      "num_input_tokens_seen": 68370040,
      "step": 4175,
      "train_runtime": 8024.7516,
      "train_tokens_per_second": 8519.895
    },
    {
      "epoch": 0.2814435766226771,
      "grad_norm": 1.035642571417639,
      "learning_rate": 9.519522442698447e-06,
      "loss": 0.38779397010803224,
      "num_input_tokens_seen": 68451960,
      "step": 4180,
      "train_runtime": 8034.3781,
      "train_tokens_per_second": 8519.883
    },
    {
      "epoch": 0.28178023161863724,
      "grad_norm": 1.5401430550193265,
      "learning_rate": 9.518390846049699e-06,
      "loss": 0.42632169723510743,
      "num_input_tokens_seen": 68533880,
      "step": 4185,
      "train_runtime": 8044.0014,
      "train_tokens_per_second": 8519.874
    },
    {
      "epoch": 0.2821168866145974,
      "grad_norm": 1.1003549336446483,
      "learning_rate": 9.517257985847361e-06,
      "loss": 0.4484898567199707,
      "num_input_tokens_seen": 68615800,
      "step": 4190,
      "train_runtime": 8053.4965,
      "train_tokens_per_second": 8520.001
    },
    {
      "epoch": 0.2824535416105575,
      "grad_norm": 1.1582757060790017,
      "learning_rate": 9.516123862408232e-06,
      "loss": 0.47380504608154295,
      "num_input_tokens_seen": 68697720,
      "step": 4195,
      "train_runtime": 8063.1198,
      "train_tokens_per_second": 8519.993
    },
    {
      "epoch": 0.28279019660651766,
      "grad_norm": 1.3195691629005468,
      "learning_rate": 9.514988476049466e-06,
      "loss": 0.42533278465270996,
      "num_input_tokens_seen": 68779640,
      "step": 4200,
      "train_runtime": 8072.7259,
      "train_tokens_per_second": 8520.002
    },
    {
      "epoch": 0.2831268516024778,
      "grad_norm": 1.0345433465101306,
      "learning_rate": 9.513851827088567e-06,
      "loss": 0.4271504878997803,
      "num_input_tokens_seen": 68861560,
      "step": 4205,
      "train_runtime": 8082.2191,
      "train_tokens_per_second": 8520.13
    },
    {
      "epoch": 0.28346350659843794,
      "grad_norm": 1.2519965474623007,
      "learning_rate": 9.512713915843402e-06,
      "loss": 0.4214357376098633,
      "num_input_tokens_seen": 68943480,
      "step": 4210,
      "train_runtime": 8091.848,
      "train_tokens_per_second": 8520.116
    },
    {
      "epoch": 0.2838001615943981,
      "grad_norm": 0.9984089449263305,
      "learning_rate": 9.511574742632177e-06,
      "loss": 0.4165010452270508,
      "num_input_tokens_seen": 69025400,
      "step": 4215,
      "train_runtime": 8101.4508,
      "train_tokens_per_second": 8520.128
    },
    {
      "epoch": 0.2841368165903582,
      "grad_norm": 1.266510295780462,
      "learning_rate": 9.510434307773464e-06,
      "loss": 0.45716071128845215,
      "num_input_tokens_seen": 69107320,
      "step": 4220,
      "train_runtime": 8110.9954,
      "train_tokens_per_second": 8520.202
    },
    {
      "epoch": 0.28447347158631836,
      "grad_norm": 0.9518541865482143,
      "learning_rate": 9.509292611586179e-06,
      "loss": 0.3699838638305664,
      "num_input_tokens_seen": 69189240,
      "step": 4225,
      "train_runtime": 8120.568,
      "train_tokens_per_second": 8520.246
    },
    {
      "epoch": 0.2848101265822785,
      "grad_norm": 1.3070615798438059,
      "learning_rate": 9.508149654389592e-06,
      "loss": 0.4135759353637695,
      "num_input_tokens_seen": 69271160,
      "step": 4230,
      "train_runtime": 8130.2044,
      "train_tokens_per_second": 8520.224
    },
    {
      "epoch": 0.28514678157823864,
      "grad_norm": 0.9528517204192211,
      "learning_rate": 9.507005436503331e-06,
      "loss": 0.3646747827529907,
      "num_input_tokens_seen": 69353080,
      "step": 4235,
      "train_runtime": 8139.7478,
      "train_tokens_per_second": 8520.298
    },
    {
      "epoch": 0.2854834365741988,
      "grad_norm": 1.0893790614231875,
      "learning_rate": 9.505859958247373e-06,
      "loss": 0.4230369567871094,
      "num_input_tokens_seen": 69435000,
      "step": 4240,
      "train_runtime": 8149.3103,
      "train_tokens_per_second": 8520.353
    },
    {
      "epoch": 0.2858200915701589,
      "grad_norm": 1.2521866110497555,
      "learning_rate": 9.504713219942045e-06,
      "loss": 0.41087875366210935,
      "num_input_tokens_seen": 69516920,
      "step": 4245,
      "train_runtime": 8158.9056,
      "train_tokens_per_second": 8520.373
    },
    {
      "epoch": 0.28615674656611906,
      "grad_norm": 1.309753836043401,
      "learning_rate": 9.503565221908029e-06,
      "loss": 0.40329618453979493,
      "num_input_tokens_seen": 69598840,
      "step": 4250,
      "train_runtime": 8168.5218,
      "train_tokens_per_second": 8520.371
    },
    {
      "epoch": 0.2864934015620792,
      "grad_norm": 0.9956711852016557,
      "learning_rate": 9.50241596446636e-06,
      "loss": 0.4689029693603516,
      "num_input_tokens_seen": 69680760,
      "step": 4255,
      "train_runtime": 8178.0222,
      "train_tokens_per_second": 8520.49
    },
    {
      "epoch": 0.28683005655803934,
      "grad_norm": 1.1043184936194501,
      "learning_rate": 9.501265447938423e-06,
      "loss": 0.3807155847549438,
      "num_input_tokens_seen": 69762680,
      "step": 4260,
      "train_runtime": 8187.6295,
      "train_tokens_per_second": 8520.498
    },
    {
      "epoch": 0.2871667115539995,
      "grad_norm": 1.099800958676872,
      "learning_rate": 9.500113672645958e-06,
      "loss": 0.41358633041381837,
      "num_input_tokens_seen": 69844600,
      "step": 4265,
      "train_runtime": 8197.2328,
      "train_tokens_per_second": 8520.509
    },
    {
      "epoch": 0.2875033665499596,
      "grad_norm": 1.0991084703420129,
      "learning_rate": 9.498960638911054e-06,
      "loss": 0.37629601955413816,
      "num_input_tokens_seen": 69926520,
      "step": 4270,
      "train_runtime": 8206.7316,
      "train_tokens_per_second": 8520.63
    },
    {
      "epoch": 0.28784002154591976,
      "grad_norm": 1.2189237205238552,
      "learning_rate": 9.497806347056153e-06,
      "loss": 0.420274019241333,
      "num_input_tokens_seen": 70008440,
      "step": 4275,
      "train_runtime": 8216.3383,
      "train_tokens_per_second": 8520.637
    },
    {
      "epoch": 0.2881766765418799,
      "grad_norm": 1.301009668544247,
      "learning_rate": 9.49665079740405e-06,
      "loss": 0.46674289703369143,
      "num_input_tokens_seen": 70090360,
      "step": 4280,
      "train_runtime": 8225.9568,
      "train_tokens_per_second": 8520.633
    },
    {
      "epoch": 0.28851333153784003,
      "grad_norm": 1.5262505904277404,
      "learning_rate": 9.495493990277889e-06,
      "loss": 0.45811662673950193,
      "num_input_tokens_seen": 70172064,
      "step": 4285,
      "train_runtime": 8235.4536,
      "train_tokens_per_second": 8520.728
    },
    {
      "epoch": 0.2888499865338002,
      "grad_norm": 1.2453655649919084,
      "learning_rate": 9.49433592600117e-06,
      "loss": 0.48780527114868166,
      "num_input_tokens_seen": 70253984,
      "step": 4290,
      "train_runtime": 8245.0437,
      "train_tokens_per_second": 8520.753
    },
    {
      "epoch": 0.2891866415297603,
      "grad_norm": 1.0720656606292307,
      "learning_rate": 9.49317660489774e-06,
      "loss": 0.443308162689209,
      "num_input_tokens_seen": 70335904,
      "step": 4295,
      "train_runtime": 8254.5646,
      "train_tokens_per_second": 8520.85
    },
    {
      "epoch": 0.28952329652572045,
      "grad_norm": 1.3307545461777348,
      "learning_rate": 9.4920160272918e-06,
      "loss": 0.4324596405029297,
      "num_input_tokens_seen": 70417824,
      "step": 4300,
      "train_runtime": 8264.0636,
      "train_tokens_per_second": 8520.968
    },
    {
      "epoch": 0.2898599515216806,
      "grad_norm": 1.1925718765879,
      "learning_rate": 9.490854193507904e-06,
      "loss": 0.4282393932342529,
      "num_input_tokens_seen": 70499744,
      "step": 4305,
      "train_runtime": 8273.6696,
      "train_tokens_per_second": 8520.976
    },
    {
      "epoch": 0.29019660651764073,
      "grad_norm": 1.229591468282557,
      "learning_rate": 9.489691103870951e-06,
      "loss": 0.4300379753112793,
      "num_input_tokens_seen": 70581664,
      "step": 4310,
      "train_runtime": 8283.2573,
      "train_tokens_per_second": 8521.003
    },
    {
      "epoch": 0.29053326151360087,
      "grad_norm": 1.0356918306606824,
      "learning_rate": 9.488526758706198e-06,
      "loss": 0.4282421112060547,
      "num_input_tokens_seen": 70663584,
      "step": 4315,
      "train_runtime": 8292.7997,
      "train_tokens_per_second": 8521.077
    },
    {
      "epoch": 0.290869916509561,
      "grad_norm": 1.2118327417205417,
      "learning_rate": 9.48736115833925e-06,
      "loss": 0.4536780834197998,
      "num_input_tokens_seen": 70745504,
      "step": 4320,
      "train_runtime": 8302.3657,
      "train_tokens_per_second": 8521.126
    },
    {
      "epoch": 0.29120657150552115,
      "grad_norm": 1.234605354085704,
      "learning_rate": 9.486194303096062e-06,
      "loss": 0.4153404712677002,
      "num_input_tokens_seen": 70827424,
      "step": 4325,
      "train_runtime": 8311.9897,
      "train_tokens_per_second": 8521.116
    },
    {
      "epoch": 0.2915432265014813,
      "grad_norm": 1.2621759088313844,
      "learning_rate": 9.485026193302945e-06,
      "loss": 0.39893364906311035,
      "num_input_tokens_seen": 70909344,
      "step": 4330,
      "train_runtime": 8321.5404,
      "train_tokens_per_second": 8521.18
    },
    {
      "epoch": 0.29187988149744143,
      "grad_norm": 1.0500468152947084,
      "learning_rate": 9.483856829286552e-06,
      "loss": 0.45873279571533204,
      "num_input_tokens_seen": 70991264,
      "step": 4335,
      "train_runtime": 8331.0923,
      "train_tokens_per_second": 8521.243
    },
    {
      "epoch": 0.29221653649340157,
      "grad_norm": 1.1101721258004036,
      "learning_rate": 9.482686211373896e-06,
      "loss": 0.47636189460754397,
      "num_input_tokens_seen": 71073184,
      "step": 4340,
      "train_runtime": 8340.6941,
      "train_tokens_per_second": 8521.255
    },
    {
      "epoch": 0.2925531914893617,
      "grad_norm": 1.1315720633779751,
      "learning_rate": 9.481514339892335e-06,
      "loss": 0.4131361484527588,
      "num_input_tokens_seen": 71155104,
      "step": 4345,
      "train_runtime": 8350.2901,
      "train_tokens_per_second": 8521.273
    },
    {
      "epoch": 0.29288984648532185,
      "grad_norm": 1.1945546688232216,
      "learning_rate": 9.48034121516958e-06,
      "loss": 0.4098196983337402,
      "num_input_tokens_seen": 71237024,
      "step": 4350,
      "train_runtime": 8359.7906,
      "train_tokens_per_second": 8521.389
    },
    {
      "epoch": 0.293226501481282,
      "grad_norm": 1.3602837340095226,
      "learning_rate": 9.47916683753369e-06,
      "loss": 0.45661320686340334,
      "num_input_tokens_seen": 71318944,
      "step": 4355,
      "train_runtime": 8369.4092,
      "train_tokens_per_second": 8521.383
    },
    {
      "epoch": 0.2935631564772421,
      "grad_norm": 1.313020495974058,
      "learning_rate": 9.477991207313077e-06,
      "loss": 0.42633466720581054,
      "num_input_tokens_seen": 71400864,
      "step": 4360,
      "train_runtime": 8379.0115,
      "train_tokens_per_second": 8521.395
    },
    {
      "epoch": 0.29389981147320227,
      "grad_norm": 1.0798765463857234,
      "learning_rate": 9.476814324836504e-06,
      "loss": 0.40662622451782227,
      "num_input_tokens_seen": 71482784,
      "step": 4365,
      "train_runtime": 8388.5063,
      "train_tokens_per_second": 8521.515
    },
    {
      "epoch": 0.2942364664691624,
      "grad_norm": 1.0979442452963588,
      "learning_rate": 9.475636190433078e-06,
      "loss": 0.443254280090332,
      "num_input_tokens_seen": 71564704,
      "step": 4370,
      "train_runtime": 8398.0925,
      "train_tokens_per_second": 8521.543
    },
    {
      "epoch": 0.29457312146512254,
      "grad_norm": 1.1309241415127171,
      "learning_rate": 9.474456804432264e-06,
      "loss": 0.4555081367492676,
      "num_input_tokens_seen": 71646624,
      "step": 4375,
      "train_runtime": 8407.7147,
      "train_tokens_per_second": 8521.534
    },
    {
      "epoch": 0.2949097764610827,
      "grad_norm": 1.3838899885424363,
      "learning_rate": 9.473276167163872e-06,
      "loss": 0.4588156700134277,
      "num_input_tokens_seen": 71728544,
      "step": 4380,
      "train_runtime": 8417.2198,
      "train_tokens_per_second": 8521.643
    },
    {
      "epoch": 0.2952464314570428,
      "grad_norm": 1.0945991102093888,
      "learning_rate": 9.472094278958066e-06,
      "loss": 0.4797795295715332,
      "num_input_tokens_seen": 71810464,
      "step": 4385,
      "train_runtime": 8426.8265,
      "train_tokens_per_second": 8521.65
    },
    {
      "epoch": 0.29558308645300296,
      "grad_norm": 1.1517728322433805,
      "learning_rate": 9.470911140145353e-06,
      "loss": 0.44533214569091795,
      "num_input_tokens_seen": 71892384,
      "step": 4390,
      "train_runtime": 8436.446,
      "train_tokens_per_second": 8521.643
    },
    {
      "epoch": 0.2959197414489631,
      "grad_norm": 1.2764403752062856,
      "learning_rate": 9.469726751056599e-06,
      "loss": 0.43787946701049807,
      "num_input_tokens_seen": 71974304,
      "step": 4395,
      "train_runtime": 8445.9946,
      "train_tokens_per_second": 8521.709
    },
    {
      "epoch": 0.29625639644492324,
      "grad_norm": 1.3536329925944859,
      "learning_rate": 9.468541112023009e-06,
      "loss": 0.43798513412475587,
      "num_input_tokens_seen": 72056224,
      "step": 4400,
      "train_runtime": 8455.5544,
      "train_tokens_per_second": 8521.762
    },
    {
      "epoch": 0.2965930514408834,
      "grad_norm": 1.0240230708718463,
      "learning_rate": 9.467354223376149e-06,
      "loss": 0.42894835472106935,
      "num_input_tokens_seen": 72138144,
      "step": 4405,
      "train_runtime": 8465.1175,
      "train_tokens_per_second": 8521.812
    },
    {
      "epoch": 0.2969297064368435,
      "grad_norm": 1.1113414164449542,
      "learning_rate": 9.466166085447923e-06,
      "loss": 0.39919605255126955,
      "num_input_tokens_seen": 72220064,
      "step": 4410,
      "train_runtime": 8474.6741,
      "train_tokens_per_second": 8521.869
    },
    {
      "epoch": 0.29726636143280366,
      "grad_norm": 1.1168378497992868,
      "learning_rate": 9.464976698570595e-06,
      "loss": 0.4007763385772705,
      "num_input_tokens_seen": 72301152,
      "step": 4415,
      "train_runtime": 8484.4177,
      "train_tokens_per_second": 8521.64
    },
    {
      "epoch": 0.2976030164287638,
      "grad_norm": 1.1947572335623229,
      "learning_rate": 9.463786063076769e-06,
      "loss": 0.4543807029724121,
      "num_input_tokens_seen": 72383072,
      "step": 4420,
      "train_runtime": 8494.0219,
      "train_tokens_per_second": 8521.649
    },
    {
      "epoch": 0.29793967142472394,
      "grad_norm": 1.0650829109848998,
      "learning_rate": 9.462594179299408e-06,
      "loss": 0.3524836540222168,
      "num_input_tokens_seen": 72464992,
      "step": 4425,
      "train_runtime": 8503.6205,
      "train_tokens_per_second": 8521.663
    },
    {
      "epoch": 0.2982763264206841,
      "grad_norm": 1.1672911332466616,
      "learning_rate": 9.461401047571811e-06,
      "loss": 0.40203056335449217,
      "num_input_tokens_seen": 72546912,
      "step": 4430,
      "train_runtime": 8513.1546,
      "train_tokens_per_second": 8521.743
    },
    {
      "epoch": 0.2986129814166442,
      "grad_norm": 1.181964953989582,
      "learning_rate": 9.460206668227639e-06,
      "loss": 0.4103087902069092,
      "num_input_tokens_seen": 72628832,
      "step": 4435,
      "train_runtime": 8522.7146,
      "train_tokens_per_second": 8521.796
    },
    {
      "epoch": 0.29894963641260436,
      "grad_norm": 0.9561388330859738,
      "learning_rate": 9.459011041600895e-06,
      "loss": 0.43447580337524416,
      "num_input_tokens_seen": 72710752,
      "step": 4440,
      "train_runtime": 8532.3282,
      "train_tokens_per_second": 8521.795
    },
    {
      "epoch": 0.2992862914085645,
      "grad_norm": 1.1570482073192936,
      "learning_rate": 9.457814168025932e-06,
      "loss": 0.4091656684875488,
      "num_input_tokens_seen": 72792672,
      "step": 4445,
      "train_runtime": 8541.9335,
      "train_tokens_per_second": 8521.803
    },
    {
      "epoch": 0.29962294640452464,
      "grad_norm": 1.2029584619870006,
      "learning_rate": 9.456616047837452e-06,
      "loss": 0.4083125591278076,
      "num_input_tokens_seen": 72874544,
      "step": 4450,
      "train_runtime": 8552.651,
      "train_tokens_per_second": 8520.697
    },
    {
      "epoch": 0.2999596014004848,
      "grad_norm": 1.1454853786555041,
      "learning_rate": 9.455416681370506e-06,
      "loss": 0.4855609893798828,
      "num_input_tokens_seen": 72956464,
      "step": 4455,
      "train_runtime": 8563.1944,
      "train_tokens_per_second": 8519.772
    },
    {
      "epoch": 0.3002962563964449,
      "grad_norm": 1.3292732405036685,
      "learning_rate": 9.45421606896049e-06,
      "loss": 0.4704150676727295,
      "num_input_tokens_seen": 73038384,
      "step": 4460,
      "train_runtime": 8574.3305,
      "train_tokens_per_second": 8518.261
    },
    {
      "epoch": 0.30063291139240506,
      "grad_norm": 1.0488907380643748,
      "learning_rate": 9.453014210943155e-06,
      "loss": 0.4177052974700928,
      "num_input_tokens_seen": 73120304,
      "step": 4465,
      "train_runtime": 8585.3894,
      "train_tokens_per_second": 8516.83
    },
    {
      "epoch": 0.3009695663883652,
      "grad_norm": 1.1940709502800941,
      "learning_rate": 9.451811107654596e-06,
      "loss": 0.4119865417480469,
      "num_input_tokens_seen": 73202224,
      "step": 4470,
      "train_runtime": 8596.4324,
      "train_tokens_per_second": 8515.419
    },
    {
      "epoch": 0.30130622138432533,
      "grad_norm": 1.378252529966046,
      "learning_rate": 9.450606759431255e-06,
      "loss": 0.43073692321777346,
      "num_input_tokens_seen": 73284144,
      "step": 4475,
      "train_runtime": 8607.6446,
      "train_tokens_per_second": 8513.844
    },
    {
      "epoch": 0.3016428763802855,
      "grad_norm": 1.1471110344423427,
      "learning_rate": 9.449401166609928e-06,
      "loss": 0.45232439041137695,
      "num_input_tokens_seen": 73366064,
      "step": 4480,
      "train_runtime": 8617.9766,
      "train_tokens_per_second": 8513.143
    },
    {
      "epoch": 0.3019795313762456,
      "grad_norm": 1.2114568035335318,
      "learning_rate": 9.448194329527752e-06,
      "loss": 0.4222346305847168,
      "num_input_tokens_seen": 73447984,
      "step": 4485,
      "train_runtime": 8627.4808,
      "train_tokens_per_second": 8513.26
    },
    {
      "epoch": 0.30231618637220575,
      "grad_norm": 1.2107262971211135,
      "learning_rate": 9.446986248522216e-06,
      "loss": 0.4250503063201904,
      "num_input_tokens_seen": 73529904,
      "step": 4490,
      "train_runtime": 8636.9819,
      "train_tokens_per_second": 8513.379
    },
    {
      "epoch": 0.3026528413681659,
      "grad_norm": 1.1475854063040676,
      "learning_rate": 9.445776923931157e-06,
      "loss": 0.40310211181640626,
      "num_input_tokens_seen": 73611824,
      "step": 4495,
      "train_runtime": 8646.4796,
      "train_tokens_per_second": 8513.502
    },
    {
      "epoch": 0.30298949636412603,
      "grad_norm": 1.1231482828569594,
      "learning_rate": 9.444566356092754e-06,
      "loss": 0.45733957290649413,
      "num_input_tokens_seen": 73693744,
      "step": 4500,
      "train_runtime": 8655.9827,
      "train_tokens_per_second": 8513.62
    },
    {
      "epoch": 0.30332615136008617,
      "grad_norm": 1.1197875383743112,
      "learning_rate": 9.443354545345545e-06,
      "loss": 0.4003597736358643,
      "num_input_tokens_seen": 73775664,
      "step": 4505,
      "train_runtime": 8665.5005,
      "train_tokens_per_second": 8513.722
    },
    {
      "epoch": 0.3036628063560463,
      "grad_norm": 0.9632339949767456,
      "learning_rate": 9.442141492028406e-06,
      "loss": 0.39724030494689944,
      "num_input_tokens_seen": 73857584,
      "step": 4510,
      "train_runtime": 8675.0099,
      "train_tokens_per_second": 8513.833
    },
    {
      "epoch": 0.30399946135200645,
      "grad_norm": 1.1698528948385651,
      "learning_rate": 9.440927196480563e-06,
      "loss": 0.39248526096343994,
      "num_input_tokens_seen": 73939504,
      "step": 4515,
      "train_runtime": 8684.5178,
      "train_tokens_per_second": 8513.945
    },
    {
      "epoch": 0.3043361163479666,
      "grad_norm": 0.967922447476968,
      "learning_rate": 9.439711659041593e-06,
      "loss": 0.3996472120285034,
      "num_input_tokens_seen": 74021424,
      "step": 4520,
      "train_runtime": 8694.0247,
      "train_tokens_per_second": 8514.057
    },
    {
      "epoch": 0.30467277134392673,
      "grad_norm": 1.10122725112793,
      "learning_rate": 9.438494880051413e-06,
      "loss": 0.43231678009033203,
      "num_input_tokens_seen": 74103344,
      "step": 4525,
      "train_runtime": 8703.5308,
      "train_tokens_per_second": 8514.17
    },
    {
      "epoch": 0.30500942633988687,
      "grad_norm": 1.093716415683515,
      "learning_rate": 9.437276859850293e-06,
      "loss": 0.44533534049987794,
      "num_input_tokens_seen": 74185264,
      "step": 4530,
      "train_runtime": 8713.0372,
      "train_tokens_per_second": 8514.283
    },
    {
      "epoch": 0.305346081335847,
      "grad_norm": 1.410241391076216,
      "learning_rate": 9.436057598778848e-06,
      "loss": 0.4771921157836914,
      "num_input_tokens_seen": 74267184,
      "step": 4535,
      "train_runtime": 8722.5456,
      "train_tokens_per_second": 8514.393
    },
    {
      "epoch": 0.30568273633180715,
      "grad_norm": 1.2930236714168413,
      "learning_rate": 9.434837097178043e-06,
      "loss": 0.4697895050048828,
      "num_input_tokens_seen": 74349104,
      "step": 4540,
      "train_runtime": 8732.0448,
      "train_tokens_per_second": 8514.512
    },
    {
      "epoch": 0.3060193913277673,
      "grad_norm": 1.281202075515235,
      "learning_rate": 9.433615355389183e-06,
      "loss": 0.4238016128540039,
      "num_input_tokens_seen": 74431024,
      "step": 4545,
      "train_runtime": 8741.5479,
      "train_tokens_per_second": 8514.628
    },
    {
      "epoch": 0.3063560463237274,
      "grad_norm": 1.1247643508011576,
      "learning_rate": 9.432392373753926e-06,
      "loss": 0.45804462432861326,
      "num_input_tokens_seen": 74512944,
      "step": 4550,
      "train_runtime": 8751.047,
      "train_tokens_per_second": 8514.746
    },
    {
      "epoch": 0.30669270131968757,
      "grad_norm": 1.0888612829623496,
      "learning_rate": 9.431168152614278e-06,
      "loss": 0.4204387187957764,
      "num_input_tokens_seen": 74594864,
      "step": 4555,
      "train_runtime": 8760.5579,
      "train_tokens_per_second": 8514.853
    },
    {
      "epoch": 0.3070293563156477,
      "grad_norm": 1.2283509751876944,
      "learning_rate": 9.429942692312585e-06,
      "loss": 0.44479918479919434,
      "num_input_tokens_seen": 74676784,
      "step": 4560,
      "train_runtime": 8770.0631,
      "train_tokens_per_second": 8514.965
    },
    {
      "epoch": 0.30736601131160785,
      "grad_norm": 2.0176496423991668,
      "learning_rate": 9.428715993191546e-06,
      "loss": 0.4067281723022461,
      "num_input_tokens_seen": 74758704,
      "step": 4565,
      "train_runtime": 8779.5724,
      "train_tokens_per_second": 8515.073
    },
    {
      "epoch": 0.307702666307568,
      "grad_norm": 1.3074500326591387,
      "learning_rate": 9.427488055594199e-06,
      "loss": 0.41740713119506834,
      "num_input_tokens_seen": 74840624,
      "step": 4570,
      "train_runtime": 8789.0761,
      "train_tokens_per_second": 8515.187
    },
    {
      "epoch": 0.3080393213035281,
      "grad_norm": 1.1774100525908313,
      "learning_rate": 9.426258879863937e-06,
      "loss": 0.4435271263122559,
      "num_input_tokens_seen": 74922544,
      "step": 4575,
      "train_runtime": 8798.576,
      "train_tokens_per_second": 8515.303
    },
    {
      "epoch": 0.30837597629948826,
      "grad_norm": 1.0477240677599067,
      "learning_rate": 9.425028466344494e-06,
      "loss": 0.3664237499237061,
      "num_input_tokens_seen": 75004464,
      "step": 4580,
      "train_runtime": 8808.0745,
      "train_tokens_per_second": 8515.421
    },
    {
      "epoch": 0.3087126312954484,
      "grad_norm": 1.156568081005157,
      "learning_rate": 9.42379681537995e-06,
      "loss": 0.40375051498413084,
      "num_input_tokens_seen": 75086384,
      "step": 4585,
      "train_runtime": 8817.5868,
      "train_tokens_per_second": 8515.525
    },
    {
      "epoch": 0.30904928629140854,
      "grad_norm": 1.0788931429914974,
      "learning_rate": 9.422563927314732e-06,
      "loss": 0.38632376194000245,
      "num_input_tokens_seen": 75168304,
      "step": 4590,
      "train_runtime": 8827.1017,
      "train_tokens_per_second": 8515.627
    },
    {
      "epoch": 0.3093859412873687,
      "grad_norm": 1.2297868512875045,
      "learning_rate": 9.421329802493615e-06,
      "loss": 0.41800642013549805,
      "num_input_tokens_seen": 75250224,
      "step": 4595,
      "train_runtime": 8836.6126,
      "train_tokens_per_second": 8515.732
    },
    {
      "epoch": 0.3097225962833288,
      "grad_norm": 1.068581912774039,
      "learning_rate": 9.420094441261717e-06,
      "loss": 0.40239553451538085,
      "num_input_tokens_seen": 75332144,
      "step": 4600,
      "train_runtime": 8846.1212,
      "train_tokens_per_second": 8515.839
    },
    {
      "epoch": 0.31005925127928896,
      "grad_norm": 1.7322348249483517,
      "learning_rate": 9.418857843964506e-06,
      "loss": 0.3976526975631714,
      "num_input_tokens_seen": 75414064,
      "step": 4605,
      "train_runtime": 8855.7345,
      "train_tokens_per_second": 8515.845
    },
    {
      "epoch": 0.3103959062752491,
      "grad_norm": 1.1394862525357652,
      "learning_rate": 9.417620010947786e-06,
      "loss": 0.44963579177856444,
      "num_input_tokens_seen": 75495984,
      "step": 4610,
      "train_runtime": 8865.2458,
      "train_tokens_per_second": 8515.949
    },
    {
      "epoch": 0.31073256127120924,
      "grad_norm": 1.0351917363302334,
      "learning_rate": 9.416380942557719e-06,
      "loss": 0.4218024253845215,
      "num_input_tokens_seen": 75577904,
      "step": 4615,
      "train_runtime": 8874.7566,
      "train_tokens_per_second": 8516.054
    },
    {
      "epoch": 0.3110692162671694,
      "grad_norm": 1.051979530577901,
      "learning_rate": 9.415140639140803e-06,
      "loss": 0.44550395011901855,
      "num_input_tokens_seen": 75659824,
      "step": 4620,
      "train_runtime": 8884.2627,
      "train_tokens_per_second": 8516.162
    },
    {
      "epoch": 0.3114058712631295,
      "grad_norm": 1.0609120320947418,
      "learning_rate": 9.413899101043887e-06,
      "loss": 0.4207293510437012,
      "num_input_tokens_seen": 75741744,
      "step": 4625,
      "train_runtime": 8893.8976,
      "train_tokens_per_second": 8516.147
    },
    {
      "epoch": 0.31174252625908966,
      "grad_norm": 1.089520764150589,
      "learning_rate": 9.412656328614162e-06,
      "loss": 0.43401222229003905,
      "num_input_tokens_seen": 75823664,
      "step": 4630,
      "train_runtime": 8903.4073,
      "train_tokens_per_second": 8516.252
    },
    {
      "epoch": 0.3120791812550498,
      "grad_norm": 1.1597453798002748,
      "learning_rate": 9.411412322199165e-06,
      "loss": 0.4324362754821777,
      "num_input_tokens_seen": 75905584,
      "step": 4635,
      "train_runtime": 8913.0639,
      "train_tokens_per_second": 8516.217
    },
    {
      "epoch": 0.31241583625100994,
      "grad_norm": 0.9777893252199009,
      "learning_rate": 9.41016708214678e-06,
      "loss": 0.41935243606567385,
      "num_input_tokens_seen": 75987504,
      "step": 4640,
      "train_runtime": 8922.7642,
      "train_tokens_per_second": 8516.139
    },
    {
      "epoch": 0.31275249124697013,
      "grad_norm": 1.1570573582593255,
      "learning_rate": 9.408920608805233e-06,
      "loss": 0.4468108654022217,
      "num_input_tokens_seen": 76069424,
      "step": 4645,
      "train_runtime": 8932.5932,
      "train_tokens_per_second": 8515.94
    },
    {
      "epoch": 0.31308914624293027,
      "grad_norm": 1.0963501094313914,
      "learning_rate": 9.407672902523097e-06,
      "loss": 0.44944448471069337,
      "num_input_tokens_seen": 76151344,
      "step": 4650,
      "train_runtime": 8942.1209,
      "train_tokens_per_second": 8516.027
    },
    {
      "epoch": 0.3134258012388904,
      "grad_norm": 1.1654436738377796,
      "learning_rate": 9.406423963649287e-06,
      "loss": 0.4017190933227539,
      "num_input_tokens_seen": 76233264,
      "step": 4655,
      "train_runtime": 8951.8066,
      "train_tokens_per_second": 8515.964
    },
    {
      "epoch": 0.31376245623485055,
      "grad_norm": 1.2936050372292518,
      "learning_rate": 9.405173792533069e-06,
      "loss": 0.4109018325805664,
      "num_input_tokens_seen": 76315184,
      "step": 4660,
      "train_runtime": 8961.4623,
      "train_tokens_per_second": 8515.93
    },
    {
      "epoch": 0.3140991112308107,
      "grad_norm": 1.0661504652375982,
      "learning_rate": 9.403922389524045e-06,
      "loss": 0.4197213172912598,
      "num_input_tokens_seen": 76397104,
      "step": 4665,
      "train_runtime": 8971.2668,
      "train_tokens_per_second": 8515.754
    },
    {
      "epoch": 0.31443576622677083,
      "grad_norm": 1.0957258839577617,
      "learning_rate": 9.40266975497217e-06,
      "loss": 0.4321625232696533,
      "num_input_tokens_seen": 76479024,
      "step": 4670,
      "train_runtime": 8981.0083,
      "train_tokens_per_second": 8515.639
    },
    {
      "epoch": 0.31477242122273097,
      "grad_norm": 1.4565411389055456,
      "learning_rate": 9.401415889227736e-06,
      "loss": 0.45740337371826173,
      "num_input_tokens_seen": 76560944,
      "step": 4675,
      "train_runtime": 8990.7033,
      "train_tokens_per_second": 8515.568
    },
    {
      "epoch": 0.3151090762186911,
      "grad_norm": 1.1111178455893924,
      "learning_rate": 9.400160792641385e-06,
      "loss": 0.4735849380493164,
      "num_input_tokens_seen": 76642864,
      "step": 4680,
      "train_runtime": 9000.3945,
      "train_tokens_per_second": 8515.501
    },
    {
      "epoch": 0.31544573121465125,
      "grad_norm": 1.0186251040194967,
      "learning_rate": 9.398904465564097e-06,
      "loss": 0.44649810791015626,
      "num_input_tokens_seen": 76724784,
      "step": 4685,
      "train_runtime": 9010.039,
      "train_tokens_per_second": 8515.477
    },
    {
      "epoch": 0.3157823862106114,
      "grad_norm": 1.23921442546418,
      "learning_rate": 9.397646908347203e-06,
      "loss": 0.3959618091583252,
      "num_input_tokens_seen": 76806704,
      "step": 4690,
      "train_runtime": 9019.6833,
      "train_tokens_per_second": 8515.455
    },
    {
      "epoch": 0.3161190412065715,
      "grad_norm": 1.0964095228675732,
      "learning_rate": 9.396388121342375e-06,
      "loss": 0.40221376419067384,
      "num_input_tokens_seen": 76888624,
      "step": 4695,
      "train_runtime": 9029.356,
      "train_tokens_per_second": 8515.405
    },
    {
      "epoch": 0.31645569620253167,
      "grad_norm": 1.208036672229273,
      "learning_rate": 9.395128104901628e-06,
      "loss": 0.3947370290756226,
      "num_input_tokens_seen": 76970240,
      "step": 4700,
      "train_runtime": 9039.3097,
      "train_tokens_per_second": 8515.057
    },
    {
      "epoch": 0.3167923511984918,
      "grad_norm": 1.075372892270335,
      "learning_rate": 9.393866859377321e-06,
      "loss": 0.45838031768798826,
      "num_input_tokens_seen": 77052160,
      "step": 4705,
      "train_runtime": 9049.1803,
      "train_tokens_per_second": 8514.822
    },
    {
      "epoch": 0.31712900619445195,
      "grad_norm": 1.0581063719077535,
      "learning_rate": 9.392604385122157e-06,
      "loss": 0.41800613403320314,
      "num_input_tokens_seen": 77134080,
      "step": 4710,
      "train_runtime": 9058.7131,
      "train_tokens_per_second": 8514.905
    },
    {
      "epoch": 0.3174656611904121,
      "grad_norm": 1.1419549625905607,
      "learning_rate": 9.391340682489185e-06,
      "loss": 0.39562578201293946,
      "num_input_tokens_seen": 77216000,
      "step": 4715,
      "train_runtime": 9068.5821,
      "train_tokens_per_second": 8514.672
    },
    {
      "epoch": 0.3178023161863722,
      "grad_norm": 1.396061070707909,
      "learning_rate": 9.390075751831794e-06,
      "loss": 0.4368928909301758,
      "num_input_tokens_seen": 77297920,
      "step": 4720,
      "train_runtime": 9078.1,
      "train_tokens_per_second": 8514.769
    },
    {
      "epoch": 0.31813897118233236,
      "grad_norm": 1.3980040632567061,
      "learning_rate": 9.388809593503718e-06,
      "loss": 0.4521182060241699,
      "num_input_tokens_seen": 77379840,
      "step": 4725,
      "train_runtime": 9087.6125,
      "train_tokens_per_second": 8514.87
    },
    {
      "epoch": 0.3184756261782925,
      "grad_norm": 1.2753897034895996,
      "learning_rate": 9.387542207859034e-06,
      "loss": 0.46914329528808596,
      "num_input_tokens_seen": 77461760,
      "step": 4730,
      "train_runtime": 9097.1222,
      "train_tokens_per_second": 8514.974
    },
    {
      "epoch": 0.31881228117425264,
      "grad_norm": 1.1219697938916575,
      "learning_rate": 9.386273595252161e-06,
      "loss": 0.4532617568969727,
      "num_input_tokens_seen": 77543680,
      "step": 4735,
      "train_runtime": 9106.6227,
      "train_tokens_per_second": 8515.087
    },
    {
      "epoch": 0.3191489361702128,
      "grad_norm": 1.1771488341428553,
      "learning_rate": 9.385003756037865e-06,
      "loss": 0.46709909439086916,
      "num_input_tokens_seen": 77625600,
      "step": 4740,
      "train_runtime": 9116.2711,
      "train_tokens_per_second": 8515.061
    },
    {
      "epoch": 0.3194855911661729,
      "grad_norm": 1.1541741266117471,
      "learning_rate": 9.383732690571253e-06,
      "loss": 0.41798577308654783,
      "num_input_tokens_seen": 77707520,
      "step": 4745,
      "train_runtime": 9125.8809,
      "train_tokens_per_second": 8515.071
    },
    {
      "epoch": 0.31982224616213306,
      "grad_norm": 1.1605520926321682,
      "learning_rate": 9.382460399207769e-06,
      "loss": 0.4560890197753906,
      "num_input_tokens_seen": 77789440,
      "step": 4750,
      "train_runtime": 9135.6264,
      "train_tokens_per_second": 8514.954
    },
    {
      "epoch": 0.3201589011580932,
      "grad_norm": 1.166244599754302,
      "learning_rate": 9.381186882303212e-06,
      "loss": 0.4143784523010254,
      "num_input_tokens_seen": 77871360,
      "step": 4755,
      "train_runtime": 9145.2342,
      "train_tokens_per_second": 8514.966
    },
    {
      "epoch": 0.32049555615405334,
      "grad_norm": 1.072874433198712,
      "learning_rate": 9.379912140213713e-06,
      "loss": 0.3885941505432129,
      "num_input_tokens_seen": 77953280,
      "step": 4760,
      "train_runtime": 9154.7358,
      "train_tokens_per_second": 8515.077
    },
    {
      "epoch": 0.3208322111500135,
      "grad_norm": 0.9712167926970543,
      "learning_rate": 9.37863617329575e-06,
      "loss": 0.42183990478515626,
      "num_input_tokens_seen": 78035200,
      "step": 4765,
      "train_runtime": 9164.3651,
      "train_tokens_per_second": 8515.069
    },
    {
      "epoch": 0.3211688661459736,
      "grad_norm": 1.228832367180628,
      "learning_rate": 9.377358981906145e-06,
      "loss": 0.42816920280456544,
      "num_input_tokens_seen": 78116832,
      "step": 4770,
      "train_runtime": 9174.3327,
      "train_tokens_per_second": 8514.715
    },
    {
      "epoch": 0.32150552114193376,
      "grad_norm": 1.1025008579512525,
      "learning_rate": 9.376080566402059e-06,
      "loss": 0.4301916599273682,
      "num_input_tokens_seen": 78198752,
      "step": 4775,
      "train_runtime": 9183.9207,
      "train_tokens_per_second": 8514.746
    },
    {
      "epoch": 0.3218421761378939,
      "grad_norm": 1.067457655930923,
      "learning_rate": 9.374800927140994e-06,
      "loss": 0.34853439331054686,
      "num_input_tokens_seen": 78280672,
      "step": 4780,
      "train_runtime": 9193.538,
      "train_tokens_per_second": 8514.749
    },
    {
      "epoch": 0.32217883113385404,
      "grad_norm": 1.3210580167279051,
      "learning_rate": 9.373520064480804e-06,
      "loss": 0.3959688663482666,
      "num_input_tokens_seen": 78362592,
      "step": 4785,
      "train_runtime": 9203.1492,
      "train_tokens_per_second": 8514.758
    },
    {
      "epoch": 0.3225154861298142,
      "grad_norm": 1.2367317375457751,
      "learning_rate": 9.372237978779672e-06,
      "loss": 0.45833454132080076,
      "num_input_tokens_seen": 78444512,
      "step": 4790,
      "train_runtime": 9212.6477,
      "train_tokens_per_second": 8514.872
    },
    {
      "epoch": 0.3228521411257743,
      "grad_norm": 1.2336616688111848,
      "learning_rate": 9.37095467039613e-06,
      "loss": 0.4255546569824219,
      "num_input_tokens_seen": 78526432,
      "step": 4795,
      "train_runtime": 9222.2651,
      "train_tokens_per_second": 8514.875
    },
    {
      "epoch": 0.32318879612173446,
      "grad_norm": 1.1335092079403708,
      "learning_rate": 9.369670139689056e-06,
      "loss": 0.3963499069213867,
      "num_input_tokens_seen": 78607672,
      "step": 4800,
      "train_runtime": 9232.1127,
      "train_tokens_per_second": 8514.592
    },
    {
      "epoch": 0.3235254511176946,
      "grad_norm": 1.1639579198463128,
      "learning_rate": 9.368384387017659e-06,
      "loss": 0.4183605194091797,
      "num_input_tokens_seen": 78689448,
      "step": 4805,
      "train_runtime": 9241.6064,
      "train_tokens_per_second": 8514.694
    },
    {
      "epoch": 0.32386210611365474,
      "grad_norm": 1.0245141207831248,
      "learning_rate": 9.367097412741497e-06,
      "loss": 0.40891523361206056,
      "num_input_tokens_seen": 78771368,
      "step": 4810,
      "train_runtime": 9251.1433,
      "train_tokens_per_second": 8514.771
    },
    {
      "epoch": 0.3241987611096149,
      "grad_norm": 1.5178852634614808,
      "learning_rate": 9.36580921722047e-06,
      "loss": 0.4070852279663086,
      "num_input_tokens_seen": 78853288,
      "step": 4815,
      "train_runtime": 9260.7898,
      "train_tokens_per_second": 8514.748
    },
    {
      "epoch": 0.324535416105575,
      "grad_norm": 1.1006849960234164,
      "learning_rate": 9.364519800814818e-06,
      "loss": 0.41195058822631836,
      "num_input_tokens_seen": 78935208,
      "step": 4820,
      "train_runtime": 9270.3954,
      "train_tokens_per_second": 8514.762
    },
    {
      "epoch": 0.32487207110153515,
      "grad_norm": 1.2478338582922655,
      "learning_rate": 9.36322916388512e-06,
      "loss": 0.43653059005737305,
      "num_input_tokens_seen": 79017128,
      "step": 4825,
      "train_runtime": 9279.9499,
      "train_tokens_per_second": 8514.823
    },
    {
      "epoch": 0.3252087260974953,
      "grad_norm": 1.122549604026152,
      "learning_rate": 9.3619373067923e-06,
      "loss": 0.3917700290679932,
      "num_input_tokens_seen": 79099048,
      "step": 4830,
      "train_runtime": 9289.519,
      "train_tokens_per_second": 8514.87
    },
    {
      "epoch": 0.32554538109345543,
      "grad_norm": 1.1705154710393235,
      "learning_rate": 9.360644229897622e-06,
      "loss": 0.4469670295715332,
      "num_input_tokens_seen": 79180968,
      "step": 4835,
      "train_runtime": 9299.1048,
      "train_tokens_per_second": 8514.902
    },
    {
      "epoch": 0.32588203608941557,
      "grad_norm": 1.1786317584327608,
      "learning_rate": 9.35934993356269e-06,
      "loss": 0.45931034088134765,
      "num_input_tokens_seen": 79262888,
      "step": 4840,
      "train_runtime": 9309.0205,
      "train_tokens_per_second": 8514.632
    },
    {
      "epoch": 0.3262186910853757,
      "grad_norm": 1.245515924874069,
      "learning_rate": 9.35805441814945e-06,
      "loss": 0.3743353128433228,
      "num_input_tokens_seen": 79344808,
      "step": 4845,
      "train_runtime": 9318.6893,
      "train_tokens_per_second": 8514.589
    },
    {
      "epoch": 0.32655534608133585,
      "grad_norm": 1.1918810509564808,
      "learning_rate": 9.356757684020188e-06,
      "loss": 0.43869791030883787,
      "num_input_tokens_seen": 79426728,
      "step": 4850,
      "train_runtime": 9328.4108,
      "train_tokens_per_second": 8514.497
    },
    {
      "epoch": 0.326892001077296,
      "grad_norm": 1.070028535887082,
      "learning_rate": 9.355459731537533e-06,
      "loss": 0.3791459560394287,
      "num_input_tokens_seen": 79508648,
      "step": 4855,
      "train_runtime": 9338.1668,
      "train_tokens_per_second": 8514.374
    },
    {
      "epoch": 0.32722865607325613,
      "grad_norm": 1.0613100101434025,
      "learning_rate": 9.354160561064451e-06,
      "loss": 0.4171855926513672,
      "num_input_tokens_seen": 79590568,
      "step": 4860,
      "train_runtime": 9347.9869,
      "train_tokens_per_second": 8514.193
    },
    {
      "epoch": 0.32756531106921627,
      "grad_norm": 1.047741924870198,
      "learning_rate": 9.352860172964254e-06,
      "loss": 0.44502959251403806,
      "num_input_tokens_seen": 79672488,
      "step": 4865,
      "train_runtime": 9357.7043,
      "train_tokens_per_second": 8514.106
    },
    {
      "epoch": 0.3279019660651764,
      "grad_norm": 1.2124235354563468,
      "learning_rate": 9.35155856760059e-06,
      "loss": 0.44103269577026366,
      "num_input_tokens_seen": 79754408,
      "step": 4870,
      "train_runtime": 9367.3635,
      "train_tokens_per_second": 8514.072
    },
    {
      "epoch": 0.32823862106113655,
      "grad_norm": 1.1818638862348958,
      "learning_rate": 9.35025574533745e-06,
      "loss": 0.43625531196594236,
      "num_input_tokens_seen": 79836328,
      "step": 4875,
      "train_runtime": 9376.8789,
      "train_tokens_per_second": 8514.169
    },
    {
      "epoch": 0.3285752760570967,
      "grad_norm": 0.9496891843233825,
      "learning_rate": 9.34895170653916e-06,
      "loss": 0.42622966766357423,
      "num_input_tokens_seen": 79918248,
      "step": 4880,
      "train_runtime": 9386.392,
      "train_tokens_per_second": 8514.267
    },
    {
      "epoch": 0.3289119310530568,
      "grad_norm": 0.9476020649742796,
      "learning_rate": 9.347646451570394e-06,
      "loss": 0.3851289987564087,
      "num_input_tokens_seen": 80000168,
      "step": 4885,
      "train_runtime": 9395.894,
      "train_tokens_per_second": 8514.375
    },
    {
      "epoch": 0.32924858604901697,
      "grad_norm": 1.1408956114987692,
      "learning_rate": 9.346339980796162e-06,
      "loss": 0.4220522403717041,
      "num_input_tokens_seen": 80082088,
      "step": 4890,
      "train_runtime": 9405.4139,
      "train_tokens_per_second": 8514.467
    },
    {
      "epoch": 0.3295852410449771,
      "grad_norm": 1.1279211119730366,
      "learning_rate": 9.345032294581813e-06,
      "loss": 0.4412722587585449,
      "num_input_tokens_seen": 80164008,
      "step": 4895,
      "train_runtime": 9415.093,
      "train_tokens_per_second": 8514.415
    },
    {
      "epoch": 0.32992189604093725,
      "grad_norm": 1.2617416427966939,
      "learning_rate": 9.343723393293038e-06,
      "loss": 0.40000200271606445,
      "num_input_tokens_seen": 80245928,
      "step": 4900,
      "train_runtime": 9424.7013,
      "train_tokens_per_second": 8514.427
    },
    {
      "epoch": 0.3302585510368974,
      "grad_norm": 1.2483522330232544,
      "learning_rate": 9.342413277295869e-06,
      "loss": 0.44048175811767576,
      "num_input_tokens_seen": 80327848,
      "step": 4905,
      "train_runtime": 9434.5383,
      "train_tokens_per_second": 8514.232
    },
    {
      "epoch": 0.3305952060328575,
      "grad_norm": 1.0187864261659325,
      "learning_rate": 9.341101946956672e-06,
      "loss": 0.4030300617218018,
      "num_input_tokens_seen": 80409768,
      "step": 4910,
      "train_runtime": 9444.0457,
      "train_tokens_per_second": 8514.335
    },
    {
      "epoch": 0.33093186102881766,
      "grad_norm": 1.1585359089966787,
      "learning_rate": 9.33978940264216e-06,
      "loss": 0.4015416145324707,
      "num_input_tokens_seen": 80491688,
      "step": 4915,
      "train_runtime": 9453.5538,
      "train_tokens_per_second": 8514.437
    },
    {
      "epoch": 0.3312685160247778,
      "grad_norm": 1.2444262285382959,
      "learning_rate": 9.338475644719377e-06,
      "loss": 0.4189976692199707,
      "num_input_tokens_seen": 80573608,
      "step": 4920,
      "train_runtime": 9463.2368,
      "train_tokens_per_second": 8514.381
    },
    {
      "epoch": 0.33160517102073794,
      "grad_norm": 1.1516164320655202,
      "learning_rate": 9.337160673555719e-06,
      "loss": 0.3897253513336182,
      "num_input_tokens_seen": 80655528,
      "step": 4925,
      "train_runtime": 9472.8931,
      "train_tokens_per_second": 8514.35
    },
    {
      "epoch": 0.3319418260166981,
      "grad_norm": 1.0373577780463643,
      "learning_rate": 9.335844489518905e-06,
      "loss": 0.4195266246795654,
      "num_input_tokens_seen": 80737448,
      "step": 4930,
      "train_runtime": 9482.5704,
      "train_tokens_per_second": 8514.3
    },
    {
      "epoch": 0.3322784810126582,
      "grad_norm": 0.9084049657833471,
      "learning_rate": 9.334527092977008e-06,
      "loss": 0.41116924285888673,
      "num_input_tokens_seen": 80819368,
      "step": 4935,
      "train_runtime": 9492.254,
      "train_tokens_per_second": 8514.244
    },
    {
      "epoch": 0.33261513600861836,
      "grad_norm": 0.9566758867738144,
      "learning_rate": 9.33320848429843e-06,
      "loss": 0.4413943290710449,
      "num_input_tokens_seen": 80901288,
      "step": 4940,
      "train_runtime": 9502.1055,
      "train_tokens_per_second": 8514.038
    },
    {
      "epoch": 0.3329517910045785,
      "grad_norm": 1.0735070559701367,
      "learning_rate": 9.331888663851917e-06,
      "loss": 0.44445343017578126,
      "num_input_tokens_seen": 80983208,
      "step": 4945,
      "train_runtime": 9511.6037,
      "train_tokens_per_second": 8514.149
    },
    {
      "epoch": 0.33328844600053864,
      "grad_norm": 1.3200536947380344,
      "learning_rate": 9.330567632006552e-06,
      "loss": 0.42911829948425295,
      "num_input_tokens_seen": 81065128,
      "step": 4950,
      "train_runtime": 9521.1013,
      "train_tokens_per_second": 8514.26
    },
    {
      "epoch": 0.3336251009964988,
      "grad_norm": 1.1086324127033846,
      "learning_rate": 9.329245389131759e-06,
      "loss": 0.43160672187805177,
      "num_input_tokens_seen": 81147048,
      "step": 4955,
      "train_runtime": 9530.6317,
      "train_tokens_per_second": 8514.341
    },
    {
      "epoch": 0.3339617559924589,
      "grad_norm": 1.095346921865322,
      "learning_rate": 9.327921935597298e-06,
      "loss": 0.40712833404541016,
      "num_input_tokens_seen": 81228968,
      "step": 4960,
      "train_runtime": 9540.4514,
      "train_tokens_per_second": 8514.164
    },
    {
      "epoch": 0.33429841098841906,
      "grad_norm": 1.0923631404674328,
      "learning_rate": 9.326597271773267e-06,
      "loss": 0.4390350341796875,
      "num_input_tokens_seen": 81310888,
      "step": 4965,
      "train_runtime": 9549.9569,
      "train_tokens_per_second": 8514.268
    },
    {
      "epoch": 0.3346350659843792,
      "grad_norm": 1.0279677565273502,
      "learning_rate": 9.325271398030107e-06,
      "loss": 0.3829920291900635,
      "num_input_tokens_seen": 81392808,
      "step": 4970,
      "train_runtime": 9559.4592,
      "train_tokens_per_second": 8514.374
    },
    {
      "epoch": 0.33497172098033934,
      "grad_norm": 1.2927721495849764,
      "learning_rate": 9.323944314738591e-06,
      "loss": 0.4332720756530762,
      "num_input_tokens_seen": 81474728,
      "step": 4975,
      "train_runtime": 9569.091,
      "train_tokens_per_second": 8514.364
    },
    {
      "epoch": 0.3353083759762995,
      "grad_norm": 1.2380686837079111,
      "learning_rate": 9.322616022269838e-06,
      "loss": 0.38703668117523193,
      "num_input_tokens_seen": 81556648,
      "step": 4980,
      "train_runtime": 9578.7142,
      "train_tokens_per_second": 8514.363
    },
    {
      "epoch": 0.3356450309722596,
      "grad_norm": 0.9169733817649471,
      "learning_rate": 9.321286520995297e-06,
      "loss": 0.42053747177124023,
      "num_input_tokens_seen": 81638568,
      "step": 4985,
      "train_runtime": 9588.2665,
      "train_tokens_per_second": 8514.424
    },
    {
      "epoch": 0.33598168596821976,
      "grad_norm": 1.237478152062952,
      "learning_rate": 9.319955811286758e-06,
      "loss": 0.4284243583679199,
      "num_input_tokens_seen": 81720488,
      "step": 4990,
      "train_runtime": 9597.82,
      "train_tokens_per_second": 8514.484
    },
    {
      "epoch": 0.3363183409641799,
      "grad_norm": 1.0339413904474422,
      "learning_rate": 9.318623893516354e-06,
      "loss": 0.4205324172973633,
      "num_input_tokens_seen": 81802408,
      "step": 4995,
      "train_runtime": 9607.4731,
      "train_tokens_per_second": 8514.456
    },
    {
      "epoch": 0.33665499596014004,
      "grad_norm": 1.1189110722843802,
      "learning_rate": 9.317290768056548e-06,
      "loss": 0.41985454559326174,
      "num_input_tokens_seen": 81884328,
      "step": 5000,
      "train_runtime": 9617.0366,
      "train_tokens_per_second": 8514.507
    },
    {
      "epoch": 0.3369916509561002,
      "grad_norm": 1.0732273673797856,
      "learning_rate": 9.315956435280147e-06,
      "loss": 0.4256202220916748,
      "num_input_tokens_seen": 81966120,
      "step": 5005,
      "train_runtime": 9627.0847,
      "train_tokens_per_second": 8514.116
    },
    {
      "epoch": 0.3373283059520603,
      "grad_norm": 1.1789800072839385,
      "learning_rate": 9.314620895560292e-06,
      "loss": 0.40423927307128904,
      "num_input_tokens_seen": 82048040,
      "step": 5010,
      "train_runtime": 9636.7196,
      "train_tokens_per_second": 8514.105
    },
    {
      "epoch": 0.33766496094802045,
      "grad_norm": 1.3786027263835399,
      "learning_rate": 9.313284149270459e-06,
      "loss": 0.4306342124938965,
      "num_input_tokens_seen": 82129960,
      "step": 5015,
      "train_runtime": 9646.3399,
      "train_tokens_per_second": 8514.106
    },
    {
      "epoch": 0.3380016159439806,
      "grad_norm": 1.2315821630963943,
      "learning_rate": 9.311946196784469e-06,
      "loss": 0.40974855422973633,
      "num_input_tokens_seen": 82211880,
      "step": 5020,
      "train_runtime": 9655.9446,
      "train_tokens_per_second": 8514.121
    },
    {
      "epoch": 0.33833827093994073,
      "grad_norm": 0.9774005742255519,
      "learning_rate": 9.310607038476476e-06,
      "loss": 0.4053037166595459,
      "num_input_tokens_seen": 82293800,
      "step": 5025,
      "train_runtime": 9665.4879,
      "train_tokens_per_second": 8514.19
    },
    {
      "epoch": 0.3386749259359009,
      "grad_norm": 1.2003477022398206,
      "learning_rate": 9.30926667472097e-06,
      "loss": 0.4145936965942383,
      "num_input_tokens_seen": 82375720,
      "step": 5030,
      "train_runtime": 9675.0983,
      "train_tokens_per_second": 8514.2
    },
    {
      "epoch": 0.339011580931861,
      "grad_norm": 1.2594585154124702,
      "learning_rate": 9.307925105892779e-06,
      "loss": 0.4268640518188477,
      "num_input_tokens_seen": 82457640,
      "step": 5035,
      "train_runtime": 9684.7774,
      "train_tokens_per_second": 8514.149
    },
    {
      "epoch": 0.33934823592782115,
      "grad_norm": 1.1818390060127701,
      "learning_rate": 9.30658233236707e-06,
      "loss": 0.42670745849609376,
      "num_input_tokens_seen": 82539560,
      "step": 5040,
      "train_runtime": 9694.4476,
      "train_tokens_per_second": 8514.107
    },
    {
      "epoch": 0.3396848909237813,
      "grad_norm": 1.1115721684177091,
      "learning_rate": 9.30523835451934e-06,
      "loss": 0.43652915954589844,
      "num_input_tokens_seen": 82621480,
      "step": 5045,
      "train_runtime": 9704.112,
      "train_tokens_per_second": 8514.069
    },
    {
      "epoch": 0.34002154591974143,
      "grad_norm": 1.1849518488814181,
      "learning_rate": 9.303893172725437e-06,
      "loss": 0.4386833667755127,
      "num_input_tokens_seen": 82703400,
      "step": 5050,
      "train_runtime": 9713.8447,
      "train_tokens_per_second": 8513.972
    },
    {
      "epoch": 0.34035820091570157,
      "grad_norm": 1.1740412434073904,
      "learning_rate": 9.302546787361533e-06,
      "loss": 0.4030099391937256,
      "num_input_tokens_seen": 82785320,
      "step": 5055,
      "train_runtime": 9723.5944,
      "train_tokens_per_second": 8513.86
    },
    {
      "epoch": 0.3406948559116617,
      "grad_norm": 1.2490536936769312,
      "learning_rate": 9.301199198804139e-06,
      "loss": 0.40512895584106445,
      "num_input_tokens_seen": 82867240,
      "step": 5060,
      "train_runtime": 9733.284,
      "train_tokens_per_second": 8513.801
    },
    {
      "epoch": 0.34103151090762185,
      "grad_norm": 1.2517057043583684,
      "learning_rate": 9.299850407430102e-06,
      "loss": 0.47108044624328616,
      "num_input_tokens_seen": 82949160,
      "step": 5065,
      "train_runtime": 9742.9559,
      "train_tokens_per_second": 8513.757
    },
    {
      "epoch": 0.341368165903582,
      "grad_norm": 0.9306484024177044,
      "learning_rate": 9.298500413616613e-06,
      "loss": 0.39539294242858886,
      "num_input_tokens_seen": 83031080,
      "step": 5070,
      "train_runtime": 9752.6283,
      "train_tokens_per_second": 8513.713
    },
    {
      "epoch": 0.3417048208995421,
      "grad_norm": 1.232070659624266,
      "learning_rate": 9.297149217741188e-06,
      "loss": 0.3893394470214844,
      "num_input_tokens_seen": 83113000,
      "step": 5075,
      "train_runtime": 9762.3268,
      "train_tokens_per_second": 8513.647
    },
    {
      "epoch": 0.34204147589550227,
      "grad_norm": 1.2062561136828711,
      "learning_rate": 9.295796820181688e-06,
      "loss": 0.4828469276428223,
      "num_input_tokens_seen": 83194920,
      "step": 5080,
      "train_runtime": 9771.9962,
      "train_tokens_per_second": 8513.605
    },
    {
      "epoch": 0.3423781308914624,
      "grad_norm": 1.274553060950057,
      "learning_rate": 9.294443221316305e-06,
      "loss": 0.41120405197143556,
      "num_input_tokens_seen": 83276840,
      "step": 5085,
      "train_runtime": 9781.4987,
      "train_tokens_per_second": 8513.71
    },
    {
      "epoch": 0.34271478588742255,
      "grad_norm": 1.0790899560842087,
      "learning_rate": 9.293088421523569e-06,
      "loss": 0.47322731018066405,
      "num_input_tokens_seen": 83358648,
      "step": 5090,
      "train_runtime": 9791.5456,
      "train_tokens_per_second": 8513.329
    },
    {
      "epoch": 0.3430514408833827,
      "grad_norm": 1.2355897199936694,
      "learning_rate": 9.291732421182346e-06,
      "loss": 0.3980757713317871,
      "num_input_tokens_seen": 83440568,
      "step": 5095,
      "train_runtime": 9801.2138,
      "train_tokens_per_second": 8513.289
    },
    {
      "epoch": 0.3433880958793428,
      "grad_norm": 1.2172581699601301,
      "learning_rate": 9.290375220671837e-06,
      "loss": 0.4107996940612793,
      "num_input_tokens_seen": 83522488,
      "step": 5100,
      "train_runtime": 9811.0398,
      "train_tokens_per_second": 8513.113
    },
    {
      "epoch": 0.34372475087530296,
      "grad_norm": 1.065703069649166,
      "learning_rate": 9.28901682037158e-06,
      "loss": 0.38751099109649656,
      "num_input_tokens_seen": 83604408,
      "step": 5105,
      "train_runtime": 9820.6513,
      "train_tokens_per_second": 8513.123
    },
    {
      "epoch": 0.3440614058712631,
      "grad_norm": 1.0937002199485533,
      "learning_rate": 9.287657220661442e-06,
      "loss": 0.4176473140716553,
      "num_input_tokens_seen": 83686328,
      "step": 5110,
      "train_runtime": 9830.289,
      "train_tokens_per_second": 8513.11
    },
    {
      "epoch": 0.34439806086722324,
      "grad_norm": 1.4276021560170589,
      "learning_rate": 9.286296421921636e-06,
      "loss": 0.4128898620605469,
      "num_input_tokens_seen": 83768168,
      "step": 5115,
      "train_runtime": 9840.4317,
      "train_tokens_per_second": 8512.652
    },
    {
      "epoch": 0.3447347158631834,
      "grad_norm": 1.3126619054511766,
      "learning_rate": 9.284934424532706e-06,
      "loss": 0.5179042339324951,
      "num_input_tokens_seen": 83850088,
      "step": 5120,
      "train_runtime": 9849.9317,
      "train_tokens_per_second": 8512.758
    },
    {
      "epoch": 0.3450713708591435,
      "grad_norm": 1.0000674808135126,
      "learning_rate": 9.283571228875525e-06,
      "loss": 0.3790269374847412,
      "num_input_tokens_seen": 83931504,
      "step": 5125,
      "train_runtime": 9859.7728,
      "train_tokens_per_second": 8512.519
    },
    {
      "epoch": 0.3454080258551037,
      "grad_norm": 1.1148116743737428,
      "learning_rate": 9.282206835331311e-06,
      "loss": 0.4774306297302246,
      "num_input_tokens_seen": 84013424,
      "step": 5130,
      "train_runtime": 9869.2808,
      "train_tokens_per_second": 8512.619
    },
    {
      "epoch": 0.34574468085106386,
      "grad_norm": 1.141446064947306,
      "learning_rate": 9.28084124428161e-06,
      "loss": 0.35935330390930176,
      "num_input_tokens_seen": 84095344,
      "step": 5135,
      "train_runtime": 9878.7936,
      "train_tokens_per_second": 8512.714
    },
    {
      "epoch": 0.346081335847024,
      "grad_norm": 1.318948311396857,
      "learning_rate": 9.279474456108305e-06,
      "loss": 0.3846935272216797,
      "num_input_tokens_seen": 84177264,
      "step": 5140,
      "train_runtime": 9888.606,
      "train_tokens_per_second": 8512.551
    },
    {
      "epoch": 0.34641799084298414,
      "grad_norm": 1.0422895540447303,
      "learning_rate": 9.278106471193615e-06,
      "loss": 0.39165849685668946,
      "num_input_tokens_seen": 84259184,
      "step": 5145,
      "train_runtime": 9898.255,
      "train_tokens_per_second": 8512.529
    },
    {
      "epoch": 0.3467546458389443,
      "grad_norm": 1.0419302309418565,
      "learning_rate": 9.276737289920093e-06,
      "loss": 0.43006553649902346,
      "num_input_tokens_seen": 84341104,
      "step": 5150,
      "train_runtime": 9907.865,
      "train_tokens_per_second": 8512.541
    },
    {
      "epoch": 0.3470913008349044,
      "grad_norm": 1.0640571762980657,
      "learning_rate": 9.275366912670624e-06,
      "loss": 0.4371912479400635,
      "num_input_tokens_seen": 84421936,
      "step": 5155,
      "train_runtime": 9917.5307,
      "train_tokens_per_second": 8512.395
    },
    {
      "epoch": 0.34742795583086455,
      "grad_norm": 1.1748080444478497,
      "learning_rate": 9.273995339828432e-06,
      "loss": 0.4296929359436035,
      "num_input_tokens_seen": 84503856,
      "step": 5160,
      "train_runtime": 9927.0399,
      "train_tokens_per_second": 8512.493
    },
    {
      "epoch": 0.3477646108268247,
      "grad_norm": 1.2190592486176866,
      "learning_rate": 9.272622571777072e-06,
      "loss": 0.4642051696777344,
      "num_input_tokens_seen": 84585776,
      "step": 5165,
      "train_runtime": 9936.6727,
      "train_tokens_per_second": 8512.485
    },
    {
      "epoch": 0.34810126582278483,
      "grad_norm": 1.3063591757602577,
      "learning_rate": 9.271248608900434e-06,
      "loss": 0.3292104721069336,
      "num_input_tokens_seen": 84667696,
      "step": 5170,
      "train_runtime": 9946.2931,
      "train_tokens_per_second": 8512.488
    },
    {
      "epoch": 0.348437920818745,
      "grad_norm": 1.2922185958423904,
      "learning_rate": 9.269873451582741e-06,
      "loss": 0.439868688583374,
      "num_input_tokens_seen": 84749616,
      "step": 5175,
      "train_runtime": 9955.8498,
      "train_tokens_per_second": 8512.545
    },
    {
      "epoch": 0.3487745758147051,
      "grad_norm": 1.3403216946350418,
      "learning_rate": 9.268497100208556e-06,
      "loss": 0.45718908309936523,
      "num_input_tokens_seen": 84831536,
      "step": 5180,
      "train_runtime": 9965.4313,
      "train_tokens_per_second": 8512.581
    },
    {
      "epoch": 0.34911123081066525,
      "grad_norm": 1.3093910715710457,
      "learning_rate": 9.267119555162765e-06,
      "loss": 0.4276247501373291,
      "num_input_tokens_seen": 84913456,
      "step": 5185,
      "train_runtime": 9974.9906,
      "train_tokens_per_second": 8512.635
    },
    {
      "epoch": 0.3494478858066254,
      "grad_norm": 1.2166336582491941,
      "learning_rate": 9.2657408168306e-06,
      "loss": 0.4264081478118896,
      "num_input_tokens_seen": 84995376,
      "step": 5190,
      "train_runtime": 9984.6016,
      "train_tokens_per_second": 8512.646
    },
    {
      "epoch": 0.34978454080258553,
      "grad_norm": 1.3502885276954129,
      "learning_rate": 9.264360885597617e-06,
      "loss": 0.44219121932983396,
      "num_input_tokens_seen": 85077296,
      "step": 5195,
      "train_runtime": 9994.6982,
      "train_tokens_per_second": 8512.243
    },
    {
      "epoch": 0.35012119579854567,
      "grad_norm": 1.0797252741298327,
      "learning_rate": 9.262979761849709e-06,
      "loss": 0.40576963424682616,
      "num_input_tokens_seen": 85159216,
      "step": 5200,
      "train_runtime": 10004.2819,
      "train_tokens_per_second": 8512.277
    },
    {
      "epoch": 0.3504578507945058,
      "grad_norm": 1.1922890533715054,
      "learning_rate": 9.261597445973106e-06,
      "loss": 0.4562631607055664,
      "num_input_tokens_seen": 85241136,
      "step": 5205,
      "train_runtime": 10013.907,
      "train_tokens_per_second": 8512.276
    },
    {
      "epoch": 0.35079450579046595,
      "grad_norm": 1.1427412790233689,
      "learning_rate": 9.260213938354365e-06,
      "loss": 0.42475113868713377,
      "num_input_tokens_seen": 85323056,
      "step": 5210,
      "train_runtime": 10023.5288,
      "train_tokens_per_second": 8512.277
    },
    {
      "epoch": 0.3511311607864261,
      "grad_norm": 1.2630162967598462,
      "learning_rate": 9.25882923938038e-06,
      "loss": 0.4418492317199707,
      "num_input_tokens_seen": 85404832,
      "step": 5215,
      "train_runtime": 10033.6792,
      "train_tokens_per_second": 8511.816
    },
    {
      "epoch": 0.35146781578238623,
      "grad_norm": 1.266251233435307,
      "learning_rate": 9.257443349438382e-06,
      "loss": 0.4357272148132324,
      "num_input_tokens_seen": 85486752,
      "step": 5220,
      "train_runtime": 10043.283,
      "train_tokens_per_second": 8511.833
    },
    {
      "epoch": 0.35180447077834637,
      "grad_norm": 0.9708317987278431,
      "learning_rate": 9.256056268915924e-06,
      "loss": 0.3974766731262207,
      "num_input_tokens_seen": 85568672,
      "step": 5225,
      "train_runtime": 10052.9596,
      "train_tokens_per_second": 8511.789
    },
    {
      "epoch": 0.3521411257743065,
      "grad_norm": 1.0225107458879212,
      "learning_rate": 9.254667998200906e-06,
      "loss": 0.44359898567199707,
      "num_input_tokens_seen": 85650592,
      "step": 5230,
      "train_runtime": 10062.662,
      "train_tokens_per_second": 8511.723
    },
    {
      "epoch": 0.35247778077026665,
      "grad_norm": 1.2824891542369972,
      "learning_rate": 9.253278537681548e-06,
      "loss": 0.48610725402832033,
      "num_input_tokens_seen": 85732512,
      "step": 5235,
      "train_runtime": 10072.5,
      "train_tokens_per_second": 8511.542
    },
    {
      "epoch": 0.3528144357662268,
      "grad_norm": 1.2040083587132242,
      "learning_rate": 9.251887887746407e-06,
      "loss": 0.42476820945739746,
      "num_input_tokens_seen": 85814432,
      "step": 5240,
      "train_runtime": 10082.0059,
      "train_tokens_per_second": 8511.643
    },
    {
      "epoch": 0.3531510907621869,
      "grad_norm": 1.1295103407252682,
      "learning_rate": 9.25049604878438e-06,
      "loss": 0.3608102321624756,
      "num_input_tokens_seen": 85896352,
      "step": 5245,
      "train_runtime": 10091.5104,
      "train_tokens_per_second": 8511.744
    },
    {
      "epoch": 0.35348774575814707,
      "grad_norm": 1.0610131279212613,
      "learning_rate": 9.249103021184684e-06,
      "loss": 0.4264249801635742,
      "num_input_tokens_seen": 85978272,
      "step": 5250,
      "train_runtime": 10101.0151,
      "train_tokens_per_second": 8511.845
    },
    {
      "epoch": 0.3538244007541072,
      "grad_norm": 1.505251915412993,
      "learning_rate": 9.24770880533688e-06,
      "loss": 0.5038589000701904,
      "num_input_tokens_seen": 86060192,
      "step": 5255,
      "train_runtime": 10110.5414,
      "train_tokens_per_second": 8511.927
    },
    {
      "epoch": 0.35416105575006734,
      "grad_norm": 1.4429671050190034,
      "learning_rate": 9.24631340163085e-06,
      "loss": 0.4451129913330078,
      "num_input_tokens_seen": 86142112,
      "step": 5260,
      "train_runtime": 10120.203,
      "train_tokens_per_second": 8511.896
    },
    {
      "epoch": 0.3544977107460275,
      "grad_norm": 1.1411650161226328,
      "learning_rate": 9.244916810456822e-06,
      "loss": 0.39274139404296876,
      "num_input_tokens_seen": 86224032,
      "step": 5265,
      "train_runtime": 10129.866,
      "train_tokens_per_second": 8511.863
    },
    {
      "epoch": 0.3548343657419876,
      "grad_norm": 1.0975172668493218,
      "learning_rate": 9.24351903220534e-06,
      "loss": 0.43754992485046384,
      "num_input_tokens_seen": 86305952,
      "step": 5270,
      "train_runtime": 10139.5608,
      "train_tokens_per_second": 8511.804
    },
    {
      "epoch": 0.35517102073794776,
      "grad_norm": 1.1669720075252092,
      "learning_rate": 9.242120067267295e-06,
      "loss": 0.3605740785598755,
      "num_input_tokens_seen": 86387872,
      "step": 5275,
      "train_runtime": 10149.1669,
      "train_tokens_per_second": 8511.819
    },
    {
      "epoch": 0.3555076757339079,
      "grad_norm": 1.1297853055669758,
      "learning_rate": 9.2407199160339e-06,
      "loss": 0.4295405387878418,
      "num_input_tokens_seen": 86469792,
      "step": 5280,
      "train_runtime": 10158.6642,
      "train_tokens_per_second": 8511.925
    },
    {
      "epoch": 0.35584433072986804,
      "grad_norm": 1.1957361458994609,
      "learning_rate": 9.239318578896701e-06,
      "loss": 0.4562390327453613,
      "num_input_tokens_seen": 86551712,
      "step": 5285,
      "train_runtime": 10168.164,
      "train_tokens_per_second": 8512.029
    },
    {
      "epoch": 0.3561809857258282,
      "grad_norm": 1.3235688124400478,
      "learning_rate": 9.237916056247583e-06,
      "loss": 0.44049768447875975,
      "num_input_tokens_seen": 86633632,
      "step": 5290,
      "train_runtime": 10177.6642,
      "train_tokens_per_second": 8512.133
    },
    {
      "epoch": 0.3565176407217883,
      "grad_norm": 1.6411484420127653,
      "learning_rate": 9.23651234847875e-06,
      "loss": 0.4670247554779053,
      "num_input_tokens_seen": 86715552,
      "step": 5295,
      "train_runtime": 10187.1976,
      "train_tokens_per_second": 8512.209
    },
    {
      "epoch": 0.35685429571774846,
      "grad_norm": 1.043616264394936,
      "learning_rate": 9.23510745598275e-06,
      "loss": 0.3600893974304199,
      "num_input_tokens_seen": 86797472,
      "step": 5300,
      "train_runtime": 10196.8827,
      "train_tokens_per_second": 8512.158
    },
    {
      "epoch": 0.3571909507137086,
      "grad_norm": 1.0313455416742217,
      "learning_rate": 9.233701379152456e-06,
      "loss": 0.35738410949707033,
      "num_input_tokens_seen": 86879392,
      "step": 5305,
      "train_runtime": 10206.5825,
      "train_tokens_per_second": 8512.094
    },
    {
      "epoch": 0.35752760570966874,
      "grad_norm": 1.0085507425322808,
      "learning_rate": 9.232294118381069e-06,
      "loss": 0.3956319332122803,
      "num_input_tokens_seen": 86961312,
      "step": 5310,
      "train_runtime": 10216.2134,
      "train_tokens_per_second": 8512.088
    },
    {
      "epoch": 0.3578642607056289,
      "grad_norm": 1.2070493125934596,
      "learning_rate": 9.230885674062129e-06,
      "loss": 0.4040237903594971,
      "num_input_tokens_seen": 87043232,
      "step": 5315,
      "train_runtime": 10225.8969,
      "train_tokens_per_second": 8512.039
    },
    {
      "epoch": 0.358200915701589,
      "grad_norm": 1.2829665593630515,
      "learning_rate": 9.2294760465895e-06,
      "loss": 0.4551395893096924,
      "num_input_tokens_seen": 87125152,
      "step": 5320,
      "train_runtime": 10235.4045,
      "train_tokens_per_second": 8512.136
    },
    {
      "epoch": 0.35853757069754916,
      "grad_norm": 1.1764460521359958,
      "learning_rate": 9.228065236357383e-06,
      "loss": 0.44333696365356445,
      "num_input_tokens_seen": 87207072,
      "step": 5325,
      "train_runtime": 10244.9095,
      "train_tokens_per_second": 8512.234
    },
    {
      "epoch": 0.3588742256935093,
      "grad_norm": 1.0294058828955253,
      "learning_rate": 9.226653243760304e-06,
      "loss": 0.4153123378753662,
      "num_input_tokens_seen": 87288992,
      "step": 5330,
      "train_runtime": 10254.4131,
      "train_tokens_per_second": 8512.334
    },
    {
      "epoch": 0.35921088068946944,
      "grad_norm": 1.0884526886376966,
      "learning_rate": 9.225240069193124e-06,
      "loss": 0.46562728881835935,
      "num_input_tokens_seen": 87370912,
      "step": 5335,
      "train_runtime": 10263.9127,
      "train_tokens_per_second": 8512.437
    },
    {
      "epoch": 0.3595475356854296,
      "grad_norm": 1.3712018745870753,
      "learning_rate": 9.223825713051034e-06,
      "loss": 0.4538878440856934,
      "num_input_tokens_seen": 87452832,
      "step": 5340,
      "train_runtime": 10273.4162,
      "train_tokens_per_second": 8512.537
    },
    {
      "epoch": 0.3598841906813897,
      "grad_norm": 1.2054665841412764,
      "learning_rate": 9.222410175729549e-06,
      "loss": 0.4127343654632568,
      "num_input_tokens_seen": 87534752,
      "step": 5345,
      "train_runtime": 10282.9173,
      "train_tokens_per_second": 8512.638
    },
    {
      "epoch": 0.36022084567734985,
      "grad_norm": 0.9116710722930033,
      "learning_rate": 9.220993457624523e-06,
      "loss": 0.37782993316650393,
      "num_input_tokens_seen": 87616672,
      "step": 5350,
      "train_runtime": 10292.5414,
      "train_tokens_per_second": 8512.637
    },
    {
      "epoch": 0.36055750067331,
      "grad_norm": 1.2387573277326809,
      "learning_rate": 9.219575559132136e-06,
      "loss": 0.44984960556030273,
      "num_input_tokens_seen": 87698592,
      "step": 5355,
      "train_runtime": 10302.0923,
      "train_tokens_per_second": 8512.697
    },
    {
      "epoch": 0.36089415566927013,
      "grad_norm": 1.129188100547281,
      "learning_rate": 9.2181564806489e-06,
      "loss": 0.39498946666717527,
      "num_input_tokens_seen": 87780512,
      "step": 5360,
      "train_runtime": 10311.652,
      "train_tokens_per_second": 8512.75
    },
    {
      "epoch": 0.3612308106652303,
      "grad_norm": 1.1111800744423066,
      "learning_rate": 9.216736222571655e-06,
      "loss": 0.3818144559860229,
      "num_input_tokens_seen": 87862432,
      "step": 5365,
      "train_runtime": 10321.2575,
      "train_tokens_per_second": 8512.764
    },
    {
      "epoch": 0.3615674656611904,
      "grad_norm": 1.1269137039843509,
      "learning_rate": 9.215314785297572e-06,
      "loss": 0.43512392044067383,
      "num_input_tokens_seen": 87944352,
      "step": 5370,
      "train_runtime": 10330.7903,
      "train_tokens_per_second": 8512.839
    },
    {
      "epoch": 0.36190412065715055,
      "grad_norm": 1.2186105432923733,
      "learning_rate": 9.21389216922415e-06,
      "loss": 0.43152508735656736,
      "num_input_tokens_seen": 88026272,
      "step": 5375,
      "train_runtime": 10340.3609,
      "train_tokens_per_second": 8512.882
    },
    {
      "epoch": 0.3622407756531107,
      "grad_norm": 1.1750737851427437,
      "learning_rate": 9.212468374749222e-06,
      "loss": 0.464593505859375,
      "num_input_tokens_seen": 88108192,
      "step": 5380,
      "train_runtime": 10349.9903,
      "train_tokens_per_second": 8512.877
    },
    {
      "epoch": 0.36257743064907083,
      "grad_norm": 1.2608237306322836,
      "learning_rate": 9.211043402270942e-06,
      "loss": 0.40706615447998046,
      "num_input_tokens_seen": 88190112,
      "step": 5385,
      "train_runtime": 10359.6329,
      "train_tokens_per_second": 8512.861
    },
    {
      "epoch": 0.36291408564503097,
      "grad_norm": 0.956295096712853,
      "learning_rate": 9.209617252187806e-06,
      "loss": 0.37904748916625974,
      "num_input_tokens_seen": 88272032,
      "step": 5390,
      "train_runtime": 10369.1861,
      "train_tokens_per_second": 8512.918
    },
    {
      "epoch": 0.3632507406409911,
      "grad_norm": 1.0221802787552514,
      "learning_rate": 9.208189924898627e-06,
      "loss": 0.3921098232269287,
      "num_input_tokens_seen": 88353952,
      "step": 5395,
      "train_runtime": 10378.7605,
      "train_tokens_per_second": 8512.958
    },
    {
      "epoch": 0.36358739563695125,
      "grad_norm": 1.2414390886790594,
      "learning_rate": 9.206761420802554e-06,
      "loss": 0.4524404525756836,
      "num_input_tokens_seen": 88435872,
      "step": 5400,
      "train_runtime": 10388.369,
      "train_tokens_per_second": 8512.97
    },
    {
      "epoch": 0.3639240506329114,
      "grad_norm": 1.2170495896438998,
      "learning_rate": 9.205331740299065e-06,
      "loss": 0.4200294971466064,
      "num_input_tokens_seen": 88517696,
      "step": 5405,
      "train_runtime": 10398.5473,
      "train_tokens_per_second": 8512.506
    },
    {
      "epoch": 0.36426070562887153,
      "grad_norm": 0.8385209466043423,
      "learning_rate": 9.203900883787967e-06,
      "loss": 0.4166131019592285,
      "num_input_tokens_seen": 88599616,
      "step": 5410,
      "train_runtime": 10408.1752,
      "train_tokens_per_second": 8512.502
    },
    {
      "epoch": 0.36459736062483167,
      "grad_norm": 1.208627992984176,
      "learning_rate": 9.202468851669388e-06,
      "loss": 0.4571394920349121,
      "num_input_tokens_seen": 88681536,
      "step": 5415,
      "train_runtime": 10417.8184,
      "train_tokens_per_second": 8512.486
    },
    {
      "epoch": 0.3649340156207918,
      "grad_norm": 1.1694924593607132,
      "learning_rate": 9.201035644343797e-06,
      "loss": 0.4228943347930908,
      "num_input_tokens_seen": 88763456,
      "step": 5420,
      "train_runtime": 10427.4573,
      "train_tokens_per_second": 8512.474
    },
    {
      "epoch": 0.36527067061675195,
      "grad_norm": 1.0804884719240444,
      "learning_rate": 9.199601262211985e-06,
      "loss": 0.4108231544494629,
      "num_input_tokens_seen": 88845376,
      "step": 5425,
      "train_runtime": 10437.1345,
      "train_tokens_per_second": 8512.43
    },
    {
      "epoch": 0.3656073256127121,
      "grad_norm": 1.0917922009905918,
      "learning_rate": 9.19816570567507e-06,
      "loss": 0.4177213668823242,
      "num_input_tokens_seen": 88927296,
      "step": 5430,
      "train_runtime": 10446.844,
      "train_tokens_per_second": 8512.36
    },
    {
      "epoch": 0.3659439806086722,
      "grad_norm": 1.33829719085257,
      "learning_rate": 9.196728975134503e-06,
      "loss": 0.4484572410583496,
      "num_input_tokens_seen": 89009216,
      "step": 5435,
      "train_runtime": 10456.5234,
      "train_tokens_per_second": 8512.315
    },
    {
      "epoch": 0.36628063560463237,
      "grad_norm": 1.1610509814529502,
      "learning_rate": 9.195291070992061e-06,
      "loss": 0.40694737434387207,
      "num_input_tokens_seen": 89091136,
      "step": 5440,
      "train_runtime": 10466.0641,
      "train_tokens_per_second": 8512.382
    },
    {
      "epoch": 0.3666172906005925,
      "grad_norm": 1.1310394593182789,
      "learning_rate": 9.193851993649849e-06,
      "loss": 0.4257976531982422,
      "num_input_tokens_seen": 89173056,
      "step": 5445,
      "train_runtime": 10475.9209,
      "train_tokens_per_second": 8512.193
    },
    {
      "epoch": 0.36695394559655264,
      "grad_norm": 1.1601240740198318,
      "learning_rate": 9.1924117435103e-06,
      "loss": 0.39375321865081786,
      "num_input_tokens_seen": 89254976,
      "step": 5450,
      "train_runtime": 10485.6098,
      "train_tokens_per_second": 8512.14
    },
    {
      "epoch": 0.3672906005925128,
      "grad_norm": 1.1950045012994264,
      "learning_rate": 9.190970320976176e-06,
      "loss": 0.39046943187713623,
      "num_input_tokens_seen": 89336896,
      "step": 5455,
      "train_runtime": 10495.3004,
      "train_tokens_per_second": 8512.086
    },
    {
      "epoch": 0.3676272555884729,
      "grad_norm": 1.2739860398216893,
      "learning_rate": 9.189527726450565e-06,
      "loss": 0.4665255069732666,
      "num_input_tokens_seen": 89418816,
      "step": 5460,
      "train_runtime": 10504.9885,
      "train_tokens_per_second": 8512.034
    },
    {
      "epoch": 0.36796391058443306,
      "grad_norm": 0.9766110173298467,
      "learning_rate": 9.188083960336885e-06,
      "loss": 0.3879652976989746,
      "num_input_tokens_seen": 89500736,
      "step": 5465,
      "train_runtime": 10514.6303,
      "train_tokens_per_second": 8512.019
    },
    {
      "epoch": 0.3683005655803932,
      "grad_norm": 1.0011693732285887,
      "learning_rate": 9.186639023038879e-06,
      "loss": 0.4169576168060303,
      "num_input_tokens_seen": 89582656,
      "step": 5470,
      "train_runtime": 10524.1548,
      "train_tokens_per_second": 8512.1
    },
    {
      "epoch": 0.36863722057635334,
      "grad_norm": 1.1929752082039538,
      "learning_rate": 9.185192914960622e-06,
      "loss": 0.40958828926086427,
      "num_input_tokens_seen": 89664576,
      "step": 5475,
      "train_runtime": 10533.6651,
      "train_tokens_per_second": 8512.192
    },
    {
      "epoch": 0.3689738755723135,
      "grad_norm": 1.2677276166185194,
      "learning_rate": 9.183745636506511e-06,
      "loss": 0.39518964290618896,
      "num_input_tokens_seen": 89746496,
      "step": 5480,
      "train_runtime": 10543.1919,
      "train_tokens_per_second": 8512.27
    },
    {
      "epoch": 0.3693105305682736,
      "grad_norm": 1.2058420746522143,
      "learning_rate": 9.182297188081274e-06,
      "loss": 0.43999524116516114,
      "num_input_tokens_seen": 89828416,
      "step": 5485,
      "train_runtime": 10552.894,
      "train_tokens_per_second": 8512.207
    },
    {
      "epoch": 0.36964718556423376,
      "grad_norm": 1.1167152759310661,
      "learning_rate": 9.180847570089964e-06,
      "loss": 0.401684045791626,
      "num_input_tokens_seen": 89910336,
      "step": 5490,
      "train_runtime": 10562.5363,
      "train_tokens_per_second": 8512.192
    },
    {
      "epoch": 0.3699838405601939,
      "grad_norm": 1.2394318549741439,
      "learning_rate": 9.179396782937965e-06,
      "loss": 0.5003046035766602,
      "num_input_tokens_seen": 89992256,
      "step": 5495,
      "train_runtime": 10572.254,
      "train_tokens_per_second": 8512.116
    },
    {
      "epoch": 0.37032049555615404,
      "grad_norm": 1.2449500971231966,
      "learning_rate": 9.17794482703098e-06,
      "loss": 0.48639936447143556,
      "num_input_tokens_seen": 90074176,
      "step": 5500,
      "train_runtime": 10581.9377,
      "train_tokens_per_second": 8512.068
    },
    {
      "epoch": 0.3706571505521142,
      "grad_norm": 1.2256891631983617,
      "learning_rate": 9.176491702775049e-06,
      "loss": 0.43704586029052733,
      "num_input_tokens_seen": 90156096,
      "step": 5505,
      "train_runtime": 10591.5883,
      "train_tokens_per_second": 8512.047
    },
    {
      "epoch": 0.3709938055480743,
      "grad_norm": 1.1574099079344593,
      "learning_rate": 9.17503741057653e-06,
      "loss": 0.45391025543212893,
      "num_input_tokens_seen": 90238016,
      "step": 5510,
      "train_runtime": 10601.2954,
      "train_tokens_per_second": 8511.98
    },
    {
      "epoch": 0.37133046054403446,
      "grad_norm": 1.3087373071061237,
      "learning_rate": 9.173581950842111e-06,
      "loss": 0.4313360214233398,
      "num_input_tokens_seen": 90319936,
      "step": 5515,
      "train_runtime": 10610.9856,
      "train_tokens_per_second": 8511.927
    },
    {
      "epoch": 0.3716671155399946,
      "grad_norm": 1.4264934006028989,
      "learning_rate": 9.172125323978811e-06,
      "loss": 0.40774807929992674,
      "num_input_tokens_seen": 90401856,
      "step": 5520,
      "train_runtime": 10620.6007,
      "train_tokens_per_second": 8511.934
    },
    {
      "epoch": 0.37200377053595474,
      "grad_norm": 1.1744152772772425,
      "learning_rate": 9.170667530393966e-06,
      "loss": 0.4286545753479004,
      "num_input_tokens_seen": 90483776,
      "step": 5525,
      "train_runtime": 10630.273,
      "train_tokens_per_second": 8511.896
    },
    {
      "epoch": 0.3723404255319149,
      "grad_norm": 1.1437863987696388,
      "learning_rate": 9.169208570495247e-06,
      "loss": 0.3891228437423706,
      "num_input_tokens_seen": 90565696,
      "step": 5530,
      "train_runtime": 10639.7867,
      "train_tokens_per_second": 8511.984
    },
    {
      "epoch": 0.372677080527875,
      "grad_norm": 1.090326123495595,
      "learning_rate": 9.167748444690646e-06,
      "loss": 0.41758337020874026,
      "num_input_tokens_seen": 90647616,
      "step": 5535,
      "train_runtime": 10649.4179,
      "train_tokens_per_second": 8511.978
    },
    {
      "epoch": 0.37301373552383515,
      "grad_norm": 1.0361597075960158,
      "learning_rate": 9.16628715338848e-06,
      "loss": 0.4008808135986328,
      "num_input_tokens_seen": 90729536,
      "step": 5540,
      "train_runtime": 10658.9948,
      "train_tokens_per_second": 8512.016
    },
    {
      "epoch": 0.3733503905197953,
      "grad_norm": 1.0021496100532417,
      "learning_rate": 9.164824696997398e-06,
      "loss": 0.41350679397583007,
      "num_input_tokens_seen": 90811456,
      "step": 5545,
      "train_runtime": 10668.5525,
      "train_tokens_per_second": 8512.069
    },
    {
      "epoch": 0.37368704551575543,
      "grad_norm": 0.9959584078980535,
      "learning_rate": 9.163361075926371e-06,
      "loss": 0.482531213760376,
      "num_input_tokens_seen": 90893376,
      "step": 5550,
      "train_runtime": 10678.1936,
      "train_tokens_per_second": 8512.055
    },
    {
      "epoch": 0.3740237005117156,
      "grad_norm": 1.0100252602494266,
      "learning_rate": 9.161896290584695e-06,
      "loss": 0.40149974822998047,
      "num_input_tokens_seen": 90975296,
      "step": 5555,
      "train_runtime": 10687.8117,
      "train_tokens_per_second": 8512.06
    },
    {
      "epoch": 0.3743603555076757,
      "grad_norm": 1.1190588446883523,
      "learning_rate": 9.160430341381991e-06,
      "loss": 0.40807452201843264,
      "num_input_tokens_seen": 91057216,
      "step": 5560,
      "train_runtime": 10697.3179,
      "train_tokens_per_second": 8512.154
    },
    {
      "epoch": 0.37469701050363585,
      "grad_norm": 1.0425294645177432,
      "learning_rate": 9.15896322872821e-06,
      "loss": 0.41860475540161135,
      "num_input_tokens_seen": 91139016,
      "step": 5565,
      "train_runtime": 10707.4659,
      "train_tokens_per_second": 8511.726
    },
    {
      "epoch": 0.375033665499596,
      "grad_norm": 1.2937936366551155,
      "learning_rate": 9.157494953033625e-06,
      "loss": 0.4222146987915039,
      "num_input_tokens_seen": 91220936,
      "step": 5570,
      "train_runtime": 10717.0678,
      "train_tokens_per_second": 8511.744
    },
    {
      "epoch": 0.37537032049555613,
      "grad_norm": 1.1366591542151088,
      "learning_rate": 9.156025514708832e-06,
      "loss": 0.41307382583618163,
      "num_input_tokens_seen": 91302856,
      "step": 5575,
      "train_runtime": 10726.5818,
      "train_tokens_per_second": 8511.831
    },
    {
      "epoch": 0.37570697549151627,
      "grad_norm": 1.3970896017517214,
      "learning_rate": 9.154554914164757e-06,
      "loss": 0.42003374099731444,
      "num_input_tokens_seen": 91384776,
      "step": 5580,
      "train_runtime": 10736.204,
      "train_tokens_per_second": 8511.833
    },
    {
      "epoch": 0.3760436304874764,
      "grad_norm": 1.161663109446674,
      "learning_rate": 9.153083151812649e-06,
      "loss": 0.38340439796447756,
      "num_input_tokens_seen": 91466696,
      "step": 5585,
      "train_runtime": 10745.8289,
      "train_tokens_per_second": 8511.833
    },
    {
      "epoch": 0.37638028548343655,
      "grad_norm": 1.1038602560376878,
      "learning_rate": 9.151610228064078e-06,
      "loss": 0.4574557304382324,
      "num_input_tokens_seen": 91548616,
      "step": 5590,
      "train_runtime": 10755.3321,
      "train_tokens_per_second": 8511.928
    },
    {
      "epoch": 0.3767169404793967,
      "grad_norm": 1.1967060674236822,
      "learning_rate": 9.150136143330949e-06,
      "loss": 0.4666589260101318,
      "num_input_tokens_seen": 91630536,
      "step": 5595,
      "train_runtime": 10764.9303,
      "train_tokens_per_second": 8511.949
    },
    {
      "epoch": 0.37705359547535683,
      "grad_norm": 1.1283696867789779,
      "learning_rate": 9.14866089802548e-06,
      "loss": 0.39041264057159425,
      "num_input_tokens_seen": 91712456,
      "step": 5600,
      "train_runtime": 10774.5065,
      "train_tokens_per_second": 8511.987
    },
    {
      "epoch": 0.37739025047131697,
      "grad_norm": 1.192382642711143,
      "learning_rate": 9.147184492560219e-06,
      "loss": 0.3695479393005371,
      "num_input_tokens_seen": 91794376,
      "step": 5605,
      "train_runtime": 10784.0134,
      "train_tokens_per_second": 8512.079
    },
    {
      "epoch": 0.3777269054672771,
      "grad_norm": 1.082924516679474,
      "learning_rate": 9.14570692734804e-06,
      "loss": 0.3999659061431885,
      "num_input_tokens_seen": 91876296,
      "step": 5610,
      "train_runtime": 10793.6346,
      "train_tokens_per_second": 8512.081
    },
    {
      "epoch": 0.37806356046323725,
      "grad_norm": 1.41559716830687,
      "learning_rate": 9.144228202802137e-06,
      "loss": 0.4685948371887207,
      "num_input_tokens_seen": 91958216,
      "step": 5615,
      "train_runtime": 10803.2303,
      "train_tokens_per_second": 8512.104
    },
    {
      "epoch": 0.37840021545919744,
      "grad_norm": 1.1783679569803516,
      "learning_rate": 9.142748319336033e-06,
      "loss": 0.4041193962097168,
      "num_input_tokens_seen": 92040136,
      "step": 5620,
      "train_runtime": 10812.7536,
      "train_tokens_per_second": 8512.183
    },
    {
      "epoch": 0.3787368704551576,
      "grad_norm": 1.0441532702145506,
      "learning_rate": 9.141267277363573e-06,
      "loss": 0.39581828117370604,
      "num_input_tokens_seen": 92122056,
      "step": 5625,
      "train_runtime": 10822.3342,
      "train_tokens_per_second": 8512.217
    },
    {
      "epoch": 0.3790735254511177,
      "grad_norm": 1.1218014634015587,
      "learning_rate": 9.139785077298921e-06,
      "loss": 0.3945466041564941,
      "num_input_tokens_seen": 92203976,
      "step": 5630,
      "train_runtime": 10831.9321,
      "train_tokens_per_second": 8512.237
    },
    {
      "epoch": 0.37941018044707786,
      "grad_norm": 1.2042247420393566,
      "learning_rate": 9.138301719556577e-06,
      "loss": 0.4022211074829102,
      "num_input_tokens_seen": 92285896,
      "step": 5635,
      "train_runtime": 10841.4808,
      "train_tokens_per_second": 8512.296
    },
    {
      "epoch": 0.379746835443038,
      "grad_norm": 1.2301612092450598,
      "learning_rate": 9.136817204551352e-06,
      "loss": 0.4323399543762207,
      "num_input_tokens_seen": 92367816,
      "step": 5640,
      "train_runtime": 10851.0415,
      "train_tokens_per_second": 8512.346
    },
    {
      "epoch": 0.38008349043899814,
      "grad_norm": 1.1090802529432227,
      "learning_rate": 9.135331532698385e-06,
      "loss": 0.439035701751709,
      "num_input_tokens_seen": 92449736,
      "step": 5645,
      "train_runtime": 10860.6789,
      "train_tokens_per_second": 8512.335
    },
    {
      "epoch": 0.3804201454349583,
      "grad_norm": 1.0324424952970102,
      "learning_rate": 9.133844704413144e-06,
      "loss": 0.4146732330322266,
      "num_input_tokens_seen": 92530560,
      "step": 5650,
      "train_runtime": 10870.3843,
      "train_tokens_per_second": 8512.17
    },
    {
      "epoch": 0.3807568004309184,
      "grad_norm": 0.9080942415195206,
      "learning_rate": 9.132356720111413e-06,
      "loss": 0.3847212553024292,
      "num_input_tokens_seen": 92612480,
      "step": 5655,
      "train_runtime": 10879.9467,
      "train_tokens_per_second": 8512.218
    },
    {
      "epoch": 0.38109345542687856,
      "grad_norm": 1.0376084888478796,
      "learning_rate": 9.1308675802093e-06,
      "loss": 0.4146879196166992,
      "num_input_tokens_seen": 92694400,
      "step": 5660,
      "train_runtime": 10889.5771,
      "train_tokens_per_second": 8512.213
    },
    {
      "epoch": 0.3814301104228387,
      "grad_norm": 1.3964350921728652,
      "learning_rate": 9.129377285123241e-06,
      "loss": 0.430298900604248,
      "num_input_tokens_seen": 92776320,
      "step": 5665,
      "train_runtime": 10899.6697,
      "train_tokens_per_second": 8511.847
    },
    {
      "epoch": 0.38176676541879884,
      "grad_norm": 1.199762499149907,
      "learning_rate": 9.127885835269996e-06,
      "loss": 0.43253579139709475,
      "num_input_tokens_seen": 92858240,
      "step": 5670,
      "train_runtime": 10909.245,
      "train_tokens_per_second": 8511.885
    },
    {
      "epoch": 0.382103420414759,
      "grad_norm": 1.2900121708917516,
      "learning_rate": 9.126393231066636e-06,
      "loss": 0.40099368095397947,
      "num_input_tokens_seen": 92940160,
      "step": 5675,
      "train_runtime": 10918.8458,
      "train_tokens_per_second": 8511.903
    },
    {
      "epoch": 0.3824400754107191,
      "grad_norm": 1.053652498581128,
      "learning_rate": 9.124899472930568e-06,
      "loss": 0.426371431350708,
      "num_input_tokens_seen": 93022080,
      "step": 5680,
      "train_runtime": 10928.3615,
      "train_tokens_per_second": 8511.988
    },
    {
      "epoch": 0.38277673040667926,
      "grad_norm": 1.1149195448569842,
      "learning_rate": 9.123404561279517e-06,
      "loss": 0.4199851989746094,
      "num_input_tokens_seen": 93104000,
      "step": 5685,
      "train_runtime": 10937.9151,
      "train_tokens_per_second": 8512.043
    },
    {
      "epoch": 0.3831133854026394,
      "grad_norm": 1.0998697852269268,
      "learning_rate": 9.121908496531527e-06,
      "loss": 0.41463699340820315,
      "num_input_tokens_seen": 93185920,
      "step": 5690,
      "train_runtime": 10947.5146,
      "train_tokens_per_second": 8512.062
    },
    {
      "epoch": 0.38345004039859953,
      "grad_norm": 0.9906219330852912,
      "learning_rate": 9.120411279104971e-06,
      "loss": 0.405642032623291,
      "num_input_tokens_seen": 93267840,
      "step": 5695,
      "train_runtime": 10957.0718,
      "train_tokens_per_second": 8512.114
    },
    {
      "epoch": 0.3837866953945597,
      "grad_norm": 1.1336862265108398,
      "learning_rate": 9.118912909418538e-06,
      "loss": 0.3942866802215576,
      "num_input_tokens_seen": 93349760,
      "step": 5700,
      "train_runtime": 10966.6484,
      "train_tokens_per_second": 8512.15
    },
    {
      "epoch": 0.3841233503905198,
      "grad_norm": 1.09920556795178,
      "learning_rate": 9.117413387891246e-06,
      "loss": 0.4711200714111328,
      "num_input_tokens_seen": 93431680,
      "step": 5705,
      "train_runtime": 10976.2505,
      "train_tokens_per_second": 8512.167
    },
    {
      "epoch": 0.38446000538647995,
      "grad_norm": 1.3981393502651933,
      "learning_rate": 9.11591271494243e-06,
      "loss": 0.4518772602081299,
      "num_input_tokens_seen": 93513600,
      "step": 5710,
      "train_runtime": 10985.8455,
      "train_tokens_per_second": 8512.19
    },
    {
      "epoch": 0.3847966603824401,
      "grad_norm": 1.0270778967946943,
      "learning_rate": 9.114410890991748e-06,
      "loss": 0.408121395111084,
      "num_input_tokens_seen": 93595520,
      "step": 5715,
      "train_runtime": 10995.3605,
      "train_tokens_per_second": 8512.274
    },
    {
      "epoch": 0.38513331537840023,
      "grad_norm": 0.8445338597529592,
      "learning_rate": 9.112907916459177e-06,
      "loss": 0.3930232048034668,
      "num_input_tokens_seen": 93677440,
      "step": 5720,
      "train_runtime": 11004.9965,
      "train_tokens_per_second": 8512.264
    },
    {
      "epoch": 0.38546997037436037,
      "grad_norm": 1.1923436615310439,
      "learning_rate": 9.111403791765025e-06,
      "loss": 0.45052785873413087,
      "num_input_tokens_seen": 93759360,
      "step": 5725,
      "train_runtime": 11014.5929,
      "train_tokens_per_second": 8512.286
    },
    {
      "epoch": 0.3858066253703205,
      "grad_norm": 1.5659869827922395,
      "learning_rate": 9.109898517329914e-06,
      "loss": 0.43639512062072755,
      "num_input_tokens_seen": 93841280,
      "step": 5730,
      "train_runtime": 11024.1221,
      "train_tokens_per_second": 8512.359
    },
    {
      "epoch": 0.38614328036628065,
      "grad_norm": 1.1708393708558016,
      "learning_rate": 9.108392093574785e-06,
      "loss": 0.41306285858154296,
      "num_input_tokens_seen": 93923200,
      "step": 5735,
      "train_runtime": 11033.7297,
      "train_tokens_per_second": 8512.371
    },
    {
      "epoch": 0.3864799353622408,
      "grad_norm": 0.9330561047430521,
      "learning_rate": 9.10688452092091e-06,
      "loss": 0.35896849632263184,
      "num_input_tokens_seen": 94005120,
      "step": 5740,
      "train_runtime": 11043.35,
      "train_tokens_per_second": 8512.373
    },
    {
      "epoch": 0.38681659035820093,
      "grad_norm": 1.0530373175127863,
      "learning_rate": 9.105375799789874e-06,
      "loss": 0.4388981819152832,
      "num_input_tokens_seen": 94087040,
      "step": 5745,
      "train_runtime": 11052.8607,
      "train_tokens_per_second": 8512.46
    },
    {
      "epoch": 0.38715324535416107,
      "grad_norm": 1.2775338659898838,
      "learning_rate": 9.103865930603586e-06,
      "loss": 0.4068880558013916,
      "num_input_tokens_seen": 94168960,
      "step": 5750,
      "train_runtime": 11062.4734,
      "train_tokens_per_second": 8512.469
    },
    {
      "epoch": 0.3874899003501212,
      "grad_norm": 1.2967549900970143,
      "learning_rate": 9.102354913784279e-06,
      "loss": 0.4256744384765625,
      "num_input_tokens_seen": 94250880,
      "step": 5755,
      "train_runtime": 11072.0715,
      "train_tokens_per_second": 8512.488
    },
    {
      "epoch": 0.38782655534608135,
      "grad_norm": 0.9537628129084056,
      "learning_rate": 9.100842749754499e-06,
      "loss": 0.37769036293029784,
      "num_input_tokens_seen": 94332800,
      "step": 5760,
      "train_runtime": 11081.5788,
      "train_tokens_per_second": 8512.578
    },
    {
      "epoch": 0.3881632103420415,
      "grad_norm": 1.00578867752884,
      "learning_rate": 9.099329438937122e-06,
      "loss": 0.4150383949279785,
      "num_input_tokens_seen": 94414720,
      "step": 5765,
      "train_runtime": 11091.1797,
      "train_tokens_per_second": 8512.595
    },
    {
      "epoch": 0.3884998653380016,
      "grad_norm": 1.0585680981754135,
      "learning_rate": 9.09781498175534e-06,
      "loss": 0.39490957260131837,
      "num_input_tokens_seen": 94496640,
      "step": 5770,
      "train_runtime": 11100.8037,
      "train_tokens_per_second": 8512.594
    },
    {
      "epoch": 0.38883652033396177,
      "grad_norm": 1.1355586787999385,
      "learning_rate": 9.096299378632666e-06,
      "loss": 0.41808195114135743,
      "num_input_tokens_seen": 94578560,
      "step": 5775,
      "train_runtime": 11110.3115,
      "train_tokens_per_second": 8512.683
    },
    {
      "epoch": 0.3891731753299219,
      "grad_norm": 1.1861933349870886,
      "learning_rate": 9.09478262999293e-06,
      "loss": 0.4557765007019043,
      "num_input_tokens_seen": 94660480,
      "step": 5780,
      "train_runtime": 11119.8778,
      "train_tokens_per_second": 8512.727
    },
    {
      "epoch": 0.38950983032588204,
      "grad_norm": 1.1263231514247836,
      "learning_rate": 9.09326473626029e-06,
      "loss": 0.41611666679382325,
      "num_input_tokens_seen": 94742400,
      "step": 5785,
      "train_runtime": 11129.5284,
      "train_tokens_per_second": 8512.706
    },
    {
      "epoch": 0.3898464853218422,
      "grad_norm": 1.4485733008088053,
      "learning_rate": 9.091745697859218e-06,
      "loss": 0.44564247131347656,
      "num_input_tokens_seen": 94824320,
      "step": 5790,
      "train_runtime": 11139.0519,
      "train_tokens_per_second": 8512.782
    },
    {
      "epoch": 0.3901831403178023,
      "grad_norm": 1.3601108040982433,
      "learning_rate": 9.090225515214512e-06,
      "loss": 0.4178616523742676,
      "num_input_tokens_seen": 94906240,
      "step": 5795,
      "train_runtime": 11148.6944,
      "train_tokens_per_second": 8512.767
    },
    {
      "epoch": 0.39051979531376246,
      "grad_norm": 1.3198211590421085,
      "learning_rate": 9.088704188751281e-06,
      "loss": 0.45279922485351565,
      "num_input_tokens_seen": 94988160,
      "step": 5800,
      "train_runtime": 11158.3107,
      "train_tokens_per_second": 8512.772
    },
    {
      "epoch": 0.3908564503097226,
      "grad_norm": 1.1225480461811317,
      "learning_rate": 9.087181718894963e-06,
      "loss": 0.41368770599365234,
      "num_input_tokens_seen": 95070080,
      "step": 5805,
      "train_runtime": 11167.8183,
      "train_tokens_per_second": 8512.861
    },
    {
      "epoch": 0.39119310530568274,
      "grad_norm": 1.0607676650003066,
      "learning_rate": 9.085658106071309e-06,
      "loss": 0.4048016548156738,
      "num_input_tokens_seen": 95151600,
      "step": 5810,
      "train_runtime": 11177.7677,
      "train_tokens_per_second": 8512.576
    },
    {
      "epoch": 0.3915297603016429,
      "grad_norm": 1.2809496637725877,
      "learning_rate": 9.084133350706394e-06,
      "loss": 0.44188880920410156,
      "num_input_tokens_seen": 95233520,
      "step": 5815,
      "train_runtime": 11187.3857,
      "train_tokens_per_second": 8512.58
    },
    {
      "epoch": 0.391866415297603,
      "grad_norm": 1.031366295994401,
      "learning_rate": 9.082607453226613e-06,
      "loss": 0.4029900550842285,
      "num_input_tokens_seen": 95315440,
      "step": 5820,
      "train_runtime": 11196.8887,
      "train_tokens_per_second": 8512.672
    },
    {
      "epoch": 0.39220307029356316,
      "grad_norm": 1.2033503242987071,
      "learning_rate": 9.081080414058674e-06,
      "loss": 0.40069580078125,
      "num_input_tokens_seen": 95397360,
      "step": 5825,
      "train_runtime": 11206.4985,
      "train_tokens_per_second": 8512.682
    },
    {
      "epoch": 0.3925397252895233,
      "grad_norm": 1.273751186901139,
      "learning_rate": 9.079552233629612e-06,
      "loss": 0.3835675954818726,
      "num_input_tokens_seen": 95479280,
      "step": 5830,
      "train_runtime": 11216.1087,
      "train_tokens_per_second": 8512.692
    },
    {
      "epoch": 0.39287638028548344,
      "grad_norm": 0.9874975533555207,
      "learning_rate": 9.078022912366776e-06,
      "loss": 0.39106600284576415,
      "num_input_tokens_seen": 95561200,
      "step": 5835,
      "train_runtime": 11225.6184,
      "train_tokens_per_second": 8512.778
    },
    {
      "epoch": 0.3932130352814436,
      "grad_norm": 1.2408756624446258,
      "learning_rate": 9.076492450697835e-06,
      "loss": 0.45532703399658203,
      "num_input_tokens_seen": 95643120,
      "step": 5840,
      "train_runtime": 11235.2291,
      "train_tokens_per_second": 8512.788
    },
    {
      "epoch": 0.3935496902774037,
      "grad_norm": 1.2557958737671358,
      "learning_rate": 9.074960849050782e-06,
      "loss": 0.4362895965576172,
      "num_input_tokens_seen": 95724800,
      "step": 5845,
      "train_runtime": 11245.3357,
      "train_tokens_per_second": 8512.4
    },
    {
      "epoch": 0.39388634527336386,
      "grad_norm": 1.0595569613343758,
      "learning_rate": 9.07342810785392e-06,
      "loss": 0.383693265914917,
      "num_input_tokens_seen": 95806720,
      "step": 5850,
      "train_runtime": 11254.8809,
      "train_tokens_per_second": 8512.46
    },
    {
      "epoch": 0.394223000269324,
      "grad_norm": 1.1761599342885374,
      "learning_rate": 9.071894227535878e-06,
      "loss": 0.4237061023712158,
      "num_input_tokens_seen": 95888640,
      "step": 5855,
      "train_runtime": 11264.4402,
      "train_tokens_per_second": 8512.508
    },
    {
      "epoch": 0.39455965526528414,
      "grad_norm": 1.1258940441752823,
      "learning_rate": 9.070359208525598e-06,
      "loss": 0.43930416107177733,
      "num_input_tokens_seen": 95970560,
      "step": 5860,
      "train_runtime": 11274.044,
      "train_tokens_per_second": 8512.523
    },
    {
      "epoch": 0.3948963102612443,
      "grad_norm": 0.9924674032631361,
      "learning_rate": 9.068823051252346e-06,
      "loss": 0.4177695274353027,
      "num_input_tokens_seen": 96052480,
      "step": 5865,
      "train_runtime": 11283.6085,
      "train_tokens_per_second": 8512.568
    },
    {
      "epoch": 0.3952329652572044,
      "grad_norm": 0.9758036304029527,
      "learning_rate": 9.067285756145702e-06,
      "loss": 0.3710019111633301,
      "num_input_tokens_seen": 96134400,
      "step": 5870,
      "train_runtime": 11293.1728,
      "train_tokens_per_second": 8512.612
    },
    {
      "epoch": 0.39556962025316456,
      "grad_norm": 1.1295092515526661,
      "learning_rate": 9.065747323635565e-06,
      "loss": 0.3830176830291748,
      "num_input_tokens_seen": 96216320,
      "step": 5875,
      "train_runtime": 11302.7972,
      "train_tokens_per_second": 8512.611
    },
    {
      "epoch": 0.3959062752491247,
      "grad_norm": 1.2805726624588074,
      "learning_rate": 9.064207754152154e-06,
      "loss": 0.46076064109802245,
      "num_input_tokens_seen": 96297784,
      "step": 5880,
      "train_runtime": 11312.3457,
      "train_tokens_per_second": 8512.627
    },
    {
      "epoch": 0.39624293024508483,
      "grad_norm": 0.9822997377649417,
      "learning_rate": 9.062667048126005e-06,
      "loss": 0.363901424407959,
      "num_input_tokens_seen": 96379704,
      "step": 5885,
      "train_runtime": 11321.9264,
      "train_tokens_per_second": 8512.659
    },
    {
      "epoch": 0.396579585241045,
      "grad_norm": 1.1388131719304027,
      "learning_rate": 9.061125205987971e-06,
      "loss": 0.4215838432312012,
      "num_input_tokens_seen": 96461624,
      "step": 5890,
      "train_runtime": 11331.5574,
      "train_tokens_per_second": 8512.654
    },
    {
      "epoch": 0.3969162402370051,
      "grad_norm": 1.1117540323601336,
      "learning_rate": 9.059582228169222e-06,
      "loss": 0.4367208480834961,
      "num_input_tokens_seen": 96543544,
      "step": 5895,
      "train_runtime": 11341.1052,
      "train_tokens_per_second": 8512.71
    },
    {
      "epoch": 0.39725289523296525,
      "grad_norm": 1.1025993214578205,
      "learning_rate": 9.058038115101248e-06,
      "loss": 0.45531654357910156,
      "num_input_tokens_seen": 96625464,
      "step": 5900,
      "train_runtime": 11350.6659,
      "train_tokens_per_second": 8512.757
    },
    {
      "epoch": 0.3975895502289254,
      "grad_norm": 1.0050039707162621,
      "learning_rate": 9.056492867215857e-06,
      "loss": 0.38019757270812987,
      "num_input_tokens_seen": 96707384,
      "step": 5905,
      "train_runtime": 11360.2878,
      "train_tokens_per_second": 8512.758
    },
    {
      "epoch": 0.39792620522488553,
      "grad_norm": 1.26060646283026,
      "learning_rate": 9.05494648494517e-06,
      "loss": 0.45593814849853515,
      "num_input_tokens_seen": 96789288,
      "step": 5910,
      "train_runtime": 11370.4236,
      "train_tokens_per_second": 8512.373
    },
    {
      "epoch": 0.39826286022084567,
      "grad_norm": 1.0556288942441687,
      "learning_rate": 9.053398968721629e-06,
      "loss": 0.43929424285888674,
      "num_input_tokens_seen": 96871208,
      "step": 5915,
      "train_runtime": 11379.936,
      "train_tokens_per_second": 8512.456
    },
    {
      "epoch": 0.3985995152168058,
      "grad_norm": 1.2091136426565192,
      "learning_rate": 9.05185031897799e-06,
      "loss": 0.43143501281738283,
      "num_input_tokens_seen": 96953128,
      "step": 5920,
      "train_runtime": 11389.5537,
      "train_tokens_per_second": 8512.461
    },
    {
      "epoch": 0.39893617021276595,
      "grad_norm": 1.2720185982114998,
      "learning_rate": 9.050300536147331e-06,
      "loss": 0.4676380157470703,
      "num_input_tokens_seen": 97035048,
      "step": 5925,
      "train_runtime": 11399.1585,
      "train_tokens_per_second": 8512.475
    },
    {
      "epoch": 0.3992728252087261,
      "grad_norm": 1.302161009272664,
      "learning_rate": 9.048749620663044e-06,
      "loss": 0.4114323616027832,
      "num_input_tokens_seen": 97116968,
      "step": 5930,
      "train_runtime": 11408.6546,
      "train_tokens_per_second": 8512.57
    },
    {
      "epoch": 0.39960948020468623,
      "grad_norm": 1.2880532396784508,
      "learning_rate": 9.047197572958834e-06,
      "loss": 0.39770643711090087,
      "num_input_tokens_seen": 97198888,
      "step": 5935,
      "train_runtime": 11418.2559,
      "train_tokens_per_second": 8512.586
    },
    {
      "epoch": 0.39994613520064637,
      "grad_norm": 1.2135275761093096,
      "learning_rate": 9.04564439346873e-06,
      "loss": 0.4088254451751709,
      "num_input_tokens_seen": 97279528,
      "step": 5940,
      "train_runtime": 11427.9454,
      "train_tokens_per_second": 8512.425
    },
    {
      "epoch": 0.4002827901966065,
      "grad_norm": 1.4046834973211288,
      "learning_rate": 9.044090082627071e-06,
      "loss": 0.4238097667694092,
      "num_input_tokens_seen": 97360936,
      "step": 5945,
      "train_runtime": 11437.7159,
      "train_tokens_per_second": 8512.271
    },
    {
      "epoch": 0.40061944519256665,
      "grad_norm": 1.0150194324109,
      "learning_rate": 9.042534640868514e-06,
      "loss": 0.4276543617248535,
      "num_input_tokens_seen": 97442856,
      "step": 5950,
      "train_runtime": 11447.3306,
      "train_tokens_per_second": 8512.278
    },
    {
      "epoch": 0.4009561001885268,
      "grad_norm": 1.197329746154737,
      "learning_rate": 9.04097806862804e-06,
      "loss": 0.4448102951049805,
      "num_input_tokens_seen": 97524776,
      "step": 5955,
      "train_runtime": 11456.9457,
      "train_tokens_per_second": 8512.284
    },
    {
      "epoch": 0.4012927551844869,
      "grad_norm": 0.9961465474632097,
      "learning_rate": 9.03942036634093e-06,
      "loss": 0.4213907241821289,
      "num_input_tokens_seen": 97606696,
      "step": 5960,
      "train_runtime": 11466.4595,
      "train_tokens_per_second": 8512.366
    },
    {
      "epoch": 0.40162941018044707,
      "grad_norm": 0.9808034601779083,
      "learning_rate": 9.037861534442797e-06,
      "loss": 0.37805423736572263,
      "num_input_tokens_seen": 97688616,
      "step": 5965,
      "train_runtime": 11476.0847,
      "train_tokens_per_second": 8512.365
    },
    {
      "epoch": 0.4019660651764072,
      "grad_norm": 1.1105093601341856,
      "learning_rate": 9.036301573369563e-06,
      "loss": 0.4404754161834717,
      "num_input_tokens_seen": 97770536,
      "step": 5970,
      "train_runtime": 11485.7055,
      "train_tokens_per_second": 8512.367
    },
    {
      "epoch": 0.40230272017236735,
      "grad_norm": 1.362576549884914,
      "learning_rate": 9.034740483557465e-06,
      "loss": 0.4189460277557373,
      "num_input_tokens_seen": 97852456,
      "step": 5975,
      "train_runtime": 11495.2098,
      "train_tokens_per_second": 8512.455
    },
    {
      "epoch": 0.4026393751683275,
      "grad_norm": 1.1200046729585538,
      "learning_rate": 9.033178265443055e-06,
      "loss": 0.4327810287475586,
      "num_input_tokens_seen": 97934376,
      "step": 5980,
      "train_runtime": 11504.821,
      "train_tokens_per_second": 8512.464
    },
    {
      "epoch": 0.4029760301642876,
      "grad_norm": 1.049933541641451,
      "learning_rate": 9.031614919463206e-06,
      "loss": 0.4123408794403076,
      "num_input_tokens_seen": 98016296,
      "step": 5985,
      "train_runtime": 11514.4352,
      "train_tokens_per_second": 8512.471
    },
    {
      "epoch": 0.40331268516024776,
      "grad_norm": 1.0656886930633176,
      "learning_rate": 9.030050446055099e-06,
      "loss": 0.4099760055541992,
      "num_input_tokens_seen": 98098216,
      "step": 5990,
      "train_runtime": 11523.9794,
      "train_tokens_per_second": 8512.53
    },
    {
      "epoch": 0.4036493401562079,
      "grad_norm": 1.005307580333884,
      "learning_rate": 9.028484845656235e-06,
      "loss": 0.4078977584838867,
      "num_input_tokens_seen": 98180136,
      "step": 5995,
      "train_runtime": 11533.5667,
      "train_tokens_per_second": 8512.556
    },
    {
      "epoch": 0.40398599515216804,
      "grad_norm": 1.1467915690163224,
      "learning_rate": 9.026918118704431e-06,
      "loss": 0.43758745193481446,
      "num_input_tokens_seen": 98262056,
      "step": 6000,
      "train_runtime": 11543.2043,
      "train_tokens_per_second": 8512.546
    },
    {
      "epoch": 0.4043226501481282,
      "grad_norm": 1.1956672421524144,
      "learning_rate": 9.025350265637816e-06,
      "loss": 0.4288797378540039,
      "num_input_tokens_seen": 98343976,
      "step": 6005,
      "train_runtime": 11552.812,
      "train_tokens_per_second": 8512.557
    },
    {
      "epoch": 0.4046593051440883,
      "grad_norm": 0.9000669566279266,
      "learning_rate": 9.023781286894834e-06,
      "loss": 0.4210239887237549,
      "num_input_tokens_seen": 98425896,
      "step": 6010,
      "train_runtime": 11562.3683,
      "train_tokens_per_second": 8512.607
    },
    {
      "epoch": 0.40499596014004846,
      "grad_norm": 1.0585504351524944,
      "learning_rate": 9.022211182914247e-06,
      "loss": 0.38548600673675537,
      "num_input_tokens_seen": 98507232,
      "step": 6015,
      "train_runtime": 11572.236,
      "train_tokens_per_second": 8512.377
    },
    {
      "epoch": 0.4053326151360086,
      "grad_norm": 1.1653514830477745,
      "learning_rate": 9.020639954135128e-06,
      "loss": 0.38202776908874514,
      "num_input_tokens_seen": 98589152,
      "step": 6020,
      "train_runtime": 11581.7725,
      "train_tokens_per_second": 8512.441
    },
    {
      "epoch": 0.40566927013196874,
      "grad_norm": 1.1597936926569645,
      "learning_rate": 9.019067600996867e-06,
      "loss": 0.36632604598999025,
      "num_input_tokens_seen": 98671072,
      "step": 6025,
      "train_runtime": 11591.3878,
      "train_tokens_per_second": 8512.447
    },
    {
      "epoch": 0.4060059251279289,
      "grad_norm": 1.0892653628419777,
      "learning_rate": 9.017494123939169e-06,
      "loss": 0.435282039642334,
      "num_input_tokens_seen": 98752992,
      "step": 6030,
      "train_runtime": 11600.951,
      "train_tokens_per_second": 8512.491
    },
    {
      "epoch": 0.406342580123889,
      "grad_norm": 0.9746308331643626,
      "learning_rate": 9.015919523402048e-06,
      "loss": 0.3911391496658325,
      "num_input_tokens_seen": 98834912,
      "step": 6035,
      "train_runtime": 11610.5008,
      "train_tokens_per_second": 8512.545
    },
    {
      "epoch": 0.40667923511984916,
      "grad_norm": 1.0747565798192027,
      "learning_rate": 9.014343799825838e-06,
      "loss": 0.4226559638977051,
      "num_input_tokens_seen": 98916832,
      "step": 6040,
      "train_runtime": 11620.0606,
      "train_tokens_per_second": 8512.592
    },
    {
      "epoch": 0.4070158901158093,
      "grad_norm": 0.9976699648150039,
      "learning_rate": 9.012766953651186e-06,
      "loss": 0.4721692085266113,
      "num_input_tokens_seen": 98998752,
      "step": 6045,
      "train_runtime": 11629.8657,
      "train_tokens_per_second": 8512.459
    },
    {
      "epoch": 0.40735254511176944,
      "grad_norm": 1.0031226020780268,
      "learning_rate": 9.01118898531905e-06,
      "loss": 0.4350112438201904,
      "num_input_tokens_seen": 99080672,
      "step": 6050,
      "train_runtime": 11639.3716,
      "train_tokens_per_second": 8512.545
    },
    {
      "epoch": 0.4076892001077296,
      "grad_norm": 1.1227038363416773,
      "learning_rate": 9.009609895270708e-06,
      "loss": 0.3763844966888428,
      "num_input_tokens_seen": 99162592,
      "step": 6055,
      "train_runtime": 11648.9813,
      "train_tokens_per_second": 8512.555
    },
    {
      "epoch": 0.4080258551036897,
      "grad_norm": 1.7071377591610721,
      "learning_rate": 9.008029683947743e-06,
      "loss": 0.4471290588378906,
      "num_input_tokens_seen": 99244512,
      "step": 6060,
      "train_runtime": 11658.5438,
      "train_tokens_per_second": 8512.599
    },
    {
      "epoch": 0.40836251009964986,
      "grad_norm": 1.0471201780620407,
      "learning_rate": 9.006448351792057e-06,
      "loss": 0.4251726150512695,
      "num_input_tokens_seen": 99326432,
      "step": 6065,
      "train_runtime": 11668.1557,
      "train_tokens_per_second": 8512.608
    },
    {
      "epoch": 0.40869916509561,
      "grad_norm": 1.1721373671733846,
      "learning_rate": 9.004865899245864e-06,
      "loss": 0.4633072853088379,
      "num_input_tokens_seen": 99408352,
      "step": 6070,
      "train_runtime": 11677.6557,
      "train_tokens_per_second": 8512.698
    },
    {
      "epoch": 0.40903582009157013,
      "grad_norm": 1.1382670232957761,
      "learning_rate": 9.003282326751694e-06,
      "loss": 0.42792701721191406,
      "num_input_tokens_seen": 99490272,
      "step": 6075,
      "train_runtime": 11687.2638,
      "train_tokens_per_second": 8512.709
    },
    {
      "epoch": 0.4093724750875303,
      "grad_norm": 1.1362339845116283,
      "learning_rate": 9.001697634752387e-06,
      "loss": 0.4438138961791992,
      "num_input_tokens_seen": 99572192,
      "step": 6080,
      "train_runtime": 11696.8893,
      "train_tokens_per_second": 8512.707
    },
    {
      "epoch": 0.4097091300834904,
      "grad_norm": 1.1360411700715967,
      "learning_rate": 9.000111823691097e-06,
      "loss": 0.4127919673919678,
      "num_input_tokens_seen": 99654112,
      "step": 6085,
      "train_runtime": 11706.4002,
      "train_tokens_per_second": 8512.789
    },
    {
      "epoch": 0.41004578507945055,
      "grad_norm": 1.0909601021719197,
      "learning_rate": 8.99852489401129e-06,
      "loss": 0.44518446922302246,
      "num_input_tokens_seen": 99736032,
      "step": 6090,
      "train_runtime": 11715.8988,
      "train_tokens_per_second": 8512.879
    },
    {
      "epoch": 0.4103824400754107,
      "grad_norm": 1.131239370793832,
      "learning_rate": 8.996936846156748e-06,
      "loss": 0.3714694261550903,
      "num_input_tokens_seen": 99817952,
      "step": 6095,
      "train_runtime": 11725.5263,
      "train_tokens_per_second": 8512.876
    },
    {
      "epoch": 0.41071909507137083,
      "grad_norm": 1.1781137176697358,
      "learning_rate": 8.995347680571563e-06,
      "loss": 0.4163857936859131,
      "num_input_tokens_seen": 99899872,
      "step": 6100,
      "train_runtime": 11735.1483,
      "train_tokens_per_second": 8512.877
    },
    {
      "epoch": 0.41105575006733097,
      "grad_norm": 1.0699215166973135,
      "learning_rate": 8.993757397700137e-06,
      "loss": 0.43062877655029297,
      "num_input_tokens_seen": 99981792,
      "step": 6105,
      "train_runtime": 11744.6538,
      "train_tokens_per_second": 8512.962
    },
    {
      "epoch": 0.41139240506329117,
      "grad_norm": 1.29612237762739,
      "learning_rate": 8.99216599798719e-06,
      "loss": 0.45260276794433596,
      "num_input_tokens_seen": 100063712,
      "step": 6110,
      "train_runtime": 11754.1654,
      "train_tokens_per_second": 8513.043
    },
    {
      "epoch": 0.4117290600592513,
      "grad_norm": 1.204314804200267,
      "learning_rate": 8.990573481877753e-06,
      "loss": 0.4066497802734375,
      "num_input_tokens_seen": 100145632,
      "step": 6115,
      "train_runtime": 11763.767,
      "train_tokens_per_second": 8513.058
    },
    {
      "epoch": 0.41206571505521145,
      "grad_norm": 0.9977375746935455,
      "learning_rate": 8.988979849817167e-06,
      "loss": 0.36495859622955323,
      "num_input_tokens_seen": 100227400,
      "step": 6120,
      "train_runtime": 11773.3871,
      "train_tokens_per_second": 8513.047
    },
    {
      "epoch": 0.4124023700511716,
      "grad_norm": 1.749781734079108,
      "learning_rate": 8.987385102251085e-06,
      "loss": 0.39964005947113035,
      "num_input_tokens_seen": 100309320,
      "step": 6125,
      "train_runtime": 11782.8921,
      "train_tokens_per_second": 8513.132
    },
    {
      "epoch": 0.4127390250471317,
      "grad_norm": 1.2876611642340392,
      "learning_rate": 8.985789239625475e-06,
      "loss": 0.4424407958984375,
      "num_input_tokens_seen": 100391240,
      "step": 6130,
      "train_runtime": 11792.4187,
      "train_tokens_per_second": 8513.202
    },
    {
      "epoch": 0.41307568004309186,
      "grad_norm": 0.9667520355986258,
      "learning_rate": 8.984192262386613e-06,
      "loss": 0.36367321014404297,
      "num_input_tokens_seen": 100473160,
      "step": 6135,
      "train_runtime": 11802.0304,
      "train_tokens_per_second": 8513.21
    },
    {
      "epoch": 0.413412335039052,
      "grad_norm": 1.0223425403297663,
      "learning_rate": 8.98259417098109e-06,
      "loss": 0.41297225952148436,
      "num_input_tokens_seen": 100555080,
      "step": 6140,
      "train_runtime": 11811.655,
      "train_tokens_per_second": 8513.208
    },
    {
      "epoch": 0.41374899003501214,
      "grad_norm": 1.0232780848087106,
      "learning_rate": 8.980994965855808e-06,
      "loss": 0.4407472610473633,
      "num_input_tokens_seen": 100636352,
      "step": 6145,
      "train_runtime": 11821.4316,
      "train_tokens_per_second": 8513.043
    },
    {
      "epoch": 0.4140856450309723,
      "grad_norm": 1.0931562200027563,
      "learning_rate": 8.979394647457976e-06,
      "loss": 0.41681413650512694,
      "num_input_tokens_seen": 100718272,
      "step": 6150,
      "train_runtime": 11831.0251,
      "train_tokens_per_second": 8513.064
    },
    {
      "epoch": 0.4144223000269324,
      "grad_norm": 0.9853753914651011,
      "learning_rate": 8.97779321623512e-06,
      "loss": 0.3723767280578613,
      "num_input_tokens_seen": 100800192,
      "step": 6155,
      "train_runtime": 11840.6422,
      "train_tokens_per_second": 8513.068
    },
    {
      "epoch": 0.41475895502289256,
      "grad_norm": 0.9918009551418224,
      "learning_rate": 8.976190672635077e-06,
      "loss": 0.4010010242462158,
      "num_input_tokens_seen": 100882112,
      "step": 6160,
      "train_runtime": 11850.1996,
      "train_tokens_per_second": 8513.115
    },
    {
      "epoch": 0.4150956100188527,
      "grad_norm": 1.1916547145715357,
      "learning_rate": 8.974587017105991e-06,
      "loss": 0.442891263961792,
      "num_input_tokens_seen": 100964032,
      "step": 6165,
      "train_runtime": 11859.7156,
      "train_tokens_per_second": 8513.192
    },
    {
      "epoch": 0.41543226501481284,
      "grad_norm": 1.1482845416245802,
      "learning_rate": 8.97298225009632e-06,
      "loss": 0.42088894844055175,
      "num_input_tokens_seen": 101045952,
      "step": 6170,
      "train_runtime": 11869.3139,
      "train_tokens_per_second": 8513.209
    },
    {
      "epoch": 0.415768920010773,
      "grad_norm": 1.0167756848096334,
      "learning_rate": 8.971376372054829e-06,
      "loss": 0.43500452041625975,
      "num_input_tokens_seen": 101127872,
      "step": 6175,
      "train_runtime": 11878.9299,
      "train_tokens_per_second": 8513.214
    },
    {
      "epoch": 0.4161055750067331,
      "grad_norm": 0.9343661858155025,
      "learning_rate": 8.969769383430602e-06,
      "loss": 0.4126449108123779,
      "num_input_tokens_seen": 101209792,
      "step": 6180,
      "train_runtime": 11888.4718,
      "train_tokens_per_second": 8513.272
    },
    {
      "epoch": 0.41644223000269326,
      "grad_norm": 1.2178555871020502,
      "learning_rate": 8.968161284673027e-06,
      "loss": 0.3899010181427002,
      "num_input_tokens_seen": 101291712,
      "step": 6185,
      "train_runtime": 11898.0507,
      "train_tokens_per_second": 8513.303
    },
    {
      "epoch": 0.4167788849986534,
      "grad_norm": 1.058988334658142,
      "learning_rate": 8.9665520762318e-06,
      "loss": 0.40732927322387696,
      "num_input_tokens_seen": 101373632,
      "step": 6190,
      "train_runtime": 11907.6691,
      "train_tokens_per_second": 8513.306
    },
    {
      "epoch": 0.41711553999461354,
      "grad_norm": 1.1678551642308865,
      "learning_rate": 8.964941758556934e-06,
      "loss": 0.4470200538635254,
      "num_input_tokens_seen": 101455552,
      "step": 6195,
      "train_runtime": 11917.2059,
      "train_tokens_per_second": 8513.367
    },
    {
      "epoch": 0.4174521949905737,
      "grad_norm": 1.0311526802400985,
      "learning_rate": 8.963330332098747e-06,
      "loss": 0.392593240737915,
      "num_input_tokens_seen": 101537472,
      "step": 6200,
      "train_runtime": 11926.7817,
      "train_tokens_per_second": 8513.401
    },
    {
      "epoch": 0.4177888499865338,
      "grad_norm": 1.1933091662301893,
      "learning_rate": 8.961717797307872e-06,
      "loss": 0.39822723865509035,
      "num_input_tokens_seen": 101619392,
      "step": 6205,
      "train_runtime": 11936.419,
      "train_tokens_per_second": 8513.39
    },
    {
      "epoch": 0.41812550498249396,
      "grad_norm": 1.3377154655802546,
      "learning_rate": 8.960104154635248e-06,
      "loss": 0.4408430099487305,
      "num_input_tokens_seen": 101701312,
      "step": 6210,
      "train_runtime": 11945.9747,
      "train_tokens_per_second": 8513.438
    },
    {
      "epoch": 0.4184621599784541,
      "grad_norm": 1.2033788961041867,
      "learning_rate": 8.958489404532125e-06,
      "loss": 0.45555682182312013,
      "num_input_tokens_seen": 101783232,
      "step": 6215,
      "train_runtime": 11955.5451,
      "train_tokens_per_second": 8513.475
    },
    {
      "epoch": 0.41879881497441424,
      "grad_norm": 1.05286587210886,
      "learning_rate": 8.956873547450062e-06,
      "loss": 0.41132454872131347,
      "num_input_tokens_seen": 101865152,
      "step": 6220,
      "train_runtime": 11965.154,
      "train_tokens_per_second": 8513.484
    },
    {
      "epoch": 0.4191354699703744,
      "grad_norm": 1.4439204900814429,
      "learning_rate": 8.95525658384093e-06,
      "loss": 0.49929208755493165,
      "num_input_tokens_seen": 101947032,
      "step": 6225,
      "train_runtime": 11975.3422,
      "train_tokens_per_second": 8513.079
    },
    {
      "epoch": 0.4194721249663345,
      "grad_norm": 1.0140066848832283,
      "learning_rate": 8.953638514156908e-06,
      "loss": 0.4201416015625,
      "num_input_tokens_seen": 102028952,
      "step": 6230,
      "train_runtime": 11984.841,
      "train_tokens_per_second": 8513.167
    },
    {
      "epoch": 0.41980877996229465,
      "grad_norm": 1.277671052054765,
      "learning_rate": 8.952019338850481e-06,
      "loss": 0.3902099132537842,
      "num_input_tokens_seen": 102110560,
      "step": 6235,
      "train_runtime": 11994.4533,
      "train_tokens_per_second": 8513.148
    },
    {
      "epoch": 0.4201454349582548,
      "grad_norm": 1.1779129081527167,
      "learning_rate": 8.950399058374447e-06,
      "loss": 0.3821418046951294,
      "num_input_tokens_seen": 102192480,
      "step": 6240,
      "train_runtime": 12004.0775,
      "train_tokens_per_second": 8513.147
    },
    {
      "epoch": 0.42048208995421493,
      "grad_norm": 0.9411751009834844,
      "learning_rate": 8.948777673181913e-06,
      "loss": 0.3889930009841919,
      "num_input_tokens_seen": 102274400,
      "step": 6245,
      "train_runtime": 12014.1675,
      "train_tokens_per_second": 8512.816
    },
    {
      "epoch": 0.42081874495017507,
      "grad_norm": 0.9815933205679878,
      "learning_rate": 8.947155183726296e-06,
      "loss": 0.38979377746582033,
      "num_input_tokens_seen": 102356320,
      "step": 6250,
      "train_runtime": 12024.0246,
      "train_tokens_per_second": 8512.651
    },
    {
      "epoch": 0.4211553999461352,
      "grad_norm": 1.183266257520633,
      "learning_rate": 8.94553159046132e-06,
      "loss": 0.40174045562744143,
      "num_input_tokens_seen": 102438240,
      "step": 6255,
      "train_runtime": 12033.5875,
      "train_tokens_per_second": 8512.693
    },
    {
      "epoch": 0.42149205494209535,
      "grad_norm": 1.1152221127002444,
      "learning_rate": 8.943906893841015e-06,
      "loss": 0.3820040702819824,
      "num_input_tokens_seen": 102520160,
      "step": 6260,
      "train_runtime": 12043.1413,
      "train_tokens_per_second": 8512.742
    },
    {
      "epoch": 0.4218287099380555,
      "grad_norm": 1.4173804263994823,
      "learning_rate": 8.942281094319721e-06,
      "loss": 0.45283708572387693,
      "num_input_tokens_seen": 102602080,
      "step": 6265,
      "train_runtime": 12052.6991,
      "train_tokens_per_second": 8512.789
    },
    {
      "epoch": 0.42216536493401563,
      "grad_norm": 1.0704309296885275,
      "learning_rate": 8.940654192352092e-06,
      "loss": 0.3768389463424683,
      "num_input_tokens_seen": 102684000,
      "step": 6270,
      "train_runtime": 12062.3213,
      "train_tokens_per_second": 8512.789
    },
    {
      "epoch": 0.42250201992997577,
      "grad_norm": 1.106154640605666,
      "learning_rate": 8.939026188393083e-06,
      "loss": 0.40515713691711425,
      "num_input_tokens_seen": 102765920,
      "step": 6275,
      "train_runtime": 12071.8702,
      "train_tokens_per_second": 8512.842
    },
    {
      "epoch": 0.4228386749259359,
      "grad_norm": 1.0230474918490051,
      "learning_rate": 8.937397082897963e-06,
      "loss": 0.44892377853393556,
      "num_input_tokens_seen": 102847840,
      "step": 6280,
      "train_runtime": 12081.756,
      "train_tokens_per_second": 8512.657
    },
    {
      "epoch": 0.42317532992189605,
      "grad_norm": 0.9848634148079274,
      "learning_rate": 8.935766876322303e-06,
      "loss": 0.4425664901733398,
      "num_input_tokens_seen": 102929760,
      "step": 6285,
      "train_runtime": 12091.3563,
      "train_tokens_per_second": 8512.673
    },
    {
      "epoch": 0.4235119849178562,
      "grad_norm": 0.9364964441057642,
      "learning_rate": 8.934135569121985e-06,
      "loss": 0.4431659698486328,
      "num_input_tokens_seen": 103011680,
      "step": 6290,
      "train_runtime": 12100.9728,
      "train_tokens_per_second": 8512.678
    },
    {
      "epoch": 0.4238486399138163,
      "grad_norm": 1.018963689692265,
      "learning_rate": 8.932503161753202e-06,
      "loss": 0.34417071342468264,
      "num_input_tokens_seen": 103093600,
      "step": 6295,
      "train_runtime": 12110.4776,
      "train_tokens_per_second": 8512.761
    },
    {
      "epoch": 0.42418529490977647,
      "grad_norm": 1.2893328423012809,
      "learning_rate": 8.930869654672449e-06,
      "loss": 0.44674053192138674,
      "num_input_tokens_seen": 103175200,
      "step": 6300,
      "train_runtime": 12120.5209,
      "train_tokens_per_second": 8512.439
    },
    {
      "epoch": 0.4245219499057366,
      "grad_norm": 1.1946774464816678,
      "learning_rate": 8.929235048336531e-06,
      "loss": 0.4422935485839844,
      "num_input_tokens_seen": 103257120,
      "step": 6305,
      "train_runtime": 12130.1367,
      "train_tokens_per_second": 8512.445
    },
    {
      "epoch": 0.42485860490169675,
      "grad_norm": 1.284075561992368,
      "learning_rate": 8.927599343202563e-06,
      "loss": 0.4552154541015625,
      "num_input_tokens_seen": 103339040,
      "step": 6310,
      "train_runtime": 12139.6313,
      "train_tokens_per_second": 8512.535
    },
    {
      "epoch": 0.4251952598976569,
      "grad_norm": 1.1781096263137223,
      "learning_rate": 8.92596253972796e-06,
      "loss": 0.43727540969848633,
      "num_input_tokens_seen": 103420960,
      "step": 6315,
      "train_runtime": 12149.2369,
      "train_tokens_per_second": 8512.548
    },
    {
      "epoch": 0.425531914893617,
      "grad_norm": 1.0604556217150998,
      "learning_rate": 8.924324638370454e-06,
      "loss": 0.4306221008300781,
      "num_input_tokens_seen": 103502880,
      "step": 6320,
      "train_runtime": 12158.8487,
      "train_tokens_per_second": 8512.556
    },
    {
      "epoch": 0.42586856988957716,
      "grad_norm": 1.015957946648672,
      "learning_rate": 8.922685639588076e-06,
      "loss": 0.3686051368713379,
      "num_input_tokens_seen": 103584800,
      "step": 6325,
      "train_runtime": 12168.3576,
      "train_tokens_per_second": 8512.636
    },
    {
      "epoch": 0.4262052248855373,
      "grad_norm": 1.1391292545619183,
      "learning_rate": 8.921045543839167e-06,
      "loss": 0.48807034492492674,
      "num_input_tokens_seen": 103666720,
      "step": 6330,
      "train_runtime": 12177.9617,
      "train_tokens_per_second": 8512.65
    },
    {
      "epoch": 0.42654187988149744,
      "grad_norm": 1.015517532573582,
      "learning_rate": 8.919404351582376e-06,
      "loss": 0.3826406955718994,
      "num_input_tokens_seen": 103748640,
      "step": 6335,
      "train_runtime": 12187.5483,
      "train_tokens_per_second": 8512.675
    },
    {
      "epoch": 0.4268785348774576,
      "grad_norm": 1.2562193378860238,
      "learning_rate": 8.917762063276657e-06,
      "loss": 0.3867073059082031,
      "num_input_tokens_seen": 103830560,
      "step": 6340,
      "train_runtime": 12197.1077,
      "train_tokens_per_second": 8512.72
    },
    {
      "epoch": 0.4272151898734177,
      "grad_norm": 0.90172479170259,
      "learning_rate": 8.916118679381268e-06,
      "loss": 0.3643355846405029,
      "num_input_tokens_seen": 103912480,
      "step": 6345,
      "train_runtime": 12206.6835,
      "train_tokens_per_second": 8512.753
    },
    {
      "epoch": 0.42755184486937786,
      "grad_norm": 1.1372482753008462,
      "learning_rate": 8.914474200355777e-06,
      "loss": 0.41728830337524414,
      "num_input_tokens_seen": 103994400,
      "step": 6350,
      "train_runtime": 12216.3139,
      "train_tokens_per_second": 8512.748
    },
    {
      "epoch": 0.427888499865338,
      "grad_norm": 1.115777136442444,
      "learning_rate": 8.912828626660059e-06,
      "loss": 0.43468389511108396,
      "num_input_tokens_seen": 104076320,
      "step": 6355,
      "train_runtime": 12225.8646,
      "train_tokens_per_second": 8512.798
    },
    {
      "epoch": 0.42822515486129814,
      "grad_norm": 1.0576298621169957,
      "learning_rate": 8.911181958754294e-06,
      "loss": 0.4372241020202637,
      "num_input_tokens_seen": 104158240,
      "step": 6360,
      "train_runtime": 12235.4315,
      "train_tokens_per_second": 8512.837
    },
    {
      "epoch": 0.4285618098572583,
      "grad_norm": 0.9569728234711598,
      "learning_rate": 8.909534197098966e-06,
      "loss": 0.45699286460876465,
      "num_input_tokens_seen": 104240160,
      "step": 6365,
      "train_runtime": 12245.0484,
      "train_tokens_per_second": 8512.842
    },
    {
      "epoch": 0.4288984648532184,
      "grad_norm": 1.0252227451104685,
      "learning_rate": 8.907885342154865e-06,
      "loss": 0.3776343584060669,
      "num_input_tokens_seen": 104322080,
      "step": 6370,
      "train_runtime": 12254.6669,
      "train_tokens_per_second": 8512.845
    },
    {
      "epoch": 0.42923511984917856,
      "grad_norm": 1.0741282249361233,
      "learning_rate": 8.906235394383089e-06,
      "loss": 0.39996817111968996,
      "num_input_tokens_seen": 104404000,
      "step": 6375,
      "train_runtime": 12264.1749,
      "train_tokens_per_second": 8512.925
    },
    {
      "epoch": 0.4295717748451387,
      "grad_norm": 1.1737738637682449,
      "learning_rate": 8.904584354245042e-06,
      "loss": 0.4400495052337646,
      "num_input_tokens_seen": 104485920,
      "step": 6380,
      "train_runtime": 12273.8057,
      "train_tokens_per_second": 8512.92
    },
    {
      "epoch": 0.42990842984109884,
      "grad_norm": 1.2689734423266212,
      "learning_rate": 8.902932222202433e-06,
      "loss": 0.4198782444000244,
      "num_input_tokens_seen": 104567840,
      "step": 6385,
      "train_runtime": 12283.4047,
      "train_tokens_per_second": 8512.936
    },
    {
      "epoch": 0.430245084837059,
      "grad_norm": 0.948036403358958,
      "learning_rate": 8.901278998717272e-06,
      "loss": 0.40378990173339846,
      "num_input_tokens_seen": 104649760,
      "step": 6390,
      "train_runtime": 12292.9053,
      "train_tokens_per_second": 8513.021
    },
    {
      "epoch": 0.4305817398330191,
      "grad_norm": 1.1388295962584367,
      "learning_rate": 8.899624684251878e-06,
      "loss": 0.42076711654663085,
      "num_input_tokens_seen": 104731680,
      "step": 6395,
      "train_runtime": 12302.5155,
      "train_tokens_per_second": 8513.03
    },
    {
      "epoch": 0.43091839482897926,
      "grad_norm": 1.1100026915305694,
      "learning_rate": 8.897969279268877e-06,
      "loss": 0.35288267135620116,
      "num_input_tokens_seen": 104813600,
      "step": 6400,
      "train_runtime": 12312.1357,
      "train_tokens_per_second": 8513.032
    },
    {
      "epoch": 0.4312550498249394,
      "grad_norm": 1.0368073465752319,
      "learning_rate": 8.896312784231196e-06,
      "loss": 0.4114940643310547,
      "num_input_tokens_seen": 104894200,
      "step": 6405,
      "train_runtime": 12321.7481,
      "train_tokens_per_second": 8512.932
    },
    {
      "epoch": 0.43159170482089954,
      "grad_norm": 1.1679925901142276,
      "learning_rate": 8.89465519960207e-06,
      "loss": 0.4647543430328369,
      "num_input_tokens_seen": 104976120,
      "step": 6410,
      "train_runtime": 12331.3108,
      "train_tokens_per_second": 8512.973
    },
    {
      "epoch": 0.4319283598168597,
      "grad_norm": 1.1076211449698854,
      "learning_rate": 8.892996525845037e-06,
      "loss": 0.4357959747314453,
      "num_input_tokens_seen": 105058040,
      "step": 6415,
      "train_runtime": 12340.8955,
      "train_tokens_per_second": 8513.0
    },
    {
      "epoch": 0.4322650148128198,
      "grad_norm": 0.9688861329770112,
      "learning_rate": 8.89133676342394e-06,
      "loss": 0.42220449447631836,
      "num_input_tokens_seen": 105139440,
      "step": 6420,
      "train_runtime": 12350.7547,
      "train_tokens_per_second": 8512.795
    },
    {
      "epoch": 0.43260166980877995,
      "grad_norm": 1.0663936468055493,
      "learning_rate": 8.889675912802923e-06,
      "loss": 0.4207108497619629,
      "num_input_tokens_seen": 105221360,
      "step": 6425,
      "train_runtime": 12360.3257,
      "train_tokens_per_second": 8512.831
    },
    {
      "epoch": 0.4329383248047401,
      "grad_norm": 1.1466198365561837,
      "learning_rate": 8.888013974446443e-06,
      "loss": 0.4255038261413574,
      "num_input_tokens_seen": 105303280,
      "step": 6430,
      "train_runtime": 12369.9635,
      "train_tokens_per_second": 8512.821
    },
    {
      "epoch": 0.43327497980070023,
      "grad_norm": 1.0941550967901406,
      "learning_rate": 8.886350948819253e-06,
      "loss": 0.38056340217590334,
      "num_input_tokens_seen": 105385200,
      "step": 6435,
      "train_runtime": 12379.587,
      "train_tokens_per_second": 8512.82
    },
    {
      "epoch": 0.4336116347966604,
      "grad_norm": 1.0857112035541459,
      "learning_rate": 8.884686836386412e-06,
      "loss": 0.4001148700714111,
      "num_input_tokens_seen": 105467120,
      "step": 6440,
      "train_runtime": 12389.0892,
      "train_tokens_per_second": 8512.903
    },
    {
      "epoch": 0.4339482897926205,
      "grad_norm": 1.0868918451305665,
      "learning_rate": 8.883021637613286e-06,
      "loss": 0.4385634422302246,
      "num_input_tokens_seen": 105549040,
      "step": 6445,
      "train_runtime": 12398.7189,
      "train_tokens_per_second": 8512.899
    },
    {
      "epoch": 0.43428494478858065,
      "grad_norm": 1.2034498725500684,
      "learning_rate": 8.881355352965537e-06,
      "loss": 0.41811456680297854,
      "num_input_tokens_seen": 105630960,
      "step": 6450,
      "train_runtime": 12408.3313,
      "train_tokens_per_second": 8512.906
    },
    {
      "epoch": 0.4346215997845408,
      "grad_norm": 1.1041050978854425,
      "learning_rate": 8.879687982909145e-06,
      "loss": 0.3799950838088989,
      "num_input_tokens_seen": 105712496,
      "step": 6455,
      "train_runtime": 12417.8341,
      "train_tokens_per_second": 8512.958
    },
    {
      "epoch": 0.43495825478050093,
      "grad_norm": 1.3227575616808325,
      "learning_rate": 8.878019527910378e-06,
      "loss": 0.43982601165771484,
      "num_input_tokens_seen": 105794416,
      "step": 6460,
      "train_runtime": 12427.4334,
      "train_tokens_per_second": 8512.974
    },
    {
      "epoch": 0.43529490977646107,
      "grad_norm": 1.153641822359412,
      "learning_rate": 8.876349988435815e-06,
      "loss": 0.4909327507019043,
      "num_input_tokens_seen": 105876336,
      "step": 6465,
      "train_runtime": 12437.0705,
      "train_tokens_per_second": 8512.964
    },
    {
      "epoch": 0.4356315647724212,
      "grad_norm": 0.9342106047924936,
      "learning_rate": 8.874679364952339e-06,
      "loss": 0.4398624897003174,
      "num_input_tokens_seen": 105958256,
      "step": 6470,
      "train_runtime": 12446.5823,
      "train_tokens_per_second": 8513.04
    },
    {
      "epoch": 0.43596821976838135,
      "grad_norm": 1.0604631922210532,
      "learning_rate": 8.873007657927135e-06,
      "loss": 0.40314712524414065,
      "num_input_tokens_seen": 106040176,
      "step": 6475,
      "train_runtime": 12456.1957,
      "train_tokens_per_second": 8513.047
    },
    {
      "epoch": 0.4363048747643415,
      "grad_norm": 0.9579434521207757,
      "learning_rate": 8.87133486782769e-06,
      "loss": 0.3969240427017212,
      "num_input_tokens_seen": 106122096,
      "step": 6480,
      "train_runtime": 12465.8104,
      "train_tokens_per_second": 8513.052
    },
    {
      "epoch": 0.4366415297603016,
      "grad_norm": 1.0216216797298414,
      "learning_rate": 8.869660995121792e-06,
      "loss": 0.37186896800994873,
      "num_input_tokens_seen": 106204016,
      "step": 6485,
      "train_runtime": 12475.3208,
      "train_tokens_per_second": 8513.129
    },
    {
      "epoch": 0.43697818475626177,
      "grad_norm": 1.2968908404297592,
      "learning_rate": 8.867986040277537e-06,
      "loss": 0.4405055999755859,
      "num_input_tokens_seen": 106285936,
      "step": 6490,
      "train_runtime": 12484.9502,
      "train_tokens_per_second": 8513.125
    },
    {
      "epoch": 0.4373148397522219,
      "grad_norm": 1.1119480060392815,
      "learning_rate": 8.86631000376332e-06,
      "loss": 0.44515080451965333,
      "num_input_tokens_seen": 106367856,
      "step": 6495,
      "train_runtime": 12494.5661,
      "train_tokens_per_second": 8513.129
    },
    {
      "epoch": 0.43765149474818205,
      "grad_norm": 1.099802544655834,
      "learning_rate": 8.864632886047841e-06,
      "loss": 0.38926076889038086,
      "num_input_tokens_seen": 106449776,
      "step": 6500,
      "train_runtime": 12504.1207,
      "train_tokens_per_second": 8513.176
    },
    {
      "epoch": 0.4379881497441422,
      "grad_norm": 1.2347452243201993,
      "learning_rate": 8.862954687600097e-06,
      "loss": 0.4016736507415771,
      "num_input_tokens_seen": 106531696,
      "step": 6505,
      "train_runtime": 12513.6927,
      "train_tokens_per_second": 8513.21
    },
    {
      "epoch": 0.4383248047401023,
      "grad_norm": 1.2793975797692636,
      "learning_rate": 8.861275408889393e-06,
      "loss": 0.4181353569030762,
      "num_input_tokens_seen": 106613616,
      "step": 6510,
      "train_runtime": 12523.263,
      "train_tokens_per_second": 8513.246
    },
    {
      "epoch": 0.43866145973606246,
      "grad_norm": 1.3228354984472719,
      "learning_rate": 8.859595050385332e-06,
      "loss": 0.45674915313720704,
      "num_input_tokens_seen": 106695536,
      "step": 6515,
      "train_runtime": 12532.7752,
      "train_tokens_per_second": 8513.321
    },
    {
      "epoch": 0.4389981147320226,
      "grad_norm": 1.067425647222133,
      "learning_rate": 8.857913612557825e-06,
      "loss": 0.4411344528198242,
      "num_input_tokens_seen": 106777456,
      "step": 6520,
      "train_runtime": 12542.3296,
      "train_tokens_per_second": 8513.367
    },
    {
      "epoch": 0.43933476972798274,
      "grad_norm": 1.0795166501082616,
      "learning_rate": 8.856231095877077e-06,
      "loss": 0.38887667655944824,
      "num_input_tokens_seen": 106859032,
      "step": 6525,
      "train_runtime": 12552.3625,
      "train_tokens_per_second": 8513.061
    },
    {
      "epoch": 0.4396714247239429,
      "grad_norm": 1.0800226693417578,
      "learning_rate": 8.8545475008136e-06,
      "loss": 0.4403986930847168,
      "num_input_tokens_seen": 106940952,
      "step": 6530,
      "train_runtime": 12561.9862,
      "train_tokens_per_second": 8513.061
    },
    {
      "epoch": 0.440008079719903,
      "grad_norm": 1.2043665376289296,
      "learning_rate": 8.852862827838207e-06,
      "loss": 0.3944054841995239,
      "num_input_tokens_seen": 107022872,
      "step": 6535,
      "train_runtime": 12571.4928,
      "train_tokens_per_second": 8513.139
    },
    {
      "epoch": 0.44034473471586316,
      "grad_norm": 1.0824575753579748,
      "learning_rate": 8.851177077422008e-06,
      "loss": 0.4239188671112061,
      "num_input_tokens_seen": 107104792,
      "step": 6540,
      "train_runtime": 12581.1113,
      "train_tokens_per_second": 8513.142
    },
    {
      "epoch": 0.4406813897118233,
      "grad_norm": 1.1356067355360853,
      "learning_rate": 8.849490250036421e-06,
      "loss": 0.43095965385437013,
      "num_input_tokens_seen": 107186712,
      "step": 6545,
      "train_runtime": 12590.748,
      "train_tokens_per_second": 8513.133
    },
    {
      "epoch": 0.44101804470778344,
      "grad_norm": 1.2404117802362014,
      "learning_rate": 8.847802346153159e-06,
      "loss": 0.41069278717041013,
      "num_input_tokens_seen": 107268512,
      "step": 6550,
      "train_runtime": 12600.7915,
      "train_tokens_per_second": 8512.839
    },
    {
      "epoch": 0.4413546997037436,
      "grad_norm": 1.12347784553722,
      "learning_rate": 8.846113366244241e-06,
      "loss": 0.45072360038757325,
      "num_input_tokens_seen": 107350432,
      "step": 6555,
      "train_runtime": 12610.4241,
      "train_tokens_per_second": 8512.833
    },
    {
      "epoch": 0.4416913546997037,
      "grad_norm": 1.087467723189835,
      "learning_rate": 8.844423310781986e-06,
      "loss": 0.38585190773010253,
      "num_input_tokens_seen": 107432352,
      "step": 6560,
      "train_runtime": 12620.0433,
      "train_tokens_per_second": 8512.835
    },
    {
      "epoch": 0.44202800969566386,
      "grad_norm": 1.073449706892645,
      "learning_rate": 8.842732180239011e-06,
      "loss": 0.4198409080505371,
      "num_input_tokens_seen": 107514272,
      "step": 6565,
      "train_runtime": 12629.5946,
      "train_tokens_per_second": 8512.884
    },
    {
      "epoch": 0.442364664691624,
      "grad_norm": 1.0636052266520402,
      "learning_rate": 8.841039975088234e-06,
      "loss": 0.41736502647399903,
      "num_input_tokens_seen": 107596192,
      "step": 6570,
      "train_runtime": 12639.1624,
      "train_tokens_per_second": 8512.921
    },
    {
      "epoch": 0.44270131968758414,
      "grad_norm": 1.0802886154269113,
      "learning_rate": 8.839346695802878e-06,
      "loss": 0.48049197196960447,
      "num_input_tokens_seen": 107678112,
      "step": 6575,
      "train_runtime": 12648.756,
      "train_tokens_per_second": 8512.941
    },
    {
      "epoch": 0.4430379746835443,
      "grad_norm": 0.9974259022220174,
      "learning_rate": 8.837652342856459e-06,
      "loss": 0.4129035472869873,
      "num_input_tokens_seen": 107760032,
      "step": 6580,
      "train_runtime": 12658.3055,
      "train_tokens_per_second": 8512.99
    },
    {
      "epoch": 0.4433746296795044,
      "grad_norm": 1.239487313785864,
      "learning_rate": 8.835956916722803e-06,
      "loss": 0.40210647583007814,
      "num_input_tokens_seen": 107841952,
      "step": 6585,
      "train_runtime": 12667.8743,
      "train_tokens_per_second": 8513.027
    },
    {
      "epoch": 0.44371128467546456,
      "grad_norm": 1.1541890630076657,
      "learning_rate": 8.834260417876024e-06,
      "loss": 0.4360334873199463,
      "num_input_tokens_seen": 107923872,
      "step": 6590,
      "train_runtime": 12677.4886,
      "train_tokens_per_second": 8513.032
    },
    {
      "epoch": 0.4440479396714247,
      "grad_norm": 1.0502641816776244,
      "learning_rate": 8.832562846790549e-06,
      "loss": 0.4100961685180664,
      "num_input_tokens_seen": 108005792,
      "step": 6595,
      "train_runtime": 12687.0356,
      "train_tokens_per_second": 8513.083
    },
    {
      "epoch": 0.4443845946673849,
      "grad_norm": 1.2195590509550258,
      "learning_rate": 8.830864203941092e-06,
      "loss": 0.39727158546447755,
      "num_input_tokens_seen": 108087712,
      "step": 6600,
      "train_runtime": 12696.6127,
      "train_tokens_per_second": 8513.114
    },
    {
      "epoch": 0.44472124966334503,
      "grad_norm": 1.2143902243240756,
      "learning_rate": 8.829164489802677e-06,
      "loss": 0.42834720611572263,
      "num_input_tokens_seen": 108169632,
      "step": 6605,
      "train_runtime": 12706.2105,
      "train_tokens_per_second": 8513.131
    },
    {
      "epoch": 0.44505790465930517,
      "grad_norm": 1.0971314037021183,
      "learning_rate": 8.827463704850622e-06,
      "loss": 0.4065227508544922,
      "num_input_tokens_seen": 108251552,
      "step": 6610,
      "train_runtime": 12715.7536,
      "train_tokens_per_second": 8513.184
    },
    {
      "epoch": 0.4453945596552653,
      "grad_norm": 1.1766645770412028,
      "learning_rate": 8.825761849560547e-06,
      "loss": 0.41199846267700196,
      "num_input_tokens_seen": 108333472,
      "step": 6615,
      "train_runtime": 12725.2704,
      "train_tokens_per_second": 8513.255
    },
    {
      "epoch": 0.44573121465122545,
      "grad_norm": 1.3077824816335455,
      "learning_rate": 8.824058924408371e-06,
      "loss": 0.4112420558929443,
      "num_input_tokens_seen": 108415392,
      "step": 6620,
      "train_runtime": 12734.8896,
      "train_tokens_per_second": 8513.257
    },
    {
      "epoch": 0.4460678696471856,
      "grad_norm": 0.9027969524111868,
      "learning_rate": 8.822354929870311e-06,
      "loss": 0.39484357833862305,
      "num_input_tokens_seen": 108497312,
      "step": 6625,
      "train_runtime": 12744.5225,
      "train_tokens_per_second": 8513.25
    },
    {
      "epoch": 0.44640452464314573,
      "grad_norm": 1.1283421273356742,
      "learning_rate": 8.820649866422884e-06,
      "loss": 0.435210657119751,
      "num_input_tokens_seen": 108579232,
      "step": 6630,
      "train_runtime": 12754.0255,
      "train_tokens_per_second": 8513.33
    },
    {
      "epoch": 0.44674117963910587,
      "grad_norm": 1.1302777773238044,
      "learning_rate": 8.818943734542905e-06,
      "loss": 0.41800646781921386,
      "num_input_tokens_seen": 108661152,
      "step": 6635,
      "train_runtime": 12763.6335,
      "train_tokens_per_second": 8513.34
    },
    {
      "epoch": 0.447077834635066,
      "grad_norm": 1.3441019760883384,
      "learning_rate": 8.817236534707486e-06,
      "loss": 0.39182395935058595,
      "num_input_tokens_seen": 108743072,
      "step": 6640,
      "train_runtime": 12773.2655,
      "train_tokens_per_second": 8513.334
    },
    {
      "epoch": 0.44741448963102615,
      "grad_norm": 1.0871737632757417,
      "learning_rate": 8.815528267394045e-06,
      "loss": 0.3928833961486816,
      "num_input_tokens_seen": 108824992,
      "step": 6645,
      "train_runtime": 12782.7657,
      "train_tokens_per_second": 8513.415
    },
    {
      "epoch": 0.4477511446269863,
      "grad_norm": 1.2071206976183995,
      "learning_rate": 8.81381893308029e-06,
      "loss": 0.4305140972137451,
      "num_input_tokens_seen": 108906912,
      "step": 6650,
      "train_runtime": 12792.3792,
      "train_tokens_per_second": 8513.421
    },
    {
      "epoch": 0.4480877996229464,
      "grad_norm": 1.1603902932062036,
      "learning_rate": 8.81210853224423e-06,
      "loss": 0.43005905151367185,
      "num_input_tokens_seen": 108988648,
      "step": 6655,
      "train_runtime": 12802.4834,
      "train_tokens_per_second": 8513.086
    },
    {
      "epoch": 0.44842445461890656,
      "grad_norm": 1.1462951684539529,
      "learning_rate": 8.810397065364177e-06,
      "loss": 0.357511568069458,
      "num_input_tokens_seen": 109070016,
      "step": 6660,
      "train_runtime": 12812.3382,
      "train_tokens_per_second": 8512.889
    },
    {
      "epoch": 0.4487611096148667,
      "grad_norm": 1.2502799824783448,
      "learning_rate": 8.808684532918735e-06,
      "loss": 0.4111231803894043,
      "num_input_tokens_seen": 109151936,
      "step": 6665,
      "train_runtime": 12821.9046,
      "train_tokens_per_second": 8512.927
    },
    {
      "epoch": 0.44909776461082684,
      "grad_norm": 1.0179560131934968,
      "learning_rate": 8.806970935386807e-06,
      "loss": 0.3937565803527832,
      "num_input_tokens_seen": 109233320,
      "step": 6670,
      "train_runtime": 12831.8346,
      "train_tokens_per_second": 8512.681
    },
    {
      "epoch": 0.449434419606787,
      "grad_norm": 1.0463392400935438,
      "learning_rate": 8.805256273247597e-06,
      "loss": 0.4055955410003662,
      "num_input_tokens_seen": 109315240,
      "step": 6675,
      "train_runtime": 12841.4629,
      "train_tokens_per_second": 8512.678
    },
    {
      "epoch": 0.4497710746027471,
      "grad_norm": 1.0671316505070245,
      "learning_rate": 8.803540546980605e-06,
      "loss": 0.41266822814941406,
      "num_input_tokens_seen": 109397160,
      "step": 6680,
      "train_runtime": 12850.9662,
      "train_tokens_per_second": 8512.758
    },
    {
      "epoch": 0.45010772959870726,
      "grad_norm": 1.2000452360906253,
      "learning_rate": 8.801823757065628e-06,
      "loss": 0.4255462646484375,
      "num_input_tokens_seen": 109479080,
      "step": 6685,
      "train_runtime": 12860.591,
      "train_tokens_per_second": 8512.757
    },
    {
      "epoch": 0.4504443845946674,
      "grad_norm": 1.029166589524834,
      "learning_rate": 8.800105903982758e-06,
      "loss": 0.41648063659667967,
      "num_input_tokens_seen": 109561000,
      "step": 6690,
      "train_runtime": 12870.2118,
      "train_tokens_per_second": 8512.758
    },
    {
      "epoch": 0.45078103959062754,
      "grad_norm": 1.0309703829553438,
      "learning_rate": 8.798386988212393e-06,
      "loss": 0.377591609954834,
      "num_input_tokens_seen": 109642528,
      "step": 6695,
      "train_runtime": 12880.0719,
      "train_tokens_per_second": 8512.571
    },
    {
      "epoch": 0.4511176945865877,
      "grad_norm": 1.154757471399098,
      "learning_rate": 8.796667010235216e-06,
      "loss": 0.4513570308685303,
      "num_input_tokens_seen": 109724448,
      "step": 6700,
      "train_runtime": 12889.7088,
      "train_tokens_per_second": 8512.562
    },
    {
      "epoch": 0.4514543495825478,
      "grad_norm": 1.5577029802134954,
      "learning_rate": 8.794945970532219e-06,
      "loss": 0.39124717712402346,
      "num_input_tokens_seen": 109806368,
      "step": 6705,
      "train_runtime": 12899.3199,
      "train_tokens_per_second": 8512.57
    },
    {
      "epoch": 0.45179100457850796,
      "grad_norm": 1.3606365147538066,
      "learning_rate": 8.79322386958468e-06,
      "loss": 0.44945969581604006,
      "num_input_tokens_seen": 109888288,
      "step": 6710,
      "train_runtime": 12908.8717,
      "train_tokens_per_second": 8512.618
    },
    {
      "epoch": 0.4521276595744681,
      "grad_norm": 1.089804582780707,
      "learning_rate": 8.791500707874183e-06,
      "loss": 0.39991543292999265,
      "num_input_tokens_seen": 109970208,
      "step": 6715,
      "train_runtime": 12918.4447,
      "train_tokens_per_second": 8512.651
    },
    {
      "epoch": 0.45246431457042824,
      "grad_norm": 1.188930380808582,
      "learning_rate": 8.789776485882601e-06,
      "loss": 0.4309090614318848,
      "num_input_tokens_seen": 110052128,
      "step": 6720,
      "train_runtime": 12928.0718,
      "train_tokens_per_second": 8512.648
    },
    {
      "epoch": 0.4528009695663884,
      "grad_norm": 1.0412378717988284,
      "learning_rate": 8.788051204092112e-06,
      "loss": 0.4171022415161133,
      "num_input_tokens_seen": 110134048,
      "step": 6725,
      "train_runtime": 12937.5949,
      "train_tokens_per_second": 8512.714
    },
    {
      "epoch": 0.4531376245623485,
      "grad_norm": 1.275931021898937,
      "learning_rate": 8.786324862985183e-06,
      "loss": 0.38310680389404295,
      "num_input_tokens_seen": 110215968,
      "step": 6730,
      "train_runtime": 12947.1004,
      "train_tokens_per_second": 8512.792
    },
    {
      "epoch": 0.45347427955830866,
      "grad_norm": 0.9720908780210759,
      "learning_rate": 8.78459746304458e-06,
      "loss": 0.39076008796691897,
      "num_input_tokens_seen": 110297888,
      "step": 6735,
      "train_runtime": 12956.7017,
      "train_tokens_per_second": 8512.806
    },
    {
      "epoch": 0.4538109345542688,
      "grad_norm": 1.1416955526814996,
      "learning_rate": 8.782869004753363e-06,
      "loss": 0.40689983367919924,
      "num_input_tokens_seen": 110379808,
      "step": 6740,
      "train_runtime": 12966.3366,
      "train_tokens_per_second": 8512.798
    },
    {
      "epoch": 0.45414758955022894,
      "grad_norm": 1.2189249829828672,
      "learning_rate": 8.781139488594892e-06,
      "loss": 0.42772645950317384,
      "num_input_tokens_seen": 110461728,
      "step": 6745,
      "train_runtime": 12975.84,
      "train_tokens_per_second": 8512.877
    },
    {
      "epoch": 0.4544842445461891,
      "grad_norm": 1.0993813793464424,
      "learning_rate": 8.779408915052821e-06,
      "loss": 0.3731199026107788,
      "num_input_tokens_seen": 110543648,
      "step": 6750,
      "train_runtime": 12985.477,
      "train_tokens_per_second": 8512.868
    },
    {
      "epoch": 0.4548208995421492,
      "grad_norm": 1.1115187223081464,
      "learning_rate": 8.777677284611096e-06,
      "loss": 0.369341516494751,
      "num_input_tokens_seen": 110625568,
      "step": 6755,
      "train_runtime": 12995.0709,
      "train_tokens_per_second": 8512.887
    },
    {
      "epoch": 0.45515755453810935,
      "grad_norm": 1.2123164821419232,
      "learning_rate": 8.775944597753969e-06,
      "loss": 0.4342747688293457,
      "num_input_tokens_seen": 110707488,
      "step": 6760,
      "train_runtime": 13004.5761,
      "train_tokens_per_second": 8512.964
    },
    {
      "epoch": 0.4554942095340695,
      "grad_norm": 1.0588014774443808,
      "learning_rate": 8.774210854965972e-06,
      "loss": 0.4596001148223877,
      "num_input_tokens_seen": 110789408,
      "step": 6765,
      "train_runtime": 13014.1988,
      "train_tokens_per_second": 8512.964
    },
    {
      "epoch": 0.45583086453002963,
      "grad_norm": 1.1275612581100531,
      "learning_rate": 8.772476056731946e-06,
      "loss": 0.3591123342514038,
      "num_input_tokens_seen": 110871328,
      "step": 6770,
      "train_runtime": 13023.8097,
      "train_tokens_per_second": 8512.972
    },
    {
      "epoch": 0.4561675195259898,
      "grad_norm": 0.9878785506130283,
      "learning_rate": 8.77074020353702e-06,
      "loss": 0.40903940200805666,
      "num_input_tokens_seen": 110953248,
      "step": 6775,
      "train_runtime": 13033.3172,
      "train_tokens_per_second": 8513.047
    },
    {
      "epoch": 0.4565041745219499,
      "grad_norm": 1.0845380738215502,
      "learning_rate": 8.76900329586662e-06,
      "loss": 0.4432262420654297,
      "num_input_tokens_seen": 111035168,
      "step": 6780,
      "train_runtime": 13042.9472,
      "train_tokens_per_second": 8513.043
    },
    {
      "epoch": 0.45684082951791005,
      "grad_norm": 1.2308943392194818,
      "learning_rate": 8.767265334206467e-06,
      "loss": 0.43184328079223633,
      "num_input_tokens_seen": 111117088,
      "step": 6785,
      "train_runtime": 13052.575,
      "train_tokens_per_second": 8513.04
    },
    {
      "epoch": 0.4571774845138702,
      "grad_norm": 3.086901847177219,
      "learning_rate": 8.765526319042577e-06,
      "loss": 0.3704275608062744,
      "num_input_tokens_seen": 111199008,
      "step": 6790,
      "train_runtime": 13062.1234,
      "train_tokens_per_second": 8513.088
    },
    {
      "epoch": 0.45751413950983033,
      "grad_norm": 1.2841455238113049,
      "learning_rate": 8.763786250861258e-06,
      "loss": 0.40385775566101073,
      "num_input_tokens_seen": 111280928,
      "step": 6795,
      "train_runtime": 13071.6864,
      "train_tokens_per_second": 8513.127
    },
    {
      "epoch": 0.45785079450579047,
      "grad_norm": 1.146576851288958,
      "learning_rate": 8.762045130149114e-06,
      "loss": 0.3794842720031738,
      "num_input_tokens_seen": 111362848,
      "step": 6800,
      "train_runtime": 13081.301,
      "train_tokens_per_second": 8513.132
    },
    {
      "epoch": 0.4581874495017506,
      "grad_norm": 1.0491963452561675,
      "learning_rate": 8.76030295739305e-06,
      "loss": 0.4401745319366455,
      "num_input_tokens_seen": 111444768,
      "step": 6805,
      "train_runtime": 13090.8529,
      "train_tokens_per_second": 8513.179
    },
    {
      "epoch": 0.45852410449771075,
      "grad_norm": 1.1024260846830471,
      "learning_rate": 8.758559733080252e-06,
      "loss": 0.44900965690612793,
      "num_input_tokens_seen": 111526688,
      "step": 6810,
      "train_runtime": 13100.4312,
      "train_tokens_per_second": 8513.207
    },
    {
      "epoch": 0.4588607594936709,
      "grad_norm": 1.1898015057960336,
      "learning_rate": 8.75681545769821e-06,
      "loss": 0.4118891716003418,
      "num_input_tokens_seen": 111608608,
      "step": 6815,
      "train_runtime": 13110.0305,
      "train_tokens_per_second": 8513.223
    },
    {
      "epoch": 0.45919741448963103,
      "grad_norm": 1.167721269744865,
      "learning_rate": 8.755070131734706e-06,
      "loss": 0.43487701416015623,
      "num_input_tokens_seen": 111690528,
      "step": 6820,
      "train_runtime": 13119.6355,
      "train_tokens_per_second": 8513.234
    },
    {
      "epoch": 0.45953406948559117,
      "grad_norm": 1.2546692501129095,
      "learning_rate": 8.753323755677812e-06,
      "loss": 0.4371074676513672,
      "num_input_tokens_seen": 111771936,
      "step": 6825,
      "train_runtime": 13129.4006,
      "train_tokens_per_second": 8513.103
    },
    {
      "epoch": 0.4598707244815513,
      "grad_norm": 0.9468318071577687,
      "learning_rate": 8.7515763300159e-06,
      "loss": 0.34091601371765134,
      "num_input_tokens_seen": 111853856,
      "step": 6830,
      "train_runtime": 13139.0282,
      "train_tokens_per_second": 8513.1
    },
    {
      "epoch": 0.46020737947751145,
      "grad_norm": 0.976717469237834,
      "learning_rate": 8.74982785523763e-06,
      "loss": 0.41602582931518556,
      "num_input_tokens_seen": 111935776,
      "step": 6835,
      "train_runtime": 13148.6596,
      "train_tokens_per_second": 8513.094
    },
    {
      "epoch": 0.4605440344734716,
      "grad_norm": 1.0413543887023766,
      "learning_rate": 8.748078331831957e-06,
      "loss": 0.4165037155151367,
      "num_input_tokens_seen": 112017696,
      "step": 6840,
      "train_runtime": 13158.1647,
      "train_tokens_per_second": 8513.17
    },
    {
      "epoch": 0.4608806894694317,
      "grad_norm": 1.230014768460489,
      "learning_rate": 8.746327760288129e-06,
      "loss": 0.44164242744445803,
      "num_input_tokens_seen": 112099616,
      "step": 6845,
      "train_runtime": 13167.7755,
      "train_tokens_per_second": 8513.178
    },
    {
      "epoch": 0.46121734446539187,
      "grad_norm": 1.1572679955205578,
      "learning_rate": 8.744576141095691e-06,
      "loss": 0.40340681076049806,
      "num_input_tokens_seen": 112181536,
      "step": 6850,
      "train_runtime": 13177.3844,
      "train_tokens_per_second": 8513.187
    },
    {
      "epoch": 0.461553999461352,
      "grad_norm": 1.0698442676282984,
      "learning_rate": 8.742823474744476e-06,
      "loss": 0.40969104766845704,
      "num_input_tokens_seen": 112263456,
      "step": 6855,
      "train_runtime": 13186.9338,
      "train_tokens_per_second": 8513.234
    },
    {
      "epoch": 0.46189065445731214,
      "grad_norm": 1.1060506903105256,
      "learning_rate": 8.74106976172461e-06,
      "loss": 0.4375500202178955,
      "num_input_tokens_seen": 112345376,
      "step": 6860,
      "train_runtime": 13196.5136,
      "train_tokens_per_second": 8513.262
    },
    {
      "epoch": 0.4622273094532723,
      "grad_norm": 0.9813802466464427,
      "learning_rate": 8.739315002526513e-06,
      "loss": 0.3655224800109863,
      "num_input_tokens_seen": 112427296,
      "step": 6865,
      "train_runtime": 13206.1377,
      "train_tokens_per_second": 8513.261
    },
    {
      "epoch": 0.4625639644492324,
      "grad_norm": 0.976479250849382,
      "learning_rate": 8.737559197640902e-06,
      "loss": 0.3623396873474121,
      "num_input_tokens_seen": 112509216,
      "step": 6870,
      "train_runtime": 13215.686,
      "train_tokens_per_second": 8513.309
    },
    {
      "epoch": 0.46290061944519256,
      "grad_norm": 1.1030753753161,
      "learning_rate": 8.735802347558778e-06,
      "loss": 0.4678095817565918,
      "num_input_tokens_seen": 112591136,
      "step": 6875,
      "train_runtime": 13225.2621,
      "train_tokens_per_second": 8513.339
    },
    {
      "epoch": 0.4632372744411527,
      "grad_norm": 1.15506054156088,
      "learning_rate": 8.734044452771442e-06,
      "loss": 0.41818628311157224,
      "num_input_tokens_seen": 112671312,
      "step": 6880,
      "train_runtime": 13234.9282,
      "train_tokens_per_second": 8513.179
    },
    {
      "epoch": 0.46357392943711284,
      "grad_norm": 1.192728058431088,
      "learning_rate": 8.73228551377048e-06,
      "loss": 0.4381543159484863,
      "num_input_tokens_seen": 112753232,
      "step": 6885,
      "train_runtime": 13244.4759,
      "train_tokens_per_second": 8513.227
    },
    {
      "epoch": 0.463910584433073,
      "grad_norm": 1.1031429064371594,
      "learning_rate": 8.730525531047776e-06,
      "loss": 0.4481761932373047,
      "num_input_tokens_seen": 112835152,
      "step": 6890,
      "train_runtime": 13254.06,
      "train_tokens_per_second": 8513.252
    },
    {
      "epoch": 0.4642472394290331,
      "grad_norm": 1.0177145859159848,
      "learning_rate": 8.728764505095503e-06,
      "loss": 0.4194039344787598,
      "num_input_tokens_seen": 112917072,
      "step": 6895,
      "train_runtime": 13263.6701,
      "train_tokens_per_second": 8513.26
    },
    {
      "epoch": 0.46458389442499326,
      "grad_norm": 0.9623025564753079,
      "learning_rate": 8.727002436406126e-06,
      "loss": 0.41778998374938964,
      "num_input_tokens_seen": 112998992,
      "step": 6900,
      "train_runtime": 13273.2961,
      "train_tokens_per_second": 8513.258
    },
    {
      "epoch": 0.4649205494209534,
      "grad_norm": 1.0995958203951612,
      "learning_rate": 8.725239325472404e-06,
      "loss": 0.4450397491455078,
      "num_input_tokens_seen": 113080912,
      "step": 6905,
      "train_runtime": 13282.7939,
      "train_tokens_per_second": 8513.338
    },
    {
      "epoch": 0.46525720441691354,
      "grad_norm": 1.2529331511291464,
      "learning_rate": 8.723475172787381e-06,
      "loss": 0.4521728515625,
      "num_input_tokens_seen": 113162832,
      "step": 6910,
      "train_runtime": 13292.3861,
      "train_tokens_per_second": 8513.357
    },
    {
      "epoch": 0.4655938594128737,
      "grad_norm": 1.234574118687801,
      "learning_rate": 8.7217099788444e-06,
      "loss": 0.4547591686248779,
      "num_input_tokens_seen": 113244752,
      "step": 6915,
      "train_runtime": 13301.9821,
      "train_tokens_per_second": 8513.374
    },
    {
      "epoch": 0.4659305144088338,
      "grad_norm": 1.1093449529661148,
      "learning_rate": 8.719943744137091e-06,
      "loss": 0.40097846984863283,
      "num_input_tokens_seen": 113326672,
      "step": 6920,
      "train_runtime": 13311.4875,
      "train_tokens_per_second": 8513.449
    },
    {
      "epoch": 0.46626716940479396,
      "grad_norm": 1.2083172226080923,
      "learning_rate": 8.718176469159378e-06,
      "loss": 0.4497039794921875,
      "num_input_tokens_seen": 113408592,
      "step": 6925,
      "train_runtime": 13321.1046,
      "train_tokens_per_second": 8513.453
    },
    {
      "epoch": 0.4666038244007541,
      "grad_norm": 1.156613976736191,
      "learning_rate": 8.716408154405469e-06,
      "loss": 0.39054417610168457,
      "num_input_tokens_seen": 113490512,
      "step": 6930,
      "train_runtime": 13330.7129,
      "train_tokens_per_second": 8513.462
    },
    {
      "epoch": 0.46694047939671424,
      "grad_norm": 1.15183486774882,
      "learning_rate": 8.714638800369872e-06,
      "loss": 0.4910159111022949,
      "num_input_tokens_seen": 113571528,
      "step": 6935,
      "train_runtime": 13340.4374,
      "train_tokens_per_second": 8513.329
    },
    {
      "epoch": 0.4672771343926744,
      "grad_norm": 1.021522750693247,
      "learning_rate": 8.712868407547378e-06,
      "loss": 0.4250202178955078,
      "num_input_tokens_seen": 113653448,
      "step": 6940,
      "train_runtime": 13350.0155,
      "train_tokens_per_second": 8513.357
    },
    {
      "epoch": 0.4676137893886345,
      "grad_norm": 1.1942353284503653,
      "learning_rate": 8.711096976433076e-06,
      "loss": 0.4348171234130859,
      "num_input_tokens_seen": 113735368,
      "step": 6945,
      "train_runtime": 13359.6493,
      "train_tokens_per_second": 8513.35
    },
    {
      "epoch": 0.46795044438459465,
      "grad_norm": 0.9561066545127234,
      "learning_rate": 8.709324507522337e-06,
      "loss": 0.3825662851333618,
      "num_input_tokens_seen": 113817288,
      "step": 6950,
      "train_runtime": 13369.2055,
      "train_tokens_per_second": 8513.392
    },
    {
      "epoch": 0.4682870993805548,
      "grad_norm": 1.2235564730885014,
      "learning_rate": 8.707551001310828e-06,
      "loss": 0.4261052131652832,
      "num_input_tokens_seen": 113899208,
      "step": 6955,
      "train_runtime": 13378.786,
      "train_tokens_per_second": 8513.419
    },
    {
      "epoch": 0.46862375437651493,
      "grad_norm": 1.022800897225525,
      "learning_rate": 8.705776458294503e-06,
      "loss": 0.3785217046737671,
      "num_input_tokens_seen": 113980744,
      "step": 6960,
      "train_runtime": 13388.4043,
      "train_tokens_per_second": 8513.393
    },
    {
      "epoch": 0.4689604093724751,
      "grad_norm": 1.1366717778551987,
      "learning_rate": 8.704000878969608e-06,
      "loss": 0.38918318748474123,
      "num_input_tokens_seen": 114062664,
      "step": 6965,
      "train_runtime": 13397.9927,
      "train_tokens_per_second": 8513.414
    },
    {
      "epoch": 0.4692970643684352,
      "grad_norm": 1.1260854493354244,
      "learning_rate": 8.702224263832679e-06,
      "loss": 0.4104766368865967,
      "num_input_tokens_seen": 114144584,
      "step": 6970,
      "train_runtime": 13407.5002,
      "train_tokens_per_second": 8513.487
    },
    {
      "epoch": 0.46963371936439535,
      "grad_norm": 0.8915383806060335,
      "learning_rate": 8.700446613380542e-06,
      "loss": 0.4249127388000488,
      "num_input_tokens_seen": 114226504,
      "step": 6975,
      "train_runtime": 13417.1229,
      "train_tokens_per_second": 8513.487
    },
    {
      "epoch": 0.4699703743603555,
      "grad_norm": 1.1679485890848296,
      "learning_rate": 8.698667928110307e-06,
      "loss": 0.3701430320739746,
      "num_input_tokens_seen": 114308424,
      "step": 6980,
      "train_runtime": 13426.7634,
      "train_tokens_per_second": 8513.476
    },
    {
      "epoch": 0.47030702935631563,
      "grad_norm": 1.475635384131704,
      "learning_rate": 8.696888208519381e-06,
      "loss": 0.3906123161315918,
      "num_input_tokens_seen": 114390344,
      "step": 6985,
      "train_runtime": 13436.2791,
      "train_tokens_per_second": 8513.543
    },
    {
      "epoch": 0.47064368435227577,
      "grad_norm": 1.0603804923126596,
      "learning_rate": 8.695107455105454e-06,
      "loss": 0.38786399364471436,
      "num_input_tokens_seen": 114472264,
      "step": 6990,
      "train_runtime": 13445.907,
      "train_tokens_per_second": 8513.54
    },
    {
      "epoch": 0.4709803393482359,
      "grad_norm": 1.0377410074749145,
      "learning_rate": 8.693325668366513e-06,
      "loss": 0.3746509552001953,
      "num_input_tokens_seen": 114554184,
      "step": 6995,
      "train_runtime": 13455.518,
      "train_tokens_per_second": 8513.547
    },
    {
      "epoch": 0.47131699434419605,
      "grad_norm": 1.100353705284781,
      "learning_rate": 8.691542848800825e-06,
      "loss": 0.473067569732666,
      "num_input_tokens_seen": 114636104,
      "step": 7000,
      "train_runtime": 13465.0233,
      "train_tokens_per_second": 8513.621
    },
    {
      "epoch": 0.4716536493401562,
      "grad_norm": 1.2697592864256517,
      "learning_rate": 8.68975899690695e-06,
      "loss": 0.4268925666809082,
      "num_input_tokens_seen": 114718024,
      "step": 7005,
      "train_runtime": 13474.5969,
      "train_tokens_per_second": 8513.652
    },
    {
      "epoch": 0.47199030433611633,
      "grad_norm": 1.2487843750290395,
      "learning_rate": 8.687974113183738e-06,
      "loss": 0.4039153575897217,
      "num_input_tokens_seen": 114799944,
      "step": 7010,
      "train_runtime": 13484.1944,
      "train_tokens_per_second": 8513.667
    },
    {
      "epoch": 0.47232695933207647,
      "grad_norm": 1.0501592000539586,
      "learning_rate": 8.686188198130326e-06,
      "loss": 0.421143627166748,
      "num_input_tokens_seen": 114881864,
      "step": 7015,
      "train_runtime": 13493.7428,
      "train_tokens_per_second": 8513.714
    },
    {
      "epoch": 0.4726636143280366,
      "grad_norm": 1.004809769287885,
      "learning_rate": 8.684401252246138e-06,
      "loss": 0.4189322471618652,
      "num_input_tokens_seen": 114963784,
      "step": 7020,
      "train_runtime": 13503.3095,
      "train_tokens_per_second": 8513.749
    },
    {
      "epoch": 0.47300026932399675,
      "grad_norm": 1.1074757860136264,
      "learning_rate": 8.68261327603089e-06,
      "loss": 0.4151728630065918,
      "num_input_tokens_seen": 115045704,
      "step": 7025,
      "train_runtime": 13512.9175,
      "train_tokens_per_second": 8513.758
    },
    {
      "epoch": 0.4733369243199569,
      "grad_norm": 1.1004470552008232,
      "learning_rate": 8.68082426998458e-06,
      "loss": 0.4558708190917969,
      "num_input_tokens_seen": 115127624,
      "step": 7030,
      "train_runtime": 13522.5282,
      "train_tokens_per_second": 8513.765
    },
    {
      "epoch": 0.473673579315917,
      "grad_norm": 1.2836047605202472,
      "learning_rate": 8.6790342346075e-06,
      "loss": 0.38639469146728517,
      "num_input_tokens_seen": 115209544,
      "step": 7035,
      "train_runtime": 13532.0476,
      "train_tokens_per_second": 8513.829
    },
    {
      "epoch": 0.47401023431187717,
      "grad_norm": 0.9447779826772432,
      "learning_rate": 8.677243170400228e-06,
      "loss": 0.38329124450683594,
      "num_input_tokens_seen": 115291464,
      "step": 7040,
      "train_runtime": 13541.6692,
      "train_tokens_per_second": 8513.83
    },
    {
      "epoch": 0.4743468893078373,
      "grad_norm": 1.1176308383968758,
      "learning_rate": 8.675451077863632e-06,
      "loss": 0.3695223808288574,
      "num_input_tokens_seen": 115373384,
      "step": 7045,
      "train_runtime": 13551.28,
      "train_tokens_per_second": 8513.837
    },
    {
      "epoch": 0.47468354430379744,
      "grad_norm": 1.1488076732587256,
      "learning_rate": 8.67365795749886e-06,
      "loss": 0.398761510848999,
      "num_input_tokens_seen": 115455304,
      "step": 7050,
      "train_runtime": 13560.7829,
      "train_tokens_per_second": 8513.911
    },
    {
      "epoch": 0.4750201992997576,
      "grad_norm": 0.9475125472864006,
      "learning_rate": 8.671863809807355e-06,
      "loss": 0.375699782371521,
      "num_input_tokens_seen": 115537224,
      "step": 7055,
      "train_runtime": 13570.3869,
      "train_tokens_per_second": 8513.923
    },
    {
      "epoch": 0.4753568542957177,
      "grad_norm": 1.0521193615660054,
      "learning_rate": 8.670068635290844e-06,
      "loss": 0.38297467231750487,
      "num_input_tokens_seen": 115619144,
      "step": 7060,
      "train_runtime": 13579.9226,
      "train_tokens_per_second": 8513.977
    },
    {
      "epoch": 0.47569350929167786,
      "grad_norm": 1.0072337928725723,
      "learning_rate": 8.668272434451343e-06,
      "loss": 0.4159541130065918,
      "num_input_tokens_seen": 115701064,
      "step": 7065,
      "train_runtime": 13589.4282,
      "train_tokens_per_second": 8514.049
    },
    {
      "epoch": 0.476030164287638,
      "grad_norm": 0.8504658470778113,
      "learning_rate": 8.666475207791154e-06,
      "loss": 0.40416779518127444,
      "num_input_tokens_seen": 115782984,
      "step": 7070,
      "train_runtime": 13599.0425,
      "train_tokens_per_second": 8514.054
    },
    {
      "epoch": 0.47636681928359814,
      "grad_norm": 0.9419151003985473,
      "learning_rate": 8.664676955812863e-06,
      "loss": 0.42242016792297366,
      "num_input_tokens_seen": 115864904,
      "step": 7075,
      "train_runtime": 13608.6573,
      "train_tokens_per_second": 8514.058
    },
    {
      "epoch": 0.4767034742795583,
      "grad_norm": 1.0400463467312289,
      "learning_rate": 8.662877679019348e-06,
      "loss": 0.39750189781188966,
      "num_input_tokens_seen": 115946824,
      "step": 7080,
      "train_runtime": 13618.2242,
      "train_tokens_per_second": 8514.093
    },
    {
      "epoch": 0.4770401292755185,
      "grad_norm": 1.3185932504879672,
      "learning_rate": 8.66107737791377e-06,
      "loss": 0.43618173599243165,
      "num_input_tokens_seen": 116028744,
      "step": 7085,
      "train_runtime": 13627.8107,
      "train_tokens_per_second": 8514.115
    },
    {
      "epoch": 0.4773767842714786,
      "grad_norm": 1.1173501830330417,
      "learning_rate": 8.659276052999574e-06,
      "loss": 0.3959606170654297,
      "num_input_tokens_seen": 116110664,
      "step": 7090,
      "train_runtime": 13637.4373,
      "train_tokens_per_second": 8514.112
    },
    {
      "epoch": 0.47771343926743876,
      "grad_norm": 1.224876714802566,
      "learning_rate": 8.657473704780504e-06,
      "loss": 0.44341154098510743,
      "num_input_tokens_seen": 116192584,
      "step": 7095,
      "train_runtime": 13646.9935,
      "train_tokens_per_second": 8514.152
    },
    {
      "epoch": 0.4780500942633989,
      "grad_norm": 1.1476791975143164,
      "learning_rate": 8.65567033376057e-06,
      "loss": 0.41486611366271975,
      "num_input_tokens_seen": 116274504,
      "step": 7100,
      "train_runtime": 13656.5547,
      "train_tokens_per_second": 8514.19
    },
    {
      "epoch": 0.47838674925935903,
      "grad_norm": 1.1542933675992026,
      "learning_rate": 8.65386594044409e-06,
      "loss": 0.40370712280273435,
      "num_input_tokens_seen": 116356424,
      "step": 7105,
      "train_runtime": 13666.1761,
      "train_tokens_per_second": 8514.19
    },
    {
      "epoch": 0.4787234042553192,
      "grad_norm": 1.1204772123800404,
      "learning_rate": 8.652060525335647e-06,
      "loss": 0.37676894664764404,
      "num_input_tokens_seen": 116438344,
      "step": 7110,
      "train_runtime": 13675.7281,
      "train_tokens_per_second": 8514.234
    },
    {
      "epoch": 0.4790600592512793,
      "grad_norm": 1.1861122530787485,
      "learning_rate": 8.650254088940125e-06,
      "loss": 0.41529150009155275,
      "num_input_tokens_seen": 116520088,
      "step": 7115,
      "train_runtime": 13685.2499,
      "train_tokens_per_second": 8514.283
    },
    {
      "epoch": 0.47939671424723945,
      "grad_norm": 1.1134866534384582,
      "learning_rate": 8.648446631762686e-06,
      "loss": 0.41733036041259763,
      "num_input_tokens_seen": 116602008,
      "step": 7120,
      "train_runtime": 13694.8605,
      "train_tokens_per_second": 8514.29
    },
    {
      "epoch": 0.4797333692431996,
      "grad_norm": 1.0674488678115832,
      "learning_rate": 8.646638154308781e-06,
      "loss": 0.4136913299560547,
      "num_input_tokens_seen": 116683928,
      "step": 7125,
      "train_runtime": 13704.42,
      "train_tokens_per_second": 8514.328
    },
    {
      "epoch": 0.48007002423915973,
      "grad_norm": 1.0500744764873513,
      "learning_rate": 8.644828657084144e-06,
      "loss": 0.3613384246826172,
      "num_input_tokens_seen": 116765848,
      "step": 7130,
      "train_runtime": 13714.0048,
      "train_tokens_per_second": 8514.351
    },
    {
      "epoch": 0.48040667923511987,
      "grad_norm": 1.070533179837511,
      "learning_rate": 8.643018140594797e-06,
      "loss": 0.3815142631530762,
      "num_input_tokens_seen": 116847768,
      "step": 7135,
      "train_runtime": 13723.6167,
      "train_tokens_per_second": 8514.357
    },
    {
      "epoch": 0.48074333423108,
      "grad_norm": 1.0354870806828491,
      "learning_rate": 8.641206605347042e-06,
      "loss": 0.4023438930511475,
      "num_input_tokens_seen": 116929688,
      "step": 7140,
      "train_runtime": 13733.2261,
      "train_tokens_per_second": 8514.364
    },
    {
      "epoch": 0.48107998922704015,
      "grad_norm": 1.2193967161581019,
      "learning_rate": 8.639394051847472e-06,
      "loss": 0.3931852340698242,
      "num_input_tokens_seen": 117011608,
      "step": 7145,
      "train_runtime": 13742.7379,
      "train_tokens_per_second": 8514.432
    },
    {
      "epoch": 0.4814166442230003,
      "grad_norm": 0.9765102744533563,
      "learning_rate": 8.63758048060296e-06,
      "loss": 0.42343864440917967,
      "num_input_tokens_seen": 117093528,
      "step": 7150,
      "train_runtime": 13752.355,
      "train_tokens_per_second": 8514.435
    },
    {
      "epoch": 0.48175329921896043,
      "grad_norm": 1.2778983493926752,
      "learning_rate": 8.635765892120666e-06,
      "loss": 0.42446084022521974,
      "num_input_tokens_seen": 117175448,
      "step": 7155,
      "train_runtime": 13761.9865,
      "train_tokens_per_second": 8514.428
    },
    {
      "epoch": 0.48208995421492057,
      "grad_norm": 1.2082889023975498,
      "learning_rate": 8.633950286908032e-06,
      "loss": 0.40214123725891116,
      "num_input_tokens_seen": 117256304,
      "step": 7160,
      "train_runtime": 13771.6464,
      "train_tokens_per_second": 8514.327
    },
    {
      "epoch": 0.4824266092108807,
      "grad_norm": 1.1572872175216393,
      "learning_rate": 8.63213366547279e-06,
      "loss": 0.4714957237243652,
      "num_input_tokens_seen": 117338224,
      "step": 7165,
      "train_runtime": 13781.284,
      "train_tokens_per_second": 8514.317
    },
    {
      "epoch": 0.48276326420684085,
      "grad_norm": 1.0850834980750348,
      "learning_rate": 8.630316028322952e-06,
      "loss": 0.3721585750579834,
      "num_input_tokens_seen": 117420144,
      "step": 7170,
      "train_runtime": 13790.8965,
      "train_tokens_per_second": 8514.323
    },
    {
      "epoch": 0.483099919202801,
      "grad_norm": 1.0729990021383262,
      "learning_rate": 8.628497375966811e-06,
      "loss": 0.4154824256896973,
      "num_input_tokens_seen": 117502064,
      "step": 7175,
      "train_runtime": 13800.4581,
      "train_tokens_per_second": 8514.36
    },
    {
      "epoch": 0.4834365741987611,
      "grad_norm": 1.2636772644263072,
      "learning_rate": 8.62667770891295e-06,
      "loss": 0.44143877029418943,
      "num_input_tokens_seen": 117583984,
      "step": 7180,
      "train_runtime": 13810.0289,
      "train_tokens_per_second": 8514.391
    },
    {
      "epoch": 0.48377322919472127,
      "grad_norm": 0.8340128507279513,
      "learning_rate": 8.624857027670232e-06,
      "loss": 0.3803141117095947,
      "num_input_tokens_seen": 117665904,
      "step": 7185,
      "train_runtime": 13819.6485,
      "train_tokens_per_second": 8514.392
    },
    {
      "epoch": 0.4841098841906814,
      "grad_norm": 1.0487234173762061,
      "learning_rate": 8.623035332747804e-06,
      "loss": 0.4380201816558838,
      "num_input_tokens_seen": 117747824,
      "step": 7190,
      "train_runtime": 13829.2096,
      "train_tokens_per_second": 8514.429
    },
    {
      "epoch": 0.48444653918664154,
      "grad_norm": 0.950867704906651,
      "learning_rate": 8.6212126246551e-06,
      "loss": 0.3612549304962158,
      "num_input_tokens_seen": 117829744,
      "step": 7195,
      "train_runtime": 13838.7885,
      "train_tokens_per_second": 8514.455
    },
    {
      "epoch": 0.4847831941826017,
      "grad_norm": 1.286102291240375,
      "learning_rate": 8.619388903901833e-06,
      "loss": 0.44208660125732424,
      "num_input_tokens_seen": 117911536,
      "step": 7200,
      "train_runtime": 13848.8939,
      "train_tokens_per_second": 8514.148
    },
    {
      "epoch": 0.4851198491785618,
      "grad_norm": 1.0714830923825631,
      "learning_rate": 8.617564170997998e-06,
      "loss": 0.4239505767822266,
      "num_input_tokens_seen": 117993456,
      "step": 7205,
      "train_runtime": 13858.5137,
      "train_tokens_per_second": 8514.149
    },
    {
      "epoch": 0.48545650417452196,
      "grad_norm": 0.9386417227181725,
      "learning_rate": 8.61573842645388e-06,
      "loss": 0.4437280178070068,
      "num_input_tokens_seen": 118075376,
      "step": 7210,
      "train_runtime": 13868.0214,
      "train_tokens_per_second": 8514.219
    },
    {
      "epoch": 0.4857931591704821,
      "grad_norm": 1.032910757982502,
      "learning_rate": 8.61391167078004e-06,
      "loss": 0.42041854858398436,
      "num_input_tokens_seen": 118157296,
      "step": 7215,
      "train_runtime": 13877.6413,
      "train_tokens_per_second": 8514.22
    },
    {
      "epoch": 0.48612981416644224,
      "grad_norm": 1.0522971293993377,
      "learning_rate": 8.612083904487324e-06,
      "loss": 0.4569072723388672,
      "num_input_tokens_seen": 118239216,
      "step": 7220,
      "train_runtime": 13887.2114,
      "train_tokens_per_second": 8514.252
    },
    {
      "epoch": 0.4864664691624024,
      "grad_norm": 1.140454863718131,
      "learning_rate": 8.610255128086867e-06,
      "loss": 0.4206111907958984,
      "num_input_tokens_seen": 118321136,
      "step": 7225,
      "train_runtime": 13896.7188,
      "train_tokens_per_second": 8514.322
    },
    {
      "epoch": 0.4868031241583625,
      "grad_norm": 0.9815223655185399,
      "learning_rate": 8.60842534209007e-06,
      "loss": 0.3820094108581543,
      "num_input_tokens_seen": 118403056,
      "step": 7230,
      "train_runtime": 13906.2788,
      "train_tokens_per_second": 8514.359
    },
    {
      "epoch": 0.48713977915432266,
      "grad_norm": 1.1886014191438083,
      "learning_rate": 8.606594547008636e-06,
      "loss": 0.40130300521850587,
      "num_input_tokens_seen": 118484808,
      "step": 7235,
      "train_runtime": 13916.4051,
      "train_tokens_per_second": 8514.038
    },
    {
      "epoch": 0.4874764341502828,
      "grad_norm": 1.3254974545920115,
      "learning_rate": 8.604762743354536e-06,
      "loss": 0.4356574535369873,
      "num_input_tokens_seen": 118566728,
      "step": 7240,
      "train_runtime": 13925.9536,
      "train_tokens_per_second": 8514.083
    },
    {
      "epoch": 0.48781308914624294,
      "grad_norm": 1.1502840279942195,
      "learning_rate": 8.602929931640029e-06,
      "loss": 0.43790154457092284,
      "num_input_tokens_seen": 118648648,
      "step": 7245,
      "train_runtime": 13935.5188,
      "train_tokens_per_second": 8514.118
    },
    {
      "epoch": 0.4881497441422031,
      "grad_norm": 1.1246458485379194,
      "learning_rate": 8.601096112377659e-06,
      "loss": 0.4570526123046875,
      "num_input_tokens_seen": 118729632,
      "step": 7250,
      "train_runtime": 13945.3128,
      "train_tokens_per_second": 8513.945
    },
    {
      "epoch": 0.4884863991381632,
      "grad_norm": 1.1511186479412663,
      "learning_rate": 8.59926128608024e-06,
      "loss": 0.40047483444213866,
      "num_input_tokens_seen": 118811552,
      "step": 7255,
      "train_runtime": 13954.8775,
      "train_tokens_per_second": 8513.98
    },
    {
      "epoch": 0.48882305413412336,
      "grad_norm": 1.2772336391299166,
      "learning_rate": 8.597425453260881e-06,
      "loss": 0.42306041717529297,
      "num_input_tokens_seen": 118893472,
      "step": 7260,
      "train_runtime": 13964.4471,
      "train_tokens_per_second": 8514.012
    },
    {
      "epoch": 0.4891597091300835,
      "grad_norm": 1.031179531499706,
      "learning_rate": 8.595588614432967e-06,
      "loss": 0.3828181982040405,
      "num_input_tokens_seen": 118975352,
      "step": 7265,
      "train_runtime": 13974.6054,
      "train_tokens_per_second": 8513.682
    },
    {
      "epoch": 0.48949636412604364,
      "grad_norm": 1.0378815605569827,
      "learning_rate": 8.59375077011016e-06,
      "loss": 0.4078453540802002,
      "num_input_tokens_seen": 119057272,
      "step": 7270,
      "train_runtime": 13984.1366,
      "train_tokens_per_second": 8513.738
    },
    {
      "epoch": 0.4898330191220038,
      "grad_norm": 1.1868820160282818,
      "learning_rate": 8.591911920806412e-06,
      "loss": 0.4260712623596191,
      "num_input_tokens_seen": 119139192,
      "step": 7275,
      "train_runtime": 13993.669,
      "train_tokens_per_second": 8513.792
    },
    {
      "epoch": 0.4901696741179639,
      "grad_norm": 1.0836129390567133,
      "learning_rate": 8.590072067035947e-06,
      "loss": 0.39525861740112306,
      "num_input_tokens_seen": 119221112,
      "step": 7280,
      "train_runtime": 14003.231,
      "train_tokens_per_second": 8513.829
    },
    {
      "epoch": 0.49050632911392406,
      "grad_norm": 0.9563280291093432,
      "learning_rate": 8.588231209313278e-06,
      "loss": 0.4255088806152344,
      "num_input_tokens_seen": 119303032,
      "step": 7285,
      "train_runtime": 14012.8318,
      "train_tokens_per_second": 8513.842
    },
    {
      "epoch": 0.4908429841098842,
      "grad_norm": 1.1460008438272773,
      "learning_rate": 8.586389348153192e-06,
      "loss": 0.4007683277130127,
      "num_input_tokens_seen": 119384952,
      "step": 7290,
      "train_runtime": 14022.3497,
      "train_tokens_per_second": 8513.905
    },
    {
      "epoch": 0.49117963910584433,
      "grad_norm": 1.186623019315618,
      "learning_rate": 8.584546484070762e-06,
      "loss": 0.41766014099121096,
      "num_input_tokens_seen": 119466872,
      "step": 7295,
      "train_runtime": 14031.9638,
      "train_tokens_per_second": 8513.91
    },
    {
      "epoch": 0.4915162941018045,
      "grad_norm": 1.1785808310709098,
      "learning_rate": 8.582702617581338e-06,
      "loss": 0.4142894744873047,
      "num_input_tokens_seen": 119548792,
      "step": 7300,
      "train_runtime": 14041.59,
      "train_tokens_per_second": 8513.907
    },
    {
      "epoch": 0.4918529490977646,
      "grad_norm": 1.039644939925374,
      "learning_rate": 8.58085774920055e-06,
      "loss": 0.38299376964569093,
      "num_input_tokens_seen": 119630712,
      "step": 7305,
      "train_runtime": 14051.0923,
      "train_tokens_per_second": 8513.98
    },
    {
      "epoch": 0.49218960409372475,
      "grad_norm": 1.3430920412589027,
      "learning_rate": 8.579011879444313e-06,
      "loss": 0.47650656700134275,
      "num_input_tokens_seen": 119712632,
      "step": 7310,
      "train_runtime": 14060.7074,
      "train_tokens_per_second": 8513.984
    },
    {
      "epoch": 0.4925262590896849,
      "grad_norm": 1.0391184805421807,
      "learning_rate": 8.577165008828817e-06,
      "loss": 0.3834456205368042,
      "num_input_tokens_seen": 119794552,
      "step": 7315,
      "train_runtime": 14070.3146,
      "train_tokens_per_second": 8513.992
    },
    {
      "epoch": 0.49286291408564503,
      "grad_norm": 1.0364062521195134,
      "learning_rate": 8.57531713787053e-06,
      "loss": 0.4184147357940674,
      "num_input_tokens_seen": 119876016,
      "step": 7320,
      "train_runtime": 14080.1854,
      "train_tokens_per_second": 8513.809
    },
    {
      "epoch": 0.49319956908160517,
      "grad_norm": 1.0917037549279804,
      "learning_rate": 8.573468267086208e-06,
      "loss": 0.38274459838867186,
      "num_input_tokens_seen": 119957936,
      "step": 7325,
      "train_runtime": 14089.7498,
      "train_tokens_per_second": 8513.844
    },
    {
      "epoch": 0.4935362240775653,
      "grad_norm": 1.1139736577446437,
      "learning_rate": 8.57161839699288e-06,
      "loss": 0.3747070074081421,
      "num_input_tokens_seen": 120039856,
      "step": 7330,
      "train_runtime": 14099.3748,
      "train_tokens_per_second": 8513.842
    },
    {
      "epoch": 0.49387287907352545,
      "grad_norm": 1.08462094244019,
      "learning_rate": 8.569767528107857e-06,
      "loss": 0.4115703582763672,
      "num_input_tokens_seen": 120121776,
      "step": 7335,
      "train_runtime": 14108.9322,
      "train_tokens_per_second": 8513.881
    },
    {
      "epoch": 0.4942095340694856,
      "grad_norm": 0.9893998342992253,
      "learning_rate": 8.567915660948727e-06,
      "loss": 0.3717326641082764,
      "num_input_tokens_seen": 120203696,
      "step": 7340,
      "train_runtime": 14118.5026,
      "train_tokens_per_second": 8513.913
    },
    {
      "epoch": 0.49454618906544573,
      "grad_norm": 1.172288041337472,
      "learning_rate": 8.566062796033358e-06,
      "loss": 0.4356529712677002,
      "num_input_tokens_seen": 120285344,
      "step": 7345,
      "train_runtime": 14128.5188,
      "train_tokens_per_second": 8513.656
    },
    {
      "epoch": 0.49488284406140587,
      "grad_norm": 1.0754760225465065,
      "learning_rate": 8.564208933879903e-06,
      "loss": 0.36146609783172606,
      "num_input_tokens_seen": 120366928,
      "step": 7350,
      "train_runtime": 14138.5202,
      "train_tokens_per_second": 8513.404
    },
    {
      "epoch": 0.495219499057366,
      "grad_norm": 1.0327675609419813,
      "learning_rate": 8.562354075006782e-06,
      "loss": 0.3748115062713623,
      "num_input_tokens_seen": 120448848,
      "step": 7355,
      "train_runtime": 14148.0212,
      "train_tokens_per_second": 8513.477
    },
    {
      "epoch": 0.49555615405332615,
      "grad_norm": 1.3361365506057734,
      "learning_rate": 8.560498219932706e-06,
      "loss": 0.44521136283874513,
      "num_input_tokens_seen": 120530768,
      "step": 7360,
      "train_runtime": 14157.6341,
      "train_tokens_per_second": 8513.482
    },
    {
      "epoch": 0.4958928090492863,
      "grad_norm": 1.1450493973700064,
      "learning_rate": 8.558641369176654e-06,
      "loss": 0.36314513683319094,
      "num_input_tokens_seen": 120612688,
      "step": 7365,
      "train_runtime": 14167.2532,
      "train_tokens_per_second": 8513.484
    },
    {
      "epoch": 0.4962294640452464,
      "grad_norm": 1.1367734140863452,
      "learning_rate": 8.556783523257892e-06,
      "loss": 0.3947627067565918,
      "num_input_tokens_seen": 120694608,
      "step": 7370,
      "train_runtime": 14176.7526,
      "train_tokens_per_second": 8513.558
    },
    {
      "epoch": 0.49656611904120657,
      "grad_norm": 1.0337388907654623,
      "learning_rate": 8.554924682695959e-06,
      "loss": 0.3784221887588501,
      "num_input_tokens_seen": 120776528,
      "step": 7375,
      "train_runtime": 14186.3704,
      "train_tokens_per_second": 8513.561
    },
    {
      "epoch": 0.4969027740371667,
      "grad_norm": 0.9373458267117616,
      "learning_rate": 8.553064848010677e-06,
      "loss": 0.39788570404052737,
      "num_input_tokens_seen": 120858024,
      "step": 7380,
      "train_runtime": 14196.3428,
      "train_tokens_per_second": 8513.321
    },
    {
      "epoch": 0.49723942903312685,
      "grad_norm": 1.1001247021412015,
      "learning_rate": 8.551204019722137e-06,
      "loss": 0.39778852462768555,
      "num_input_tokens_seen": 120939944,
      "step": 7385,
      "train_runtime": 14205.8607,
      "train_tokens_per_second": 8513.384
    },
    {
      "epoch": 0.497576084029087,
      "grad_norm": 1.3117029074552518,
      "learning_rate": 8.549342198350718e-06,
      "loss": 0.4360301971435547,
      "num_input_tokens_seen": 121021864,
      "step": 7390,
      "train_runtime": 14215.4241,
      "train_tokens_per_second": 8513.419
    },
    {
      "epoch": 0.4979127390250471,
      "grad_norm": 1.1228624111771512,
      "learning_rate": 8.54747938441707e-06,
      "loss": 0.4019890785217285,
      "num_input_tokens_seen": 121103784,
      "step": 7395,
      "train_runtime": 14224.9982,
      "train_tokens_per_second": 8513.448
    },
    {
      "epoch": 0.49824939402100726,
      "grad_norm": 0.9244869826862014,
      "learning_rate": 8.545615578442126e-06,
      "loss": 0.36618263721466066,
      "num_input_tokens_seen": 121185704,
      "step": 7400,
      "train_runtime": 14234.5513,
      "train_tokens_per_second": 8513.49
    },
    {
      "epoch": 0.4985860490169674,
      "grad_norm": 1.2689522730218072,
      "learning_rate": 8.54375078094709e-06,
      "loss": 0.3851290464401245,
      "num_input_tokens_seen": 121267624,
      "step": 7405,
      "train_runtime": 14244.1127,
      "train_tokens_per_second": 8513.526
    },
    {
      "epoch": 0.49892270401292754,
      "grad_norm": 0.9469010441726762,
      "learning_rate": 8.541884992453449e-06,
      "loss": 0.4072246551513672,
      "num_input_tokens_seen": 121349544,
      "step": 7410,
      "train_runtime": 14253.7059,
      "train_tokens_per_second": 8513.543
    },
    {
      "epoch": 0.4992593590088877,
      "grad_norm": 1.0521491216712286,
      "learning_rate": 8.540018213482966e-06,
      "loss": 0.40680108070373533,
      "num_input_tokens_seen": 121431464,
      "step": 7415,
      "train_runtime": 14263.2163,
      "train_tokens_per_second": 8513.61
    },
    {
      "epoch": 0.4995960140048478,
      "grad_norm": 1.1807332480570796,
      "learning_rate": 8.538150444557676e-06,
      "loss": 0.393707275390625,
      "num_input_tokens_seen": 121513384,
      "step": 7420,
      "train_runtime": 14272.7692,
      "train_tokens_per_second": 8513.652
    },
    {
      "epoch": 0.49993266900080796,
      "grad_norm": 1.1932029774448356,
      "learning_rate": 8.536281686199896e-06,
      "loss": 0.4214649677276611,
      "num_input_tokens_seen": 121594864,
      "step": 7425,
      "train_runtime": 14282.3842,
      "train_tokens_per_second": 8513.625
    },
    {
      "epoch": 0.5002693239967682,
      "grad_norm": 0.9934450596245259,
      "learning_rate": 8.534411938932218e-06,
      "loss": 0.370166277885437,
      "num_input_tokens_seen": 121676784,
      "step": 7430,
      "train_runtime": 14291.9903,
      "train_tokens_per_second": 8513.635
    },
    {
      "epoch": 0.5006059789927283,
      "grad_norm": 0.9904254850886245,
      "learning_rate": 8.532541203277515e-06,
      "loss": 0.38652637004852297,
      "num_input_tokens_seen": 121758704,
      "step": 7435,
      "train_runtime": 14301.4978,
      "train_tokens_per_second": 8513.703
    },
    {
      "epoch": 0.5009426339886884,
      "grad_norm": 1.2608563449088985,
      "learning_rate": 8.530669479758926e-06,
      "loss": 0.38618447780609133,
      "num_input_tokens_seen": 121840624,
      "step": 7440,
      "train_runtime": 14311.1365,
      "train_tokens_per_second": 8513.693
    },
    {
      "epoch": 0.5012792889846486,
      "grad_norm": 1.192320962388343,
      "learning_rate": 8.528796768899878e-06,
      "loss": 0.3955865383148193,
      "num_input_tokens_seen": 121922544,
      "step": 7445,
      "train_runtime": 14320.7067,
      "train_tokens_per_second": 8513.724
    },
    {
      "epoch": 0.5016159439806087,
      "grad_norm": 1.102841987813258,
      "learning_rate": 8.526923071224064e-06,
      "loss": 0.41722803115844725,
      "num_input_tokens_seen": 122004464,
      "step": 7450,
      "train_runtime": 14330.2172,
      "train_tokens_per_second": 8513.79
    },
    {
      "epoch": 0.5019525989765689,
      "grad_norm": 1.2155378473074556,
      "learning_rate": 8.525048387255461e-06,
      "loss": 0.4317512035369873,
      "num_input_tokens_seen": 122086384,
      "step": 7455,
      "train_runtime": 14339.834,
      "train_tokens_per_second": 8513.793
    },
    {
      "epoch": 0.502289253972529,
      "grad_norm": 1.116546345297734,
      "learning_rate": 8.523172717518315e-06,
      "loss": 0.3858393907546997,
      "num_input_tokens_seen": 122168304,
      "step": 7460,
      "train_runtime": 14349.4434,
      "train_tokens_per_second": 8513.801
    },
    {
      "epoch": 0.5026259089684891,
      "grad_norm": 1.1425167900133222,
      "learning_rate": 8.521296062537156e-06,
      "loss": 0.37255263328552246,
      "num_input_tokens_seen": 122250224,
      "step": 7465,
      "train_runtime": 14358.9396,
      "train_tokens_per_second": 8513.875
    },
    {
      "epoch": 0.5029625639644493,
      "grad_norm": 1.094601341615734,
      "learning_rate": 8.51941842283678e-06,
      "loss": 0.3961034774780273,
      "num_input_tokens_seen": 122332144,
      "step": 7470,
      "train_runtime": 14368.5655,
      "train_tokens_per_second": 8513.873
    },
    {
      "epoch": 0.5032992189604094,
      "grad_norm": 1.2381558983664145,
      "learning_rate": 8.517539798942265e-06,
      "loss": 0.41904258728027344,
      "num_input_tokens_seen": 122414064,
      "step": 7475,
      "train_runtime": 14378.1927,
      "train_tokens_per_second": 8513.87
    },
    {
      "epoch": 0.5036358739563696,
      "grad_norm": 1.022679899377744,
      "learning_rate": 8.515660191378962e-06,
      "loss": 0.4166261672973633,
      "num_input_tokens_seen": 122495984,
      "step": 7480,
      "train_runtime": 14387.7482,
      "train_tokens_per_second": 8513.909
    },
    {
      "epoch": 0.5039725289523297,
      "grad_norm": 0.9562003523698371,
      "learning_rate": 8.513779600672495e-06,
      "loss": 0.381156849861145,
      "num_input_tokens_seen": 122577904,
      "step": 7485,
      "train_runtime": 14397.2989,
      "train_tokens_per_second": 8513.951
    },
    {
      "epoch": 0.5043091839482898,
      "grad_norm": 1.0958214728297355,
      "learning_rate": 8.511898027348771e-06,
      "loss": 0.40671114921569823,
      "num_input_tokens_seen": 122659824,
      "step": 7490,
      "train_runtime": 14406.949,
      "train_tokens_per_second": 8513.935
    },
    {
      "epoch": 0.50464583894425,
      "grad_norm": 1.006219692790885,
      "learning_rate": 8.51001547193396e-06,
      "loss": 0.36387972831726073,
      "num_input_tokens_seen": 122741744,
      "step": 7495,
      "train_runtime": 14416.5124,
      "train_tokens_per_second": 8513.969
    },
    {
      "epoch": 0.5049824939402101,
      "grad_norm": 1.0560562232861326,
      "learning_rate": 8.508131934954515e-06,
      "loss": 0.407363224029541,
      "num_input_tokens_seen": 122823664,
      "step": 7500,
      "train_runtime": 14426.0836,
      "train_tokens_per_second": 8513.999
    },
    {
      "epoch": 0.5053191489361702,
      "grad_norm": 1.1322988393201339,
      "learning_rate": 8.50624741693716e-06,
      "loss": 0.398145055770874,
      "num_input_tokens_seen": 122904576,
      "step": 7505,
      "train_runtime": 14435.8636,
      "train_tokens_per_second": 8513.836
    },
    {
      "epoch": 0.5056558039321304,
      "grad_norm": 1.236874967815779,
      "learning_rate": 8.5043619184089e-06,
      "loss": 0.43453192710876465,
      "num_input_tokens_seen": 122986496,
      "step": 7510,
      "train_runtime": 14445.4718,
      "train_tokens_per_second": 8513.844
    },
    {
      "epoch": 0.5059924589280905,
      "grad_norm": 1.1015228695270525,
      "learning_rate": 8.502475439897e-06,
      "loss": 0.44918174743652345,
      "num_input_tokens_seen": 123068416,
      "step": 7515,
      "train_runtime": 14454.9853,
      "train_tokens_per_second": 8513.908
    },
    {
      "epoch": 0.5063291139240507,
      "grad_norm": 1.1648339183415957,
      "learning_rate": 8.500587981929013e-06,
      "loss": 0.4361621379852295,
      "num_input_tokens_seen": 123150336,
      "step": 7520,
      "train_runtime": 14464.6057,
      "train_tokens_per_second": 8513.909
    },
    {
      "epoch": 0.5066657689200108,
      "grad_norm": 1.2601449428228826,
      "learning_rate": 8.498699545032758e-06,
      "loss": 0.4147230625152588,
      "num_input_tokens_seen": 123232256,
      "step": 7525,
      "train_runtime": 14474.2108,
      "train_tokens_per_second": 8513.919
    },
    {
      "epoch": 0.507002423915971,
      "grad_norm": 1.1389473370576613,
      "learning_rate": 8.496810129736334e-06,
      "loss": 0.4178779125213623,
      "num_input_tokens_seen": 123314176,
      "step": 7530,
      "train_runtime": 14483.7122,
      "train_tokens_per_second": 8513.99
    },
    {
      "epoch": 0.5073390789119311,
      "grad_norm": 1.1736520178945613,
      "learning_rate": 8.494919736568105e-06,
      "loss": 0.4100910186767578,
      "num_input_tokens_seen": 123396096,
      "step": 7535,
      "train_runtime": 14493.2591,
      "train_tokens_per_second": 8514.034
    },
    {
      "epoch": 0.5076757339078912,
      "grad_norm": 1.1503149634064376,
      "learning_rate": 8.49302836605672e-06,
      "loss": 0.4106430530548096,
      "num_input_tokens_seen": 123478016,
      "step": 7540,
      "train_runtime": 14502.8555,
      "train_tokens_per_second": 8514.049
    },
    {
      "epoch": 0.5080123889038514,
      "grad_norm": 1.0415316552232528,
      "learning_rate": 8.491136018731088e-06,
      "loss": 0.38987834453582765,
      "num_input_tokens_seen": 123559936,
      "step": 7545,
      "train_runtime": 14512.3574,
      "train_tokens_per_second": 8514.119
    },
    {
      "epoch": 0.5083490438998115,
      "grad_norm": 1.4061073523819743,
      "learning_rate": 8.4892426951204e-06,
      "loss": 0.4293052673339844,
      "num_input_tokens_seen": 123641856,
      "step": 7550,
      "train_runtime": 14521.9596,
      "train_tokens_per_second": 8514.13
    },
    {
      "epoch": 0.5086856988957716,
      "grad_norm": 1.0514520649990067,
      "learning_rate": 8.48734839575412e-06,
      "loss": 0.4163840293884277,
      "num_input_tokens_seen": 123723776,
      "step": 7555,
      "train_runtime": 14531.5725,
      "train_tokens_per_second": 8514.135
    },
    {
      "epoch": 0.5090223538917318,
      "grad_norm": 1.0137600926478274,
      "learning_rate": 8.485453121161983e-06,
      "loss": 0.4081639289855957,
      "num_input_tokens_seen": 123805696,
      "step": 7560,
      "train_runtime": 14541.075,
      "train_tokens_per_second": 8514.205
    },
    {
      "epoch": 0.5093590088876919,
      "grad_norm": 1.2416754789525652,
      "learning_rate": 8.483556871873993e-06,
      "loss": 0.45699353218078614,
      "num_input_tokens_seen": 123887616,
      "step": 7565,
      "train_runtime": 14550.6982,
      "train_tokens_per_second": 8514.204
    },
    {
      "epoch": 0.5096956638836521,
      "grad_norm": 1.4121956994097453,
      "learning_rate": 8.481659648420433e-06,
      "loss": 0.41403894424438475,
      "num_input_tokens_seen": 123969536,
      "step": 7570,
      "train_runtime": 14560.2616,
      "train_tokens_per_second": 8514.238
    },
    {
      "epoch": 0.5100323188796122,
      "grad_norm": 1.4085681753202348,
      "learning_rate": 8.479761451331855e-06,
      "loss": 0.3721931457519531,
      "num_input_tokens_seen": 124051456,
      "step": 7575,
      "train_runtime": 14569.8227,
      "train_tokens_per_second": 8514.274
    },
    {
      "epoch": 0.5103689738755723,
      "grad_norm": 1.1086129773800872,
      "learning_rate": 8.477862281139082e-06,
      "loss": 0.3570922136306763,
      "num_input_tokens_seen": 124133376,
      "step": 7580,
      "train_runtime": 14579.3877,
      "train_tokens_per_second": 8514.307
    },
    {
      "epoch": 0.5107056288715325,
      "grad_norm": 1.1234705839853985,
      "learning_rate": 8.475962138373212e-06,
      "loss": 0.3972013473510742,
      "num_input_tokens_seen": 124215296,
      "step": 7585,
      "train_runtime": 14589.0048,
      "train_tokens_per_second": 8514.309
    },
    {
      "epoch": 0.5110422838674926,
      "grad_norm": 1.2565410855180659,
      "learning_rate": 8.474061023565614e-06,
      "loss": 0.3991823196411133,
      "num_input_tokens_seen": 124297216,
      "step": 7590,
      "train_runtime": 14598.5579,
      "train_tokens_per_second": 8514.349
    },
    {
      "epoch": 0.5113789388634528,
      "grad_norm": 1.051317822984955,
      "learning_rate": 8.472158937247931e-06,
      "loss": 0.4757996559143066,
      "num_input_tokens_seen": 124379136,
      "step": 7595,
      "train_runtime": 14608.1237,
      "train_tokens_per_second": 8514.381
    },
    {
      "epoch": 0.5117155938594129,
      "grad_norm": 1.0161562700606335,
      "learning_rate": 8.47025587995207e-06,
      "loss": 0.40506410598754883,
      "num_input_tokens_seen": 124460424,
      "step": 7600,
      "train_runtime": 14618.0122,
      "train_tokens_per_second": 8514.183
    },
    {
      "epoch": 0.512052248855373,
      "grad_norm": 1.226592316382326,
      "learning_rate": 8.46835185221022e-06,
      "loss": 0.4246683597564697,
      "num_input_tokens_seen": 124542096,
      "step": 7605,
      "train_runtime": 14628.0776,
      "train_tokens_per_second": 8513.907
    },
    {
      "epoch": 0.5123889038513332,
      "grad_norm": 1.0715556671424453,
      "learning_rate": 8.466446854554835e-06,
      "loss": 0.3678670167922974,
      "num_input_tokens_seen": 124624016,
      "step": 7610,
      "train_runtime": 14637.5751,
      "train_tokens_per_second": 8513.98
    },
    {
      "epoch": 0.5127255588472933,
      "grad_norm": 1.1759496527438944,
      "learning_rate": 8.464540887518638e-06,
      "loss": 0.37921743392944335,
      "num_input_tokens_seen": 124705936,
      "step": 7615,
      "train_runtime": 14647.1889,
      "train_tokens_per_second": 8513.984
    },
    {
      "epoch": 0.5130622138432535,
      "grad_norm": 1.1904960727690406,
      "learning_rate": 8.462633951634631e-06,
      "loss": 0.4109151840209961,
      "num_input_tokens_seen": 124787856,
      "step": 7620,
      "train_runtime": 14656.8001,
      "train_tokens_per_second": 8513.99
    },
    {
      "epoch": 0.5133988688392136,
      "grad_norm": 1.170591458302791,
      "learning_rate": 8.46072604743608e-06,
      "loss": 0.3838709831237793,
      "num_input_tokens_seen": 124869776,
      "step": 7625,
      "train_runtime": 14666.3168,
      "train_tokens_per_second": 8514.051
    },
    {
      "epoch": 0.5137355238351737,
      "grad_norm": 1.149511442684928,
      "learning_rate": 8.458817175456528e-06,
      "loss": 0.43401031494140624,
      "num_input_tokens_seen": 124951696,
      "step": 7630,
      "train_runtime": 14675.9412,
      "train_tokens_per_second": 8514.05
    },
    {
      "epoch": 0.5140721788311339,
      "grad_norm": 0.9969102361101756,
      "learning_rate": 8.45690733622978e-06,
      "loss": 0.4065235137939453,
      "num_input_tokens_seen": 125033616,
      "step": 7635,
      "train_runtime": 14685.5389,
      "train_tokens_per_second": 8514.064
    },
    {
      "epoch": 0.514408833827094,
      "grad_norm": 1.169078753562118,
      "learning_rate": 8.45499653028992e-06,
      "loss": 0.3705174446105957,
      "num_input_tokens_seen": 125115536,
      "step": 7640,
      "train_runtime": 14695.091,
      "train_tokens_per_second": 8514.104
    },
    {
      "epoch": 0.5147454888230542,
      "grad_norm": 1.1574682839640904,
      "learning_rate": 8.453084758171297e-06,
      "loss": 0.394371485710144,
      "num_input_tokens_seen": 125197456,
      "step": 7645,
      "train_runtime": 14704.6724,
      "train_tokens_per_second": 8514.128
    },
    {
      "epoch": 0.5150821438190143,
      "grad_norm": 1.1009532473612849,
      "learning_rate": 8.451172020408532e-06,
      "loss": 0.4080966949462891,
      "num_input_tokens_seen": 125279376,
      "step": 7650,
      "train_runtime": 14714.2574,
      "train_tokens_per_second": 8514.149
    },
    {
      "epoch": 0.5154187988149744,
      "grad_norm": 0.8870361133509053,
      "learning_rate": 8.449258317536517e-06,
      "loss": 0.3594048976898193,
      "num_input_tokens_seen": 125361296,
      "step": 7655,
      "train_runtime": 14723.8058,
      "train_tokens_per_second": 8514.191
    },
    {
      "epoch": 0.5157554538109346,
      "grad_norm": 1.134507684408562,
      "learning_rate": 8.447343650090412e-06,
      "loss": 0.3328833103179932,
      "num_input_tokens_seen": 125443216,
      "step": 7660,
      "train_runtime": 14733.3534,
      "train_tokens_per_second": 8514.234
    },
    {
      "epoch": 0.5160921088068947,
      "grad_norm": 1.2748615614901015,
      "learning_rate": 8.445428018605647e-06,
      "loss": 0.42148466110229493,
      "num_input_tokens_seen": 125525136,
      "step": 7665,
      "train_runtime": 14742.9891,
      "train_tokens_per_second": 8514.226
    },
    {
      "epoch": 0.5164287638028549,
      "grad_norm": 1.002596939662975,
      "learning_rate": 8.443511423617924e-06,
      "loss": 0.34884042739868165,
      "num_input_tokens_seen": 125607056,
      "step": 7670,
      "train_runtime": 14752.547,
      "train_tokens_per_second": 8514.262
    },
    {
      "epoch": 0.516765418798815,
      "grad_norm": 1.052001678835538,
      "learning_rate": 8.441593865663213e-06,
      "loss": 0.43817939758300783,
      "num_input_tokens_seen": 125688976,
      "step": 7675,
      "train_runtime": 14762.1061,
      "train_tokens_per_second": 8514.298
    },
    {
      "epoch": 0.5171020737947751,
      "grad_norm": 1.1351482272355,
      "learning_rate": 8.439675345277751e-06,
      "loss": 0.42105445861816404,
      "num_input_tokens_seen": 125770896,
      "step": 7680,
      "train_runtime": 14771.7066,
      "train_tokens_per_second": 8514.31
    },
    {
      "epoch": 0.5174387287907353,
      "grad_norm": 1.1901174202066036,
      "learning_rate": 8.437755862998047e-06,
      "loss": 0.4355765342712402,
      "num_input_tokens_seen": 125852816,
      "step": 7685,
      "train_runtime": 14781.3252,
      "train_tokens_per_second": 8514.312
    },
    {
      "epoch": 0.5177753837866954,
      "grad_norm": 1.0236365396992753,
      "learning_rate": 8.435835419360878e-06,
      "loss": 0.42244529724121094,
      "num_input_tokens_seen": 125934736,
      "step": 7690,
      "train_runtime": 14790.8314,
      "train_tokens_per_second": 8514.378
    },
    {
      "epoch": 0.5181120387826555,
      "grad_norm": 1.1594444861105413,
      "learning_rate": 8.433914014903291e-06,
      "loss": 0.4402174949645996,
      "num_input_tokens_seen": 126016656,
      "step": 7695,
      "train_runtime": 14800.4551,
      "train_tokens_per_second": 8514.377
    },
    {
      "epoch": 0.5184486937786157,
      "grad_norm": 1.1906217605368197,
      "learning_rate": 8.431991650162599e-06,
      "loss": 0.435107946395874,
      "num_input_tokens_seen": 126098576,
      "step": 7700,
      "train_runtime": 14810.0676,
      "train_tokens_per_second": 8514.382
    },
    {
      "epoch": 0.5187853487745758,
      "grad_norm": 1.0965618945326503,
      "learning_rate": 8.430068325676386e-06,
      "loss": 0.40163512229919435,
      "num_input_tokens_seen": 126180496,
      "step": 7705,
      "train_runtime": 14819.5779,
      "train_tokens_per_second": 8514.446
    },
    {
      "epoch": 0.519122003770536,
      "grad_norm": 1.1728459691071083,
      "learning_rate": 8.428144041982502e-06,
      "loss": 0.44342746734619143,
      "num_input_tokens_seen": 126262416,
      "step": 7710,
      "train_runtime": 14829.1828,
      "train_tokens_per_second": 8514.455
    },
    {
      "epoch": 0.5194586587664961,
      "grad_norm": 1.0108491268326742,
      "learning_rate": 8.42621879961907e-06,
      "loss": 0.390611457824707,
      "num_input_tokens_seen": 126344336,
      "step": 7715,
      "train_runtime": 14838.799,
      "train_tokens_per_second": 8514.458
    },
    {
      "epoch": 0.5197953137624562,
      "grad_norm": 1.118819823690714,
      "learning_rate": 8.424292599124476e-06,
      "loss": 0.37361648082733157,
      "num_input_tokens_seen": 126426256,
      "step": 7720,
      "train_runtime": 14848.3039,
      "train_tokens_per_second": 8514.525
    },
    {
      "epoch": 0.5201319687584164,
      "grad_norm": 1.1369171134293954,
      "learning_rate": 8.422365441037374e-06,
      "loss": 0.43616280555725095,
      "num_input_tokens_seen": 126508176,
      "step": 7725,
      "train_runtime": 14857.9225,
      "train_tokens_per_second": 8514.527
    },
    {
      "epoch": 0.5204686237543765,
      "grad_norm": 1.4487266242533883,
      "learning_rate": 8.420437325896692e-06,
      "loss": 0.43758788108825686,
      "num_input_tokens_seen": 126590096,
      "step": 7730,
      "train_runtime": 14867.5405,
      "train_tokens_per_second": 8514.528
    },
    {
      "epoch": 0.5208052787503367,
      "grad_norm": 1.2897657021822937,
      "learning_rate": 8.418508254241617e-06,
      "loss": 0.4439405918121338,
      "num_input_tokens_seen": 126672016,
      "step": 7735,
      "train_runtime": 14877.0949,
      "train_tokens_per_second": 8514.567
    },
    {
      "epoch": 0.5211419337462968,
      "grad_norm": 1.0161605021645348,
      "learning_rate": 8.416578226611612e-06,
      "loss": 0.430496883392334,
      "num_input_tokens_seen": 126753936,
      "step": 7740,
      "train_runtime": 14886.6549,
      "train_tokens_per_second": 8514.602
    },
    {
      "epoch": 0.5214785887422569,
      "grad_norm": 2.4794220348839815,
      "learning_rate": 8.414647243546398e-06,
      "loss": 0.362098240852356,
      "num_input_tokens_seen": 126835856,
      "step": 7745,
      "train_runtime": 14896.2695,
      "train_tokens_per_second": 8514.605
    },
    {
      "epoch": 0.5218152437382171,
      "grad_norm": 1.0744479991866056,
      "learning_rate": 8.412715305585973e-06,
      "loss": 0.4332843780517578,
      "num_input_tokens_seen": 126917776,
      "step": 7750,
      "train_runtime": 14905.7788,
      "train_tokens_per_second": 8514.669
    },
    {
      "epoch": 0.5221518987341772,
      "grad_norm": 1.0299533253274453,
      "learning_rate": 8.410782413270594e-06,
      "loss": 0.4018895626068115,
      "num_input_tokens_seen": 126999696,
      "step": 7755,
      "train_runtime": 14915.3389,
      "train_tokens_per_second": 8514.704
    },
    {
      "epoch": 0.5224885537301374,
      "grad_norm": 0.9738685643835907,
      "learning_rate": 8.408848567140787e-06,
      "loss": 0.37138638496398924,
      "num_input_tokens_seen": 127081616,
      "step": 7760,
      "train_runtime": 14924.9413,
      "train_tokens_per_second": 8514.715
    },
    {
      "epoch": 0.5228252087260975,
      "grad_norm": 1.1176389575374654,
      "learning_rate": 8.406913767737353e-06,
      "loss": 0.37472920417785643,
      "num_input_tokens_seen": 127163536,
      "step": 7765,
      "train_runtime": 14934.5544,
      "train_tokens_per_second": 8514.719
    },
    {
      "epoch": 0.5231618637220576,
      "grad_norm": 0.9714705872381953,
      "learning_rate": 8.404978015601344e-06,
      "loss": 0.3790732860565186,
      "num_input_tokens_seen": 127245456,
      "step": 7770,
      "train_runtime": 14944.0599,
      "train_tokens_per_second": 8514.785
    },
    {
      "epoch": 0.5234985187180178,
      "grad_norm": 0.9600456981761779,
      "learning_rate": 8.403041311274091e-06,
      "loss": 0.43526811599731446,
      "num_input_tokens_seen": 127327376,
      "step": 7775,
      "train_runtime": 14953.6791,
      "train_tokens_per_second": 8514.786
    },
    {
      "epoch": 0.5238351737139779,
      "grad_norm": 1.2338879245815817,
      "learning_rate": 8.401103655297188e-06,
      "loss": 0.4038038730621338,
      "num_input_tokens_seen": 127409296,
      "step": 7780,
      "train_runtime": 14963.2945,
      "train_tokens_per_second": 8514.789
    },
    {
      "epoch": 0.5241718287099381,
      "grad_norm": 1.0691889461069746,
      "learning_rate": 8.399165048212494e-06,
      "loss": 0.3563144445419312,
      "num_input_tokens_seen": 127491216,
      "step": 7785,
      "train_runtime": 14972.8082,
      "train_tokens_per_second": 8514.85
    },
    {
      "epoch": 0.5245084837058982,
      "grad_norm": 1.1385710990935967,
      "learning_rate": 8.397225490562131e-06,
      "loss": 0.43369412422180176,
      "num_input_tokens_seen": 127573136,
      "step": 7790,
      "train_runtime": 14982.4149,
      "train_tokens_per_second": 8514.858
    },
    {
      "epoch": 0.5248451387018583,
      "grad_norm": 1.223294071187143,
      "learning_rate": 8.395284982888494e-06,
      "loss": 0.41280813217163087,
      "num_input_tokens_seen": 127655056,
      "step": 7795,
      "train_runtime": 14991.9758,
      "train_tokens_per_second": 8514.892
    },
    {
      "epoch": 0.5251817936978185,
      "grad_norm": 0.9535070192871427,
      "learning_rate": 8.393343525734239e-06,
      "loss": 0.3936163902282715,
      "num_input_tokens_seen": 127736976,
      "step": 7800,
      "train_runtime": 15001.4877,
      "train_tokens_per_second": 8514.954
    },
    {
      "epoch": 0.5255184486937786,
      "grad_norm": 0.9435666397116484,
      "learning_rate": 8.391401119642286e-06,
      "loss": 0.41623945236206056,
      "num_input_tokens_seen": 127818456,
      "step": 7805,
      "train_runtime": 15012.0566,
      "train_tokens_per_second": 8514.387
    },
    {
      "epoch": 0.5258551036897388,
      "grad_norm": 1.083011259322557,
      "learning_rate": 8.389457765155824e-06,
      "loss": 0.41552085876464845,
      "num_input_tokens_seen": 127900376,
      "step": 7810,
      "train_runtime": 15021.6738,
      "train_tokens_per_second": 8514.389
    },
    {
      "epoch": 0.5261917586856989,
      "grad_norm": 4.338742590823421,
      "learning_rate": 8.387513462818309e-06,
      "loss": 0.3654113531112671,
      "num_input_tokens_seen": 127982296,
      "step": 7815,
      "train_runtime": 15031.2149,
      "train_tokens_per_second": 8514.435
    },
    {
      "epoch": 0.526528413681659,
      "grad_norm": 1.1511438745946707,
      "learning_rate": 8.385568213173453e-06,
      "loss": 0.41886024475097655,
      "num_input_tokens_seen": 128064216,
      "step": 7820,
      "train_runtime": 15040.7736,
      "train_tokens_per_second": 8514.47
    },
    {
      "epoch": 0.5268650686776192,
      "grad_norm": 1.1613206966659348,
      "learning_rate": 8.383622016765242e-06,
      "loss": 0.3846519231796265,
      "num_input_tokens_seen": 128146136,
      "step": 7825,
      "train_runtime": 15050.4147,
      "train_tokens_per_second": 8514.459
    },
    {
      "epoch": 0.5272017236735793,
      "grad_norm": 0.9456586616915154,
      "learning_rate": 8.381674874137926e-06,
      "loss": 0.40617856979370115,
      "num_input_tokens_seen": 128228056,
      "step": 7830,
      "train_runtime": 15059.9195,
      "train_tokens_per_second": 8514.525
    },
    {
      "epoch": 0.5275383786695395,
      "grad_norm": 0.9907939493310004,
      "learning_rate": 8.379726785836013e-06,
      "loss": 0.39137630462646483,
      "num_input_tokens_seen": 128309976,
      "step": 7835,
      "train_runtime": 15069.494,
      "train_tokens_per_second": 8514.551
    },
    {
      "epoch": 0.5278750336654996,
      "grad_norm": 1.0580065035771204,
      "learning_rate": 8.377777752404283e-06,
      "loss": 0.408371639251709,
      "num_input_tokens_seen": 128391896,
      "step": 7840,
      "train_runtime": 15079.1017,
      "train_tokens_per_second": 8514.559
    },
    {
      "epoch": 0.5282116886614597,
      "grad_norm": 0.9206436853299813,
      "learning_rate": 8.375827774387774e-06,
      "loss": 0.38453381061553954,
      "num_input_tokens_seen": 128473816,
      "step": 7845,
      "train_runtime": 15088.7078,
      "train_tokens_per_second": 8514.567
    },
    {
      "epoch": 0.5285483436574199,
      "grad_norm": 1.0508030338292391,
      "learning_rate": 8.373876852331793e-06,
      "loss": 0.36423392295837403,
      "num_input_tokens_seen": 128555736,
      "step": 7850,
      "train_runtime": 15098.2095,
      "train_tokens_per_second": 8514.635
    },
    {
      "epoch": 0.52888499865338,
      "grad_norm": 1.15160136577757,
      "learning_rate": 8.371924986781909e-06,
      "loss": 0.4098381042480469,
      "num_input_tokens_seen": 128637656,
      "step": 7855,
      "train_runtime": 15107.827,
      "train_tokens_per_second": 8514.637
    },
    {
      "epoch": 0.5292216536493402,
      "grad_norm": 1.1825219931886948,
      "learning_rate": 8.369972178283955e-06,
      "loss": 0.40064449310302735,
      "num_input_tokens_seen": 128719576,
      "step": 7860,
      "train_runtime": 15117.4451,
      "train_tokens_per_second": 8514.638
    },
    {
      "epoch": 0.5295583086453003,
      "grad_norm": 1.0940321916862281,
      "learning_rate": 8.368018427384027e-06,
      "loss": 0.391787052154541,
      "num_input_tokens_seen": 128801496,
      "step": 7865,
      "train_runtime": 15126.9378,
      "train_tokens_per_second": 8514.711
    },
    {
      "epoch": 0.5298949636412604,
      "grad_norm": 1.1835413355219258,
      "learning_rate": 8.366063734628485e-06,
      "loss": 0.4081437110900879,
      "num_input_tokens_seen": 128883416,
      "step": 7870,
      "train_runtime": 15136.5397,
      "train_tokens_per_second": 8514.721
    },
    {
      "epoch": 0.5302316186372206,
      "grad_norm": 0.9996802014263585,
      "learning_rate": 8.364108100563955e-06,
      "loss": 0.42919244766235354,
      "num_input_tokens_seen": 128965336,
      "step": 7875,
      "train_runtime": 15146.1628,
      "train_tokens_per_second": 8514.72
    },
    {
      "epoch": 0.5305682736331807,
      "grad_norm": 1.1484126987769074,
      "learning_rate": 8.362151525737321e-06,
      "loss": 0.4685671806335449,
      "num_input_tokens_seen": 129047256,
      "step": 7880,
      "train_runtime": 15155.6672,
      "train_tokens_per_second": 8514.786
    },
    {
      "epoch": 0.5309049286291408,
      "grad_norm": 1.0358679131040192,
      "learning_rate": 8.360194010695733e-06,
      "loss": 0.4028486728668213,
      "num_input_tokens_seen": 129129176,
      "step": 7885,
      "train_runtime": 15165.2903,
      "train_tokens_per_second": 8514.784
    },
    {
      "epoch": 0.531241583625101,
      "grad_norm": 0.912741404151414,
      "learning_rate": 8.358235555986607e-06,
      "loss": 0.40334763526916506,
      "num_input_tokens_seen": 129211096,
      "step": 7890,
      "train_runtime": 15174.8985,
      "train_tokens_per_second": 8514.791
    },
    {
      "epoch": 0.5315782386210611,
      "grad_norm": 1.2006011177044957,
      "learning_rate": 8.356276162157613e-06,
      "loss": 0.38882741928100584,
      "num_input_tokens_seen": 129293016,
      "step": 7895,
      "train_runtime": 15184.3892,
      "train_tokens_per_second": 8514.864
    },
    {
      "epoch": 0.5319148936170213,
      "grad_norm": 1.0717149846197158,
      "learning_rate": 8.354315829756696e-06,
      "loss": 0.40518712997436523,
      "num_input_tokens_seen": 129374936,
      "step": 7900,
      "train_runtime": 15193.9604,
      "train_tokens_per_second": 8514.892
    },
    {
      "epoch": 0.5322515486129814,
      "grad_norm": 1.0655116678261556,
      "learning_rate": 8.352354559332053e-06,
      "loss": 0.4009148597717285,
      "num_input_tokens_seen": 129456856,
      "step": 7905,
      "train_runtime": 15203.5782,
      "train_tokens_per_second": 8514.894
    },
    {
      "epoch": 0.5325882036089415,
      "grad_norm": 0.9334511920460219,
      "learning_rate": 8.350392351432146e-06,
      "loss": 0.3942744016647339,
      "num_input_tokens_seen": 129538776,
      "step": 7910,
      "train_runtime": 15213.1339,
      "train_tokens_per_second": 8514.93
    },
    {
      "epoch": 0.5329248586049017,
      "grad_norm": 1.1258439812669117,
      "learning_rate": 8.348429206605702e-06,
      "loss": 0.44385499954223634,
      "num_input_tokens_seen": 129620696,
      "step": 7915,
      "train_runtime": 15222.7245,
      "train_tokens_per_second": 8514.947
    },
    {
      "epoch": 0.5332615136008618,
      "grad_norm": 1.3343220681970893,
      "learning_rate": 8.346465125401706e-06,
      "loss": 0.45760436058044435,
      "num_input_tokens_seen": 129702616,
      "step": 7920,
      "train_runtime": 15232.3419,
      "train_tokens_per_second": 8514.949
    },
    {
      "epoch": 0.533598168596822,
      "grad_norm": 1.0356189415285877,
      "learning_rate": 8.344500108369412e-06,
      "loss": 0.37995114326477053,
      "num_input_tokens_seen": 129784536,
      "step": 7925,
      "train_runtime": 15241.8894,
      "train_tokens_per_second": 8514.99
    },
    {
      "epoch": 0.5339348235927821,
      "grad_norm": 1.0732582233897185,
      "learning_rate": 8.342534156058326e-06,
      "loss": 0.41960716247558594,
      "num_input_tokens_seen": 129866456,
      "step": 7930,
      "train_runtime": 15251.4738,
      "train_tokens_per_second": 8515.01
    },
    {
      "epoch": 0.5342714785887422,
      "grad_norm": 1.1016138279542076,
      "learning_rate": 8.34056726901822e-06,
      "loss": 0.43289852142333984,
      "num_input_tokens_seen": 129948376,
      "step": 7935,
      "train_runtime": 15261.1022,
      "train_tokens_per_second": 8515.006
    },
    {
      "epoch": 0.5346081335847024,
      "grad_norm": 1.1068074206313496,
      "learning_rate": 8.338599447799132e-06,
      "loss": 0.392047381401062,
      "num_input_tokens_seen": 130030256,
      "step": 7940,
      "train_runtime": 15271.2788,
      "train_tokens_per_second": 8514.693
    },
    {
      "epoch": 0.5349447885806625,
      "grad_norm": 1.2457357079696774,
      "learning_rate": 8.336630692951353e-06,
      "loss": 0.4123573303222656,
      "num_input_tokens_seen": 130112176,
      "step": 7945,
      "train_runtime": 15280.7864,
      "train_tokens_per_second": 8514.757
    },
    {
      "epoch": 0.5352814435766227,
      "grad_norm": 1.2842677776209142,
      "learning_rate": 8.334661005025441e-06,
      "loss": 0.43415274620056155,
      "num_input_tokens_seen": 130194096,
      "step": 7950,
      "train_runtime": 15290.3978,
      "train_tokens_per_second": 8514.762
    },
    {
      "epoch": 0.5356180985725828,
      "grad_norm": 1.4423745560841592,
      "learning_rate": 8.332690384572213e-06,
      "loss": 0.41657099723815916,
      "num_input_tokens_seen": 130275808,
      "step": 7955,
      "train_runtime": 15300.4388,
      "train_tokens_per_second": 8514.514
    },
    {
      "epoch": 0.5359547535685429,
      "grad_norm": 1.0984299512056714,
      "learning_rate": 8.330718832142746e-06,
      "loss": 0.41459245681762696,
      "num_input_tokens_seen": 130357728,
      "step": 7960,
      "train_runtime": 15309.9581,
      "train_tokens_per_second": 8514.571
    },
    {
      "epoch": 0.5362914085645031,
      "grad_norm": 0.9669071736599849,
      "learning_rate": 8.328746348288379e-06,
      "loss": 0.37427334785461425,
      "num_input_tokens_seen": 130439648,
      "step": 7965,
      "train_runtime": 15319.5356,
      "train_tokens_per_second": 8514.595
    },
    {
      "epoch": 0.5366280635604632,
      "grad_norm": 0.9892503970482152,
      "learning_rate": 8.32677293356071e-06,
      "loss": 0.4380988121032715,
      "num_input_tokens_seen": 130521568,
      "step": 7970,
      "train_runtime": 15329.0841,
      "train_tokens_per_second": 8514.636
    },
    {
      "epoch": 0.5369647185564234,
      "grad_norm": 1.1168836736842425,
      "learning_rate": 8.324798588511598e-06,
      "loss": 0.42977170944213866,
      "num_input_tokens_seen": 130603488,
      "step": 7975,
      "train_runtime": 15338.6317,
      "train_tokens_per_second": 8514.677
    },
    {
      "epoch": 0.5373013735523835,
      "grad_norm": 1.0791828086949395,
      "learning_rate": 8.322823313693162e-06,
      "loss": 0.397434139251709,
      "num_input_tokens_seen": 130685408,
      "step": 7980,
      "train_runtime": 15348.2002,
      "train_tokens_per_second": 8514.706
    },
    {
      "epoch": 0.5376380285483436,
      "grad_norm": 1.007406136457486,
      "learning_rate": 8.320847109657782e-06,
      "loss": 0.3888430118560791,
      "num_input_tokens_seen": 130767328,
      "step": 7985,
      "train_runtime": 15357.8053,
      "train_tokens_per_second": 8514.715
    },
    {
      "epoch": 0.5379746835443038,
      "grad_norm": 1.082052825430211,
      "learning_rate": 8.318869976958099e-06,
      "loss": 0.3849940776824951,
      "num_input_tokens_seen": 130849248,
      "step": 7990,
      "train_runtime": 15367.316,
      "train_tokens_per_second": 8514.776
    },
    {
      "epoch": 0.5383113385402639,
      "grad_norm": 1.0941201171194601,
      "learning_rate": 8.316891916147007e-06,
      "loss": 0.415631103515625,
      "num_input_tokens_seen": 130931168,
      "step": 7995,
      "train_runtime": 15376.8901,
      "train_tokens_per_second": 8514.802
    },
    {
      "epoch": 0.5386479935362241,
      "grad_norm": 1.1879735067484143,
      "learning_rate": 8.314912927777671e-06,
      "loss": 0.38719117641448975,
      "num_input_tokens_seen": 131013088,
      "step": 8000,
      "train_runtime": 15386.4349,
      "train_tokens_per_second": 8514.844
    },
    {
      "epoch": 0.5389846485321842,
      "grad_norm": 1.1603011279009203,
      "learning_rate": 8.312933012403503e-06,
      "loss": 0.3738299608230591,
      "num_input_tokens_seen": 131094704,
      "step": 8005,
      "train_runtime": 15396.4263,
      "train_tokens_per_second": 8514.619
    },
    {
      "epoch": 0.5393213035281443,
      "grad_norm": 1.09370800126259,
      "learning_rate": 8.310952170578184e-06,
      "loss": 0.4297513008117676,
      "num_input_tokens_seen": 131176488,
      "step": 8010,
      "train_runtime": 15405.9236,
      "train_tokens_per_second": 8514.679
    },
    {
      "epoch": 0.5396579585241045,
      "grad_norm": 0.9369478310002465,
      "learning_rate": 8.308970402855647e-06,
      "loss": 0.3963186740875244,
      "num_input_tokens_seen": 131257600,
      "step": 8015,
      "train_runtime": 15415.7222,
      "train_tokens_per_second": 8514.528
    },
    {
      "epoch": 0.5399946135200646,
      "grad_norm": 1.134479725860133,
      "learning_rate": 8.306987709790087e-06,
      "loss": 0.3912180423736572,
      "num_input_tokens_seen": 131339520,
      "step": 8020,
      "train_runtime": 15425.316,
      "train_tokens_per_second": 8514.543
    },
    {
      "epoch": 0.5403312685160248,
      "grad_norm": 1.1266395450417102,
      "learning_rate": 8.305004091935962e-06,
      "loss": 0.4190827369689941,
      "num_input_tokens_seen": 131421440,
      "step": 8025,
      "train_runtime": 15434.8135,
      "train_tokens_per_second": 8514.611
    },
    {
      "epoch": 0.5406679235119849,
      "grad_norm": 1.1529973751031597,
      "learning_rate": 8.303019549847979e-06,
      "loss": 0.40465679168701174,
      "num_input_tokens_seen": 131503360,
      "step": 8030,
      "train_runtime": 15444.4385,
      "train_tokens_per_second": 8514.609
    },
    {
      "epoch": 0.541004578507945,
      "grad_norm": 1.0982987736364158,
      "learning_rate": 8.301034084081114e-06,
      "loss": 0.36406764984130857,
      "num_input_tokens_seen": 131585280,
      "step": 8035,
      "train_runtime": 15454.063,
      "train_tokens_per_second": 8514.607
    },
    {
      "epoch": 0.5413412335039052,
      "grad_norm": 1.218315581324223,
      "learning_rate": 8.299047695190592e-06,
      "loss": 0.4487174987792969,
      "num_input_tokens_seen": 131667200,
      "step": 8040,
      "train_runtime": 15463.6106,
      "train_tokens_per_second": 8514.648
    },
    {
      "epoch": 0.5416778884998653,
      "grad_norm": 1.0684170688859114,
      "learning_rate": 8.297060383731903e-06,
      "loss": 0.4363774299621582,
      "num_input_tokens_seen": 131749120,
      "step": 8045,
      "train_runtime": 15473.1692,
      "train_tokens_per_second": 8514.682
    },
    {
      "epoch": 0.5420145434958255,
      "grad_norm": 1.0694419111160638,
      "learning_rate": 8.29507215026079e-06,
      "loss": 0.48887271881103517,
      "num_input_tokens_seen": 131831040,
      "step": 8050,
      "train_runtime": 15482.7818,
      "train_tokens_per_second": 8514.687
    },
    {
      "epoch": 0.5423511984917856,
      "grad_norm": 1.1739994395454219,
      "learning_rate": 8.293082995333257e-06,
      "loss": 0.38370614051818847,
      "num_input_tokens_seen": 131912960,
      "step": 8055,
      "train_runtime": 15492.3338,
      "train_tokens_per_second": 8514.725
    },
    {
      "epoch": 0.5426878534877457,
      "grad_norm": 1.0747809436120817,
      "learning_rate": 8.291092919505568e-06,
      "loss": 0.37469985485076907,
      "num_input_tokens_seen": 131994880,
      "step": 8060,
      "train_runtime": 15501.9113,
      "train_tokens_per_second": 8514.749
    },
    {
      "epoch": 0.5430245084837059,
      "grad_norm": 1.087282152785749,
      "learning_rate": 8.289101923334237e-06,
      "loss": 0.41078600883483884,
      "num_input_tokens_seen": 132076800,
      "step": 8065,
      "train_runtime": 15511.5098,
      "train_tokens_per_second": 8514.761
    },
    {
      "epoch": 0.543361163479666,
      "grad_norm": 1.0794235384323319,
      "learning_rate": 8.28711000737604e-06,
      "loss": 0.4023710250854492,
      "num_input_tokens_seen": 132158720,
      "step": 8070,
      "train_runtime": 15521.1278,
      "train_tokens_per_second": 8514.763
    },
    {
      "epoch": 0.5436978184756261,
      "grad_norm": 1.1681171234475678,
      "learning_rate": 8.285117172188013e-06,
      "loss": 0.41117172241210936,
      "num_input_tokens_seen": 132240640,
      "step": 8075,
      "train_runtime": 15530.627,
      "train_tokens_per_second": 8514.829
    },
    {
      "epoch": 0.5440344734715863,
      "grad_norm": 1.4301508589590353,
      "learning_rate": 8.283123418327443e-06,
      "loss": 0.441893196105957,
      "num_input_tokens_seen": 132322560,
      "step": 8080,
      "train_runtime": 15540.2613,
      "train_tokens_per_second": 8514.822
    },
    {
      "epoch": 0.5443711284675464,
      "grad_norm": 1.1237617370958826,
      "learning_rate": 8.281128746351878e-06,
      "loss": 0.3751915216445923,
      "num_input_tokens_seen": 132404480,
      "step": 8085,
      "train_runtime": 15549.8673,
      "train_tokens_per_second": 8514.83
    },
    {
      "epoch": 0.5447077834635066,
      "grad_norm": 1.0153107993859913,
      "learning_rate": 8.279133156819123e-06,
      "loss": 0.42142629623413086,
      "num_input_tokens_seen": 132486400,
      "step": 8090,
      "train_runtime": 15559.371,
      "train_tokens_per_second": 8514.894
    },
    {
      "epoch": 0.5450444384594667,
      "grad_norm": 1.0425703749255202,
      "learning_rate": 8.277136650287237e-06,
      "loss": 0.3608701705932617,
      "num_input_tokens_seen": 132568320,
      "step": 8095,
      "train_runtime": 15568.9703,
      "train_tokens_per_second": 8514.906
    },
    {
      "epoch": 0.5453810934554268,
      "grad_norm": 1.202851962020897,
      "learning_rate": 8.275139227314537e-06,
      "loss": 0.38962588310241697,
      "num_input_tokens_seen": 132650240,
      "step": 8100,
      "train_runtime": 15578.5583,
      "train_tokens_per_second": 8514.924
    },
    {
      "epoch": 0.545717748451387,
      "grad_norm": 1.165431153100288,
      "learning_rate": 8.273140888459594e-06,
      "loss": 0.49074134826660154,
      "num_input_tokens_seen": 132732160,
      "step": 8105,
      "train_runtime": 15588.0595,
      "train_tokens_per_second": 8514.989
    },
    {
      "epoch": 0.5460544034473471,
      "grad_norm": 1.0915022551517708,
      "learning_rate": 8.27114163428124e-06,
      "loss": 0.42902326583862305,
      "num_input_tokens_seen": 132814080,
      "step": 8110,
      "train_runtime": 15597.6721,
      "train_tokens_per_second": 8514.994
    },
    {
      "epoch": 0.5463910584433073,
      "grad_norm": 1.3121638899400923,
      "learning_rate": 8.26914146533856e-06,
      "loss": 0.4181682586669922,
      "num_input_tokens_seen": 132896000,
      "step": 8115,
      "train_runtime": 15607.2846,
      "train_tokens_per_second": 8514.998
    },
    {
      "epoch": 0.5467277134392674,
      "grad_norm": 1.010913172848287,
      "learning_rate": 8.267140382190892e-06,
      "loss": 0.4042576789855957,
      "num_input_tokens_seen": 132977920,
      "step": 8120,
      "train_runtime": 15616.8311,
      "train_tokens_per_second": 8515.039
    },
    {
      "epoch": 0.5470643684352275,
      "grad_norm": 0.8908586542642979,
      "learning_rate": 8.265138385397835e-06,
      "loss": 0.38632705211639407,
      "num_input_tokens_seen": 133059840,
      "step": 8125,
      "train_runtime": 15626.3839,
      "train_tokens_per_second": 8515.076
    },
    {
      "epoch": 0.5474010234311877,
      "grad_norm": 1.0372526380637344,
      "learning_rate": 8.263135475519241e-06,
      "loss": 0.37788074016571044,
      "num_input_tokens_seen": 133141760,
      "step": 8130,
      "train_runtime": 15636.0184,
      "train_tokens_per_second": 8515.068
    },
    {
      "epoch": 0.5477376784271478,
      "grad_norm": 1.1930017493634117,
      "learning_rate": 8.261131653115216e-06,
      "loss": 0.45987420082092284,
      "num_input_tokens_seen": 133223680,
      "step": 8135,
      "train_runtime": 15645.5662,
      "train_tokens_per_second": 8515.108
    },
    {
      "epoch": 0.548074333423108,
      "grad_norm": 1.296295611389516,
      "learning_rate": 8.259126918746122e-06,
      "loss": 0.44203839302062986,
      "num_input_tokens_seen": 133305600,
      "step": 8140,
      "train_runtime": 15655.1162,
      "train_tokens_per_second": 8515.146
    },
    {
      "epoch": 0.5484109884190681,
      "grad_norm": 0.9413920383729196,
      "learning_rate": 8.25712127297258e-06,
      "loss": 0.39719343185424805,
      "num_input_tokens_seen": 133387520,
      "step": 8145,
      "train_runtime": 15664.7193,
      "train_tokens_per_second": 8515.155
    },
    {
      "epoch": 0.5487476434150282,
      "grad_norm": 1.2425527147559778,
      "learning_rate": 8.255114716355459e-06,
      "loss": 0.4204086780548096,
      "num_input_tokens_seen": 133469440,
      "step": 8150,
      "train_runtime": 15674.3385,
      "train_tokens_per_second": 8515.156
    },
    {
      "epoch": 0.5490842984109884,
      "grad_norm": 1.147087165863958,
      "learning_rate": 8.253107249455887e-06,
      "loss": 0.4582479476928711,
      "num_input_tokens_seen": 133551360,
      "step": 8155,
      "train_runtime": 15683.8424,
      "train_tokens_per_second": 8515.219
    },
    {
      "epoch": 0.5494209534069485,
      "grad_norm": 1.129391935198579,
      "learning_rate": 8.251098872835247e-06,
      "loss": 0.4315167427062988,
      "num_input_tokens_seen": 133633280,
      "step": 8160,
      "train_runtime": 15693.4747,
      "train_tokens_per_second": 8515.213
    },
    {
      "epoch": 0.5497576084029087,
      "grad_norm": 0.9320857272693351,
      "learning_rate": 8.249089587055175e-06,
      "loss": 0.4003553867340088,
      "num_input_tokens_seen": 133715200,
      "step": 8165,
      "train_runtime": 15703.0956,
      "train_tokens_per_second": 8515.213
    },
    {
      "epoch": 0.5500942633988688,
      "grad_norm": 1.0377867113794772,
      "learning_rate": 8.247079392677558e-06,
      "loss": 0.3834519624710083,
      "num_input_tokens_seen": 133797120,
      "step": 8170,
      "train_runtime": 15712.5915,
      "train_tokens_per_second": 8515.28
    },
    {
      "epoch": 0.5504309183948289,
      "grad_norm": 0.998422792105205,
      "learning_rate": 8.245068290264544e-06,
      "loss": 0.3995727777481079,
      "num_input_tokens_seen": 133879040,
      "step": 8175,
      "train_runtime": 15722.2019,
      "train_tokens_per_second": 8515.286
    },
    {
      "epoch": 0.5507675733907891,
      "grad_norm": 1.1580946570460344,
      "learning_rate": 8.243056280378531e-06,
      "loss": 0.407499885559082,
      "num_input_tokens_seen": 133960960,
      "step": 8180,
      "train_runtime": 15731.8057,
      "train_tokens_per_second": 8515.295
    },
    {
      "epoch": 0.5511042283867492,
      "grad_norm": 1.128168683340213,
      "learning_rate": 8.241043363582171e-06,
      "loss": 0.3831015586853027,
      "num_input_tokens_seen": 134042880,
      "step": 8185,
      "train_runtime": 15741.3054,
      "train_tokens_per_second": 8515.36
    },
    {
      "epoch": 0.5514408833827094,
      "grad_norm": 0.9495002195935207,
      "learning_rate": 8.239029540438369e-06,
      "loss": 0.39702615737915037,
      "num_input_tokens_seen": 134124800,
      "step": 8190,
      "train_runtime": 15750.9322,
      "train_tokens_per_second": 8515.356
    },
    {
      "epoch": 0.5517775383786695,
      "grad_norm": 1.215112842302422,
      "learning_rate": 8.237014811510284e-06,
      "loss": 0.42287359237670896,
      "num_input_tokens_seen": 134206720,
      "step": 8195,
      "train_runtime": 15760.5505,
      "train_tokens_per_second": 8515.357
    },
    {
      "epoch": 0.5521141933746296,
      "grad_norm": 1.136897041061364,
      "learning_rate": 8.234999177361329e-06,
      "loss": 0.39497807025909426,
      "num_input_tokens_seen": 134288640,
      "step": 8200,
      "train_runtime": 15770.0993,
      "train_tokens_per_second": 8515.396
    },
    {
      "epoch": 0.5524508483705898,
      "grad_norm": 1.1064154200172382,
      "learning_rate": 8.232982638555169e-06,
      "loss": 0.4235065460205078,
      "num_input_tokens_seen": 134370560,
      "step": 8205,
      "train_runtime": 15779.6787,
      "train_tokens_per_second": 8515.418
    },
    {
      "epoch": 0.5527875033665499,
      "grad_norm": 0.9042497169932397,
      "learning_rate": 8.230965195655724e-06,
      "loss": 0.4461686134338379,
      "num_input_tokens_seen": 134452480,
      "step": 8210,
      "train_runtime": 15789.3164,
      "train_tokens_per_second": 8515.409
    },
    {
      "epoch": 0.55312415836251,
      "grad_norm": 1.3018571780098658,
      "learning_rate": 8.228946849227167e-06,
      "loss": 0.4377298355102539,
      "num_input_tokens_seen": 134534400,
      "step": 8215,
      "train_runtime": 15798.8574,
      "train_tokens_per_second": 8515.451
    },
    {
      "epoch": 0.5534608133584702,
      "grad_norm": 1.1811034704338477,
      "learning_rate": 8.226927599833916e-06,
      "loss": 0.45536174774169924,
      "num_input_tokens_seen": 134616320,
      "step": 8220,
      "train_runtime": 15808.411,
      "train_tokens_per_second": 8515.487
    },
    {
      "epoch": 0.5537974683544303,
      "grad_norm": 1.0954188556074986,
      "learning_rate": 8.224907448040655e-06,
      "loss": 0.3999060392379761,
      "num_input_tokens_seen": 134698240,
      "step": 8225,
      "train_runtime": 15818.0195,
      "train_tokens_per_second": 8515.493
    },
    {
      "epoch": 0.5541341233503905,
      "grad_norm": 1.063079796337644,
      "learning_rate": 8.222886394412309e-06,
      "loss": 0.41303586959838867,
      "num_input_tokens_seen": 134780160,
      "step": 8230,
      "train_runtime": 15827.5878,
      "train_tokens_per_second": 8515.521
    },
    {
      "epoch": 0.5544707783463506,
      "grad_norm": 1.0655073623036964,
      "learning_rate": 8.220864439514057e-06,
      "loss": 0.397322940826416,
      "num_input_tokens_seen": 134862080,
      "step": 8235,
      "train_runtime": 15837.1677,
      "train_tokens_per_second": 8515.543
    },
    {
      "epoch": 0.5548074333423108,
      "grad_norm": 0.9894027425023626,
      "learning_rate": 8.218841583911338e-06,
      "loss": 0.3787353754043579,
      "num_input_tokens_seen": 134944000,
      "step": 8240,
      "train_runtime": 15846.7949,
      "train_tokens_per_second": 8515.539
    },
    {
      "epoch": 0.5551440883382709,
      "grad_norm": 1.0601564088082225,
      "learning_rate": 8.216817828169834e-06,
      "loss": 0.447930908203125,
      "num_input_tokens_seen": 135025920,
      "step": 8245,
      "train_runtime": 15856.4057,
      "train_tokens_per_second": 8515.544
    },
    {
      "epoch": 0.555480743334231,
      "grad_norm": 1.2588880236796391,
      "learning_rate": 8.214793172855481e-06,
      "loss": 0.3701935768127441,
      "num_input_tokens_seen": 135107840,
      "step": 8250,
      "train_runtime": 15865.9158,
      "train_tokens_per_second": 8515.603
    },
    {
      "epoch": 0.5558173983301912,
      "grad_norm": 0.9490130021044374,
      "learning_rate": 8.21276761853447e-06,
      "loss": 0.3731944799423218,
      "num_input_tokens_seen": 135189760,
      "step": 8255,
      "train_runtime": 15875.5234,
      "train_tokens_per_second": 8515.61
    },
    {
      "epoch": 0.5561540533261513,
      "grad_norm": 1.251321832324545,
      "learning_rate": 8.210741165773238e-06,
      "loss": 0.4381718158721924,
      "num_input_tokens_seen": 135271680,
      "step": 8260,
      "train_runtime": 15885.1492,
      "train_tokens_per_second": 8515.607
    },
    {
      "epoch": 0.5564907083221114,
      "grad_norm": 0.9312390697015035,
      "learning_rate": 8.208713815138476e-06,
      "loss": 0.36029701232910155,
      "num_input_tokens_seen": 135353600,
      "step": 8265,
      "train_runtime": 15894.6592,
      "train_tokens_per_second": 8515.665
    },
    {
      "epoch": 0.5568273633180716,
      "grad_norm": 1.028868278023732,
      "learning_rate": 8.20668556719713e-06,
      "loss": 0.4155407428741455,
      "num_input_tokens_seen": 135435520,
      "step": 8270,
      "train_runtime": 15904.2646,
      "train_tokens_per_second": 8515.673
    },
    {
      "epoch": 0.5571640183140317,
      "grad_norm": 1.1147861517781337,
      "learning_rate": 8.204656422516388e-06,
      "loss": 0.38791751861572266,
      "num_input_tokens_seen": 135517440,
      "step": 8275,
      "train_runtime": 15913.8872,
      "train_tokens_per_second": 8515.672
    },
    {
      "epoch": 0.5575006733099919,
      "grad_norm": 0.9851998521723224,
      "learning_rate": 8.2026263816637e-06,
      "loss": 0.38785037994384763,
      "num_input_tokens_seen": 135599360,
      "step": 8280,
      "train_runtime": 15923.3915,
      "train_tokens_per_second": 8515.734
    },
    {
      "epoch": 0.557837328305952,
      "grad_norm": 1.0481688906657467,
      "learning_rate": 8.200595445206753e-06,
      "loss": 0.394207239151001,
      "num_input_tokens_seen": 135681280,
      "step": 8285,
      "train_runtime": 15932.9599,
      "train_tokens_per_second": 8515.761
    },
    {
      "epoch": 0.5581739833019121,
      "grad_norm": 1.0208972317513216,
      "learning_rate": 8.198563613713497e-06,
      "loss": 0.3883861780166626,
      "num_input_tokens_seen": 135763200,
      "step": 8290,
      "train_runtime": 15942.5793,
      "train_tokens_per_second": 8515.761
    },
    {
      "epoch": 0.5585106382978723,
      "grad_norm": 1.2473166851438382,
      "learning_rate": 8.196530887752127e-06,
      "loss": 0.39716956615447996,
      "num_input_tokens_seen": 135845120,
      "step": 8295,
      "train_runtime": 15952.1198,
      "train_tokens_per_second": 8515.804
    },
    {
      "epoch": 0.5588472932938324,
      "grad_norm": 1.0952235866043176,
      "learning_rate": 8.194497267891086e-06,
      "loss": 0.40042991638183595,
      "num_input_tokens_seen": 135927040,
      "step": 8300,
      "train_runtime": 15961.6803,
      "train_tokens_per_second": 8515.835
    },
    {
      "epoch": 0.5591839482897926,
      "grad_norm": 1.0947836991189972,
      "learning_rate": 8.19246275469907e-06,
      "loss": 0.40491552352905275,
      "num_input_tokens_seen": 136008960,
      "step": 8305,
      "train_runtime": 15971.2091,
      "train_tokens_per_second": 8515.884
    },
    {
      "epoch": 0.5595206032857528,
      "grad_norm": 1.1216655334904622,
      "learning_rate": 8.190427348745027e-06,
      "loss": 0.4139406204223633,
      "num_input_tokens_seen": 136090880,
      "step": 8310,
      "train_runtime": 15980.7606,
      "train_tokens_per_second": 8515.92
    },
    {
      "epoch": 0.559857258281713,
      "grad_norm": 1.0068119576639218,
      "learning_rate": 8.188391050598146e-06,
      "loss": 0.38076441287994384,
      "num_input_tokens_seen": 136172800,
      "step": 8315,
      "train_runtime": 15990.3213,
      "train_tokens_per_second": 8515.951
    },
    {
      "epoch": 0.5601939132776731,
      "grad_norm": 1.1053441836255748,
      "learning_rate": 8.186353860827878e-06,
      "loss": 0.4128708839416504,
      "num_input_tokens_seen": 136254720,
      "step": 8320,
      "train_runtime": 15999.9076,
      "train_tokens_per_second": 8515.969
    },
    {
      "epoch": 0.5605305682736332,
      "grad_norm": 0.9637674457152479,
      "learning_rate": 8.184315780003911e-06,
      "loss": 0.4170865058898926,
      "num_input_tokens_seen": 136336640,
      "step": 8325,
      "train_runtime": 16009.503,
      "train_tokens_per_second": 8515.982
    },
    {
      "epoch": 0.5608672232695934,
      "grad_norm": 1.0354654125007865,
      "learning_rate": 8.182276808696193e-06,
      "loss": 0.4565864562988281,
      "num_input_tokens_seen": 136418152,
      "step": 8330,
      "train_runtime": 16019.3618,
      "train_tokens_per_second": 8515.829
    },
    {
      "epoch": 0.5612038782655535,
      "grad_norm": 1.0225488422945266,
      "learning_rate": 8.18023694747491e-06,
      "loss": 0.4021279335021973,
      "num_input_tokens_seen": 136500072,
      "step": 8335,
      "train_runtime": 16028.9707,
      "train_tokens_per_second": 8515.835
    },
    {
      "epoch": 0.5615405332615137,
      "grad_norm": 1.093760478812342,
      "learning_rate": 8.178196196910508e-06,
      "loss": 0.4026789665222168,
      "num_input_tokens_seen": 136581992,
      "step": 8340,
      "train_runtime": 16038.5869,
      "train_tokens_per_second": 8515.837
    },
    {
      "epoch": 0.5618771882574738,
      "grad_norm": 1.0122911387763887,
      "learning_rate": 8.176154557573672e-06,
      "loss": 0.42737445831298826,
      "num_input_tokens_seen": 136663912,
      "step": 8345,
      "train_runtime": 16048.0818,
      "train_tokens_per_second": 8515.903
    },
    {
      "epoch": 0.5622138432534339,
      "grad_norm": 1.0331605060435294,
      "learning_rate": 8.174112030035344e-06,
      "loss": 0.40136094093322755,
      "num_input_tokens_seen": 136745832,
      "step": 8350,
      "train_runtime": 16057.6923,
      "train_tokens_per_second": 8515.908
    },
    {
      "epoch": 0.5625504982493941,
      "grad_norm": 1.002759520312792,
      "learning_rate": 8.172068614866707e-06,
      "loss": 0.3645493984222412,
      "num_input_tokens_seen": 136827432,
      "step": 8355,
      "train_runtime": 16067.6962,
      "train_tokens_per_second": 8515.685
    },
    {
      "epoch": 0.5628871532453542,
      "grad_norm": 1.0833871580549466,
      "learning_rate": 8.170024312639197e-06,
      "loss": 0.4153899192810059,
      "num_input_tokens_seen": 136909352,
      "step": 8360,
      "train_runtime": 16077.2473,
      "train_tokens_per_second": 8515.721
    },
    {
      "epoch": 0.5632238082413143,
      "grad_norm": 1.0192862467305925,
      "learning_rate": 8.167979123924496e-06,
      "loss": 0.4209133148193359,
      "num_input_tokens_seen": 136991272,
      "step": 8365,
      "train_runtime": 16087.1285,
      "train_tokens_per_second": 8515.583
    },
    {
      "epoch": 0.5635604632372745,
      "grad_norm": 1.2858711411046493,
      "learning_rate": 8.165933049294537e-06,
      "loss": 0.4077274322509766,
      "num_input_tokens_seen": 137072024,
      "step": 8370,
      "train_runtime": 16096.885,
      "train_tokens_per_second": 8515.438
    },
    {
      "epoch": 0.5638971182332346,
      "grad_norm": 1.1460087242369617,
      "learning_rate": 8.163886089321493e-06,
      "loss": 0.3919187068939209,
      "num_input_tokens_seen": 137153944,
      "step": 8375,
      "train_runtime": 16106.3819,
      "train_tokens_per_second": 8515.503
    },
    {
      "epoch": 0.5642337732291948,
      "grad_norm": 0.8672555647354055,
      "learning_rate": 8.161838244577795e-06,
      "loss": 0.34131522178649903,
      "num_input_tokens_seen": 137235864,
      "step": 8380,
      "train_runtime": 16115.9481,
      "train_tokens_per_second": 8515.532
    },
    {
      "epoch": 0.5645704282251549,
      "grad_norm": 0.9377123412922483,
      "learning_rate": 8.159789515636114e-06,
      "loss": 0.34625658988952634,
      "num_input_tokens_seen": 137317784,
      "step": 8385,
      "train_runtime": 16125.5526,
      "train_tokens_per_second": 8515.54
    },
    {
      "epoch": 0.564907083221115,
      "grad_norm": 0.9405778679261492,
      "learning_rate": 8.15773990306937e-06,
      "loss": 0.4015627384185791,
      "num_input_tokens_seen": 137399704,
      "step": 8390,
      "train_runtime": 16135.1557,
      "train_tokens_per_second": 8515.549
    },
    {
      "epoch": 0.5652437382170752,
      "grad_norm": 1.2537078342565309,
      "learning_rate": 8.155689407450732e-06,
      "loss": 0.4399870872497559,
      "num_input_tokens_seen": 137481624,
      "step": 8395,
      "train_runtime": 16144.9293,
      "train_tokens_per_second": 8515.468
    },
    {
      "epoch": 0.5655803932130353,
      "grad_norm": 1.0991754506388083,
      "learning_rate": 8.153638029353615e-06,
      "loss": 0.38633031845092775,
      "num_input_tokens_seen": 137563544,
      "step": 8400,
      "train_runtime": 16155.0863,
      "train_tokens_per_second": 8515.185
    },
    {
      "epoch": 0.5659170482089955,
      "grad_norm": 1.216148387761579,
      "learning_rate": 8.151585769351678e-06,
      "loss": 0.439790678024292,
      "num_input_tokens_seen": 137645464,
      "step": 8405,
      "train_runtime": 16164.7012,
      "train_tokens_per_second": 8515.188
    },
    {
      "epoch": 0.5662537032049556,
      "grad_norm": 1.0347596494425426,
      "learning_rate": 8.149532628018833e-06,
      "loss": 0.4122410774230957,
      "num_input_tokens_seen": 137727384,
      "step": 8410,
      "train_runtime": 16174.2142,
      "train_tokens_per_second": 8515.244
    },
    {
      "epoch": 0.5665903582009157,
      "grad_norm": 1.0617093879670934,
      "learning_rate": 8.147478605929228e-06,
      "loss": 0.3843526840209961,
      "num_input_tokens_seen": 137809304,
      "step": 8415,
      "train_runtime": 16183.7906,
      "train_tokens_per_second": 8515.267
    },
    {
      "epoch": 0.5669270131968759,
      "grad_norm": 0.8339780925419288,
      "learning_rate": 8.145423703657269e-06,
      "loss": 0.3812393665313721,
      "num_input_tokens_seen": 137891224,
      "step": 8420,
      "train_runtime": 16193.3383,
      "train_tokens_per_second": 8515.306
    },
    {
      "epoch": 0.567263668192836,
      "grad_norm": 1.1570116059363915,
      "learning_rate": 8.143367921777601e-06,
      "loss": 0.4203656196594238,
      "num_input_tokens_seen": 137973144,
      "step": 8425,
      "train_runtime": 16202.8851,
      "train_tokens_per_second": 8515.344
    },
    {
      "epoch": 0.5676003231887962,
      "grad_norm": 1.188685145973767,
      "learning_rate": 8.141311260865117e-06,
      "loss": 0.4478424549102783,
      "num_input_tokens_seen": 138055064,
      "step": 8430,
      "train_runtime": 16212.4328,
      "train_tokens_per_second": 8515.382
    },
    {
      "epoch": 0.5679369781847563,
      "grad_norm": 1.1609766761831504,
      "learning_rate": 8.139253721494956e-06,
      "loss": 0.4388432502746582,
      "num_input_tokens_seen": 138136984,
      "step": 8435,
      "train_runtime": 16222.0181,
      "train_tokens_per_second": 8515.401
    },
    {
      "epoch": 0.5682736331807164,
      "grad_norm": 0.9635940416776257,
      "learning_rate": 8.137195304242502e-06,
      "loss": 0.3791311740875244,
      "num_input_tokens_seen": 138218904,
      "step": 8440,
      "train_runtime": 16231.559,
      "train_tokens_per_second": 8515.442
    },
    {
      "epoch": 0.5686102881766766,
      "grad_norm": 1.0454592869554213,
      "learning_rate": 8.135136009683387e-06,
      "loss": 0.4135592460632324,
      "num_input_tokens_seen": 138300824,
      "step": 8445,
      "train_runtime": 16241.1288,
      "train_tokens_per_second": 8515.469
    },
    {
      "epoch": 0.5689469431726367,
      "grad_norm": 0.9649908772521978,
      "learning_rate": 8.13307583839348e-06,
      "loss": 0.3912086009979248,
      "num_input_tokens_seen": 138382744,
      "step": 8450,
      "train_runtime": 16250.7339,
      "train_tokens_per_second": 8515.477
    },
    {
      "epoch": 0.5692835981685969,
      "grad_norm": 1.2089434182110042,
      "learning_rate": 8.131014790948908e-06,
      "loss": 0.41001319885253906,
      "num_input_tokens_seen": 138464664,
      "step": 8455,
      "train_runtime": 16260.3488,
      "train_tokens_per_second": 8515.479
    },
    {
      "epoch": 0.569620253164557,
      "grad_norm": 1.035059389896462,
      "learning_rate": 8.128952867926035e-06,
      "loss": 0.4166522979736328,
      "num_input_tokens_seen": 138546584,
      "step": 8460,
      "train_runtime": 16269.8429,
      "train_tokens_per_second": 8515.545
    },
    {
      "epoch": 0.5699569081605171,
      "grad_norm": 1.0769384821166454,
      "learning_rate": 8.126890069901468e-06,
      "loss": 0.3649028539657593,
      "num_input_tokens_seen": 138628056,
      "step": 8465,
      "train_runtime": 16279.7845,
      "train_tokens_per_second": 8515.35
    },
    {
      "epoch": 0.5702935631564773,
      "grad_norm": 0.9445676556681547,
      "learning_rate": 8.124826397452064e-06,
      "loss": 0.36431715488433836,
      "num_input_tokens_seen": 138709976,
      "step": 8470,
      "train_runtime": 16289.384,
      "train_tokens_per_second": 8515.36
    },
    {
      "epoch": 0.5706302181524374,
      "grad_norm": 1.0434210861291087,
      "learning_rate": 8.122761851154923e-06,
      "loss": 0.4426891326904297,
      "num_input_tokens_seen": 138791896,
      "step": 8475,
      "train_runtime": 16298.8898,
      "train_tokens_per_second": 8515.42
    },
    {
      "epoch": 0.5709668731483976,
      "grad_norm": 1.2080629431989753,
      "learning_rate": 8.120696431587387e-06,
      "loss": 0.43604393005371095,
      "num_input_tokens_seen": 138873816,
      "step": 8480,
      "train_runtime": 16308.4808,
      "train_tokens_per_second": 8515.435
    },
    {
      "epoch": 0.5713035281443577,
      "grad_norm": 0.9921951513823293,
      "learning_rate": 8.118630139327046e-06,
      "loss": 0.3636801719665527,
      "num_input_tokens_seen": 138955736,
      "step": 8485,
      "train_runtime": 16318.0818,
      "train_tokens_per_second": 8515.445
    },
    {
      "epoch": 0.5716401831403178,
      "grad_norm": 1.001180983458411,
      "learning_rate": 8.116562974951731e-06,
      "loss": 0.4192249298095703,
      "num_input_tokens_seen": 139037656,
      "step": 8490,
      "train_runtime": 16327.5813,
      "train_tokens_per_second": 8515.508
    },
    {
      "epoch": 0.571976838136278,
      "grad_norm": 1.0525338591128508,
      "learning_rate": 8.11449493903952e-06,
      "loss": 0.42520461082458494,
      "num_input_tokens_seen": 139119576,
      "step": 8495,
      "train_runtime": 16337.1878,
      "train_tokens_per_second": 8515.516
    },
    {
      "epoch": 0.5723134931322381,
      "grad_norm": 1.0207354424198112,
      "learning_rate": 8.112426032168727e-06,
      "loss": 0.42925405502319336,
      "num_input_tokens_seen": 139201496,
      "step": 8500,
      "train_runtime": 16346.7863,
      "train_tokens_per_second": 8515.527
    },
    {
      "epoch": 0.5726501481281983,
      "grad_norm": 0.9988596087245816,
      "learning_rate": 8.11035625491792e-06,
      "loss": 0.3458098888397217,
      "num_input_tokens_seen": 139283416,
      "step": 8505,
      "train_runtime": 16356.2825,
      "train_tokens_per_second": 8515.591
    },
    {
      "epoch": 0.5729868031241584,
      "grad_norm": 1.1021616592664691,
      "learning_rate": 8.108285607865907e-06,
      "loss": 0.3985898017883301,
      "num_input_tokens_seen": 139365336,
      "step": 8510,
      "train_runtime": 16365.9054,
      "train_tokens_per_second": 8515.59
    },
    {
      "epoch": 0.5733234581201185,
      "grad_norm": 1.1712494267431734,
      "learning_rate": 8.106214091591731e-06,
      "loss": 0.3703757286071777,
      "num_input_tokens_seen": 139447256,
      "step": 8515,
      "train_runtime": 16375.5113,
      "train_tokens_per_second": 8515.597
    },
    {
      "epoch": 0.5736601131160787,
      "grad_norm": 1.093000837421383,
      "learning_rate": 8.104141706674693e-06,
      "loss": 0.4086282730102539,
      "num_input_tokens_seen": 139529176,
      "step": 8520,
      "train_runtime": 16385.0589,
      "train_tokens_per_second": 8515.635
    },
    {
      "epoch": 0.5739967681120388,
      "grad_norm": 1.0450591929601252,
      "learning_rate": 8.102068453694324e-06,
      "loss": 0.4082907199859619,
      "num_input_tokens_seen": 139611096,
      "step": 8525,
      "train_runtime": 16394.6144,
      "train_tokens_per_second": 8515.668
    },
    {
      "epoch": 0.574333423107999,
      "grad_norm": 1.107825749936303,
      "learning_rate": 8.099994333230404e-06,
      "loss": 0.43042984008789065,
      "num_input_tokens_seen": 139692160,
      "step": 8530,
      "train_runtime": 16404.2368,
      "train_tokens_per_second": 8515.615
    },
    {
      "epoch": 0.5746700781039591,
      "grad_norm": 1.0740003004339804,
      "learning_rate": 8.097919345862952e-06,
      "loss": 0.431429386138916,
      "num_input_tokens_seen": 139774080,
      "step": 8535,
      "train_runtime": 16413.7928,
      "train_tokens_per_second": 8515.648
    },
    {
      "epoch": 0.5750067330999192,
      "grad_norm": 1.0071801574721193,
      "learning_rate": 8.095843492172235e-06,
      "loss": 0.457363224029541,
      "num_input_tokens_seen": 139856000,
      "step": 8540,
      "train_runtime": 16423.3624,
      "train_tokens_per_second": 8515.674
    },
    {
      "epoch": 0.5753433880958794,
      "grad_norm": 1.1858038524580972,
      "learning_rate": 8.093766772738758e-06,
      "loss": 0.35404119491577146,
      "num_input_tokens_seen": 139937920,
      "step": 8545,
      "train_runtime": 16432.9508,
      "train_tokens_per_second": 8515.69
    },
    {
      "epoch": 0.5756800430918395,
      "grad_norm": 1.1218353978462647,
      "learning_rate": 8.091689188143267e-06,
      "loss": 0.43300938606262207,
      "num_input_tokens_seen": 140019840,
      "step": 8550,
      "train_runtime": 16442.5534,
      "train_tokens_per_second": 8515.699
    },
    {
      "epoch": 0.5760166980877997,
      "grad_norm": 1.1819882681438072,
      "learning_rate": 8.089610738966754e-06,
      "loss": 0.40578222274780273,
      "num_input_tokens_seen": 140101760,
      "step": 8555,
      "train_runtime": 16452.0459,
      "train_tokens_per_second": 8515.765
    },
    {
      "epoch": 0.5763533530837598,
      "grad_norm": 1.0832353614909445,
      "learning_rate": 8.08753142579045e-06,
      "loss": 0.44884405136108396,
      "num_input_tokens_seen": 140183680,
      "step": 8560,
      "train_runtime": 16461.6743,
      "train_tokens_per_second": 8515.761
    },
    {
      "epoch": 0.5766900080797199,
      "grad_norm": 1.2386040112502172,
      "learning_rate": 8.085451249195828e-06,
      "loss": 0.4220566749572754,
      "num_input_tokens_seen": 140265600,
      "step": 8565,
      "train_runtime": 16471.2899,
      "train_tokens_per_second": 8515.763
    },
    {
      "epoch": 0.5770266630756801,
      "grad_norm": 1.0554858397791547,
      "learning_rate": 8.083370209764603e-06,
      "loss": 0.41909942626953123,
      "num_input_tokens_seen": 140347520,
      "step": 8570,
      "train_runtime": 16480.7942,
      "train_tokens_per_second": 8515.823
    },
    {
      "epoch": 0.5773633180716402,
      "grad_norm": 1.1104800182050676,
      "learning_rate": 8.081288308078731e-06,
      "loss": 0.42294774055480955,
      "num_input_tokens_seen": 140429440,
      "step": 8575,
      "train_runtime": 16490.3143,
      "train_tokens_per_second": 8515.874
    },
    {
      "epoch": 0.5776999730676003,
      "grad_norm": 1.1051530000632381,
      "learning_rate": 8.07920554472041e-06,
      "loss": 0.4210039138793945,
      "num_input_tokens_seen": 140511360,
      "step": 8580,
      "train_runtime": 16499.926,
      "train_tokens_per_second": 8515.878
    },
    {
      "epoch": 0.5780366280635605,
      "grad_norm": 1.0191434675396989,
      "learning_rate": 8.077121920272075e-06,
      "loss": 0.4264688014984131,
      "num_input_tokens_seen": 140593280,
      "step": 8585,
      "train_runtime": 16509.422,
      "train_tokens_per_second": 8515.942
    },
    {
      "epoch": 0.5783732830595206,
      "grad_norm": 1.0362255714015647,
      "learning_rate": 8.07503743531641e-06,
      "loss": 0.39792191982269287,
      "num_input_tokens_seen": 140675200,
      "step": 8590,
      "train_runtime": 16519.0268,
      "train_tokens_per_second": 8515.95
    },
    {
      "epoch": 0.5787099380554808,
      "grad_norm": 0.9901996047435158,
      "learning_rate": 8.072952090436328e-06,
      "loss": 0.39703428745269775,
      "num_input_tokens_seen": 140757120,
      "step": 8595,
      "train_runtime": 16528.6333,
      "train_tokens_per_second": 8515.956
    },
    {
      "epoch": 0.5790465930514409,
      "grad_norm": 1.1658109670741534,
      "learning_rate": 8.070865886214996e-06,
      "loss": 0.3993873596191406,
      "num_input_tokens_seen": 140839040,
      "step": 8600,
      "train_runtime": 16538.1346,
      "train_tokens_per_second": 8516.017
    },
    {
      "epoch": 0.579383248047401,
      "grad_norm": 1.5803958388213797,
      "learning_rate": 8.06877882323581e-06,
      "loss": 0.4232943534851074,
      "num_input_tokens_seen": 140920456,
      "step": 8605,
      "train_runtime": 16548.1051,
      "train_tokens_per_second": 8515.806
    },
    {
      "epoch": 0.5797199030433612,
      "grad_norm": 1.0122686848330142,
      "learning_rate": 8.066690902082413e-06,
      "loss": 0.37360339164733886,
      "num_input_tokens_seen": 141002376,
      "step": 8610,
      "train_runtime": 16557.7357,
      "train_tokens_per_second": 8515.801
    },
    {
      "epoch": 0.5800565580393213,
      "grad_norm": 1.1620112001563005,
      "learning_rate": 8.064602123338683e-06,
      "loss": 0.47429394721984863,
      "num_input_tokens_seen": 141084296,
      "step": 8615,
      "train_runtime": 16567.2822,
      "train_tokens_per_second": 8515.838
    },
    {
      "epoch": 0.5803932130352815,
      "grad_norm": 1.0034119537333923,
      "learning_rate": 8.06251248758874e-06,
      "loss": 0.3860313892364502,
      "num_input_tokens_seen": 141166216,
      "step": 8620,
      "train_runtime": 16576.8436,
      "train_tokens_per_second": 8515.868
    },
    {
      "epoch": 0.5807298680312416,
      "grad_norm": 1.1757793870916455,
      "learning_rate": 8.060421995416949e-06,
      "loss": 0.45552716255187986,
      "num_input_tokens_seen": 141248136,
      "step": 8625,
      "train_runtime": 16586.4488,
      "train_tokens_per_second": 8515.876
    },
    {
      "epoch": 0.5810665230272017,
      "grad_norm": 1.0256521162660452,
      "learning_rate": 8.058330647407904e-06,
      "loss": 0.41297268867492676,
      "num_input_tokens_seen": 141330056,
      "step": 8630,
      "train_runtime": 16595.9959,
      "train_tokens_per_second": 8515.913
    },
    {
      "epoch": 0.5814031780231619,
      "grad_norm": 1.2880608266686615,
      "learning_rate": 8.056238444146445e-06,
      "loss": 0.4045238971710205,
      "num_input_tokens_seen": 141411976,
      "step": 8635,
      "train_runtime": 16605.5613,
      "train_tokens_per_second": 8515.941
    },
    {
      "epoch": 0.581739833019122,
      "grad_norm": 1.2046836027747234,
      "learning_rate": 8.054145386217652e-06,
      "loss": 0.38072407245635986,
      "num_input_tokens_seen": 141493896,
      "step": 8640,
      "train_runtime": 16615.1774,
      "train_tokens_per_second": 8515.943
    },
    {
      "epoch": 0.5820764880150822,
      "grad_norm": 1.1102974135196302,
      "learning_rate": 8.05205147420684e-06,
      "loss": 0.39758896827697754,
      "num_input_tokens_seen": 141575816,
      "step": 8645,
      "train_runtime": 16624.7163,
      "train_tokens_per_second": 8515.984
    },
    {
      "epoch": 0.5824131430110423,
      "grad_norm": 1.2483932835768232,
      "learning_rate": 8.049956708699567e-06,
      "loss": 0.4569540500640869,
      "num_input_tokens_seen": 141657736,
      "step": 8650,
      "train_runtime": 16634.2846,
      "train_tokens_per_second": 8516.01
    },
    {
      "epoch": 0.5827497980070024,
      "grad_norm": 1.2277607863094366,
      "learning_rate": 8.047861090281623e-06,
      "loss": 0.42917470932006835,
      "num_input_tokens_seen": 141739656,
      "step": 8655,
      "train_runtime": 16643.9087,
      "train_tokens_per_second": 8516.008
    },
    {
      "epoch": 0.5830864530029626,
      "grad_norm": 0.9789017895602151,
      "learning_rate": 8.045764619539047e-06,
      "loss": 0.3731838703155518,
      "num_input_tokens_seen": 141821576,
      "step": 8660,
      "train_runtime": 16653.4508,
      "train_tokens_per_second": 8516.047
    },
    {
      "epoch": 0.5834231079989227,
      "grad_norm": 0.9938106148276316,
      "learning_rate": 8.043667297058106e-06,
      "loss": 0.37630701065063477,
      "num_input_tokens_seen": 141903152,
      "step": 8665,
      "train_runtime": 16663.4289,
      "train_tokens_per_second": 8515.843
    },
    {
      "epoch": 0.5837597629948829,
      "grad_norm": 1.036323871116762,
      "learning_rate": 8.04156912342531e-06,
      "loss": 0.3958569049835205,
      "num_input_tokens_seen": 141985072,
      "step": 8670,
      "train_runtime": 16673.0301,
      "train_tokens_per_second": 8515.853
    },
    {
      "epoch": 0.584096417990843,
      "grad_norm": 1.1458800093797599,
      "learning_rate": 8.03947009922741e-06,
      "loss": 0.38584723472595217,
      "num_input_tokens_seen": 142066992,
      "step": 8675,
      "train_runtime": 16682.6386,
      "train_tokens_per_second": 8515.859
    },
    {
      "epoch": 0.5844330729868031,
      "grad_norm": 1.158501181976702,
      "learning_rate": 8.037370225051387e-06,
      "loss": 0.44844517707824705,
      "num_input_tokens_seen": 142148912,
      "step": 8680,
      "train_runtime": 16692.1412,
      "train_tokens_per_second": 8515.918
    },
    {
      "epoch": 0.5847697279827633,
      "grad_norm": 0.9600191700789568,
      "learning_rate": 8.035269501484467e-06,
      "loss": 0.4162872314453125,
      "num_input_tokens_seen": 142230832,
      "step": 8685,
      "train_runtime": 16701.7342,
      "train_tokens_per_second": 8515.932
    },
    {
      "epoch": 0.5851063829787234,
      "grad_norm": 1.274696459544308,
      "learning_rate": 8.03316792911411e-06,
      "loss": 0.42826480865478517,
      "num_input_tokens_seen": 142312752,
      "step": 8690,
      "train_runtime": 16711.3564,
      "train_tokens_per_second": 8515.931
    },
    {
      "epoch": 0.5854430379746836,
      "grad_norm": 0.9915725663577176,
      "learning_rate": 8.031065508528011e-06,
      "loss": 0.4237533092498779,
      "num_input_tokens_seen": 142394672,
      "step": 8695,
      "train_runtime": 16720.8563,
      "train_tokens_per_second": 8515.992
    },
    {
      "epoch": 0.5857796929706437,
      "grad_norm": 1.1667693029616208,
      "learning_rate": 8.028962240314111e-06,
      "loss": 0.3992583751678467,
      "num_input_tokens_seen": 142476592,
      "step": 8700,
      "train_runtime": 16730.4879,
      "train_tokens_per_second": 8515.985
    },
    {
      "epoch": 0.5861163479666038,
      "grad_norm": 1.025836964791227,
      "learning_rate": 8.026858125060577e-06,
      "loss": 0.4569999217987061,
      "num_input_tokens_seen": 142558512,
      "step": 8705,
      "train_runtime": 16740.1111,
      "train_tokens_per_second": 8515.984
    },
    {
      "epoch": 0.586453002962564,
      "grad_norm": 1.0371677893707931,
      "learning_rate": 8.024753163355821e-06,
      "loss": 0.4456187725067139,
      "num_input_tokens_seen": 142640432,
      "step": 8710,
      "train_runtime": 16749.6536,
      "train_tokens_per_second": 8516.023
    },
    {
      "epoch": 0.5867896579585241,
      "grad_norm": 0.9378559991517206,
      "learning_rate": 8.022647355788489e-06,
      "loss": 0.39503302574157717,
      "num_input_tokens_seen": 142722352,
      "step": 8715,
      "train_runtime": 16759.2186,
      "train_tokens_per_second": 8516.05
    },
    {
      "epoch": 0.5871263129544843,
      "grad_norm": 1.163375931648933,
      "learning_rate": 8.02054070294746e-06,
      "loss": 0.4853264808654785,
      "num_input_tokens_seen": 142804272,
      "step": 8720,
      "train_runtime": 16768.8449,
      "train_tokens_per_second": 8516.047
    },
    {
      "epoch": 0.5874629679504444,
      "grad_norm": 1.095509933219262,
      "learning_rate": 8.018433205421859e-06,
      "loss": 0.4156604766845703,
      "num_input_tokens_seen": 142886192,
      "step": 8725,
      "train_runtime": 16778.4028,
      "train_tokens_per_second": 8516.078
    },
    {
      "epoch": 0.5877996229464045,
      "grad_norm": 1.3020357628735537,
      "learning_rate": 8.016324863801036e-06,
      "loss": 0.4378086566925049,
      "num_input_tokens_seen": 142968112,
      "step": 8730,
      "train_runtime": 16787.9655,
      "train_tokens_per_second": 8516.107
    },
    {
      "epoch": 0.5881362779423647,
      "grad_norm": 1.078432618842846,
      "learning_rate": 8.014215678674581e-06,
      "loss": 0.44609713554382324,
      "num_input_tokens_seen": 143050032,
      "step": 8735,
      "train_runtime": 16797.5715,
      "train_tokens_per_second": 8516.114
    },
    {
      "epoch": 0.5884729329383248,
      "grad_norm": 1.0370888103589315,
      "learning_rate": 8.012105650632325e-06,
      "loss": 0.423793888092041,
      "num_input_tokens_seen": 143131952,
      "step": 8740,
      "train_runtime": 16807.1644,
      "train_tokens_per_second": 8516.127
    },
    {
      "epoch": 0.588809587934285,
      "grad_norm": 1.0256358159518624,
      "learning_rate": 8.009994780264328e-06,
      "loss": 0.44912014007568357,
      "num_input_tokens_seen": 143213520,
      "step": 8745,
      "train_runtime": 16817.0839,
      "train_tokens_per_second": 8515.954
    },
    {
      "epoch": 0.5891462429302451,
      "grad_norm": 1.0969824086782243,
      "learning_rate": 8.00788306816089e-06,
      "loss": 0.410349178314209,
      "num_input_tokens_seen": 143295440,
      "step": 8750,
      "train_runtime": 16826.6846,
      "train_tokens_per_second": 8515.964
    },
    {
      "epoch": 0.5894828979262052,
      "grad_norm": 0.9856419279846029,
      "learning_rate": 8.005770514912543e-06,
      "loss": 0.3697677135467529,
      "num_input_tokens_seen": 143377360,
      "step": 8755,
      "train_runtime": 16836.3125,
      "train_tokens_per_second": 8515.96
    },
    {
      "epoch": 0.5898195529221654,
      "grad_norm": 0.9360986976639557,
      "learning_rate": 8.003657121110055e-06,
      "loss": 0.385546875,
      "num_input_tokens_seen": 143459280,
      "step": 8760,
      "train_runtime": 16845.8155,
      "train_tokens_per_second": 8516.019
    },
    {
      "epoch": 0.5901562079181255,
      "grad_norm": 1.1284432762507783,
      "learning_rate": 8.001542887344432e-06,
      "loss": 0.3685739040374756,
      "num_input_tokens_seen": 143541200,
      "step": 8765,
      "train_runtime": 16855.4275,
      "train_tokens_per_second": 8516.022
    },
    {
      "epoch": 0.5904928629140856,
      "grad_norm": 1.06202140118589,
      "learning_rate": 7.999427814206911e-06,
      "loss": 0.4321716785430908,
      "num_input_tokens_seen": 143623120,
      "step": 8770,
      "train_runtime": 16865.0382,
      "train_tokens_per_second": 8516.027
    },
    {
      "epoch": 0.5908295179100458,
      "grad_norm": 1.0940511992821984,
      "learning_rate": 7.997311902288968e-06,
      "loss": 0.3686883211135864,
      "num_input_tokens_seen": 143705040,
      "step": 8775,
      "train_runtime": 16874.5362,
      "train_tokens_per_second": 8516.088
    },
    {
      "epoch": 0.5911661729060059,
      "grad_norm": 1.088494315058972,
      "learning_rate": 7.995195152182307e-06,
      "loss": 0.345081090927124,
      "num_input_tokens_seen": 143786960,
      "step": 8780,
      "train_runtime": 16884.1568,
      "train_tokens_per_second": 8516.088
    },
    {
      "epoch": 0.5915028279019661,
      "grad_norm": 1.0276910108809911,
      "learning_rate": 7.993077564478875e-06,
      "loss": 0.3647735595703125,
      "num_input_tokens_seen": 143868880,
      "step": 8785,
      "train_runtime": 16893.7519,
      "train_tokens_per_second": 8516.1
    },
    {
      "epoch": 0.5918394828979262,
      "grad_norm": 1.147177125159569,
      "learning_rate": 7.990959139770844e-06,
      "loss": 0.4304629325866699,
      "num_input_tokens_seen": 143950800,
      "step": 8790,
      "train_runtime": 16903.2976,
      "train_tokens_per_second": 8516.137
    },
    {
      "epoch": 0.5921761378938863,
      "grad_norm": 1.2226901292474706,
      "learning_rate": 7.988839878650628e-06,
      "loss": 0.39246320724487305,
      "num_input_tokens_seen": 144032720,
      "step": 8795,
      "train_runtime": 16912.8631,
      "train_tokens_per_second": 8516.164
    },
    {
      "epoch": 0.5925127928898465,
      "grad_norm": 1.1972028910689259,
      "learning_rate": 7.986719781710872e-06,
      "loss": 0.4191708564758301,
      "num_input_tokens_seen": 144114640,
      "step": 8800,
      "train_runtime": 16922.4559,
      "train_tokens_per_second": 8516.178
    },
    {
      "epoch": 0.5928494478858066,
      "grad_norm": 1.3410555921335736,
      "learning_rate": 7.984598849544452e-06,
      "loss": 0.3840888500213623,
      "num_input_tokens_seen": 144196560,
      "step": 8805,
      "train_runtime": 16931.9987,
      "train_tokens_per_second": 8516.216
    },
    {
      "epoch": 0.5931861028817668,
      "grad_norm": 1.0682039281784645,
      "learning_rate": 7.98247708274448e-06,
      "loss": 0.41105403900146487,
      "num_input_tokens_seen": 144278480,
      "step": 8810,
      "train_runtime": 16941.5738,
      "train_tokens_per_second": 8516.238
    },
    {
      "epoch": 0.5935227578777269,
      "grad_norm": 1.0713984532002692,
      "learning_rate": 7.980354481904303e-06,
      "loss": 0.43819942474365237,
      "num_input_tokens_seen": 144360400,
      "step": 8815,
      "train_runtime": 16951.1784,
      "train_tokens_per_second": 8516.246
    },
    {
      "epoch": 0.593859412873687,
      "grad_norm": 1.030341191742034,
      "learning_rate": 7.978231047617498e-06,
      "loss": 0.4091555118560791,
      "num_input_tokens_seen": 144442320,
      "step": 8820,
      "train_runtime": 16960.786,
      "train_tokens_per_second": 8516.252
    },
    {
      "epoch": 0.5941960678696472,
      "grad_norm": 1.0722964082921886,
      "learning_rate": 7.976106780477878e-06,
      "loss": 0.4092889308929443,
      "num_input_tokens_seen": 144524240,
      "step": 8825,
      "train_runtime": 16970.2868,
      "train_tokens_per_second": 8516.311
    },
    {
      "epoch": 0.5945327228656073,
      "grad_norm": 1.1111788890125827,
      "learning_rate": 7.973981681079487e-06,
      "loss": 0.4075948238372803,
      "num_input_tokens_seen": 144606160,
      "step": 8830,
      "train_runtime": 16979.9055,
      "train_tokens_per_second": 8516.311
    },
    {
      "epoch": 0.5948693778615675,
      "grad_norm": 1.0426300544539855,
      "learning_rate": 7.971855750016603e-06,
      "loss": 0.3943407297134399,
      "num_input_tokens_seen": 144688080,
      "step": 8835,
      "train_runtime": 16989.5353,
      "train_tokens_per_second": 8516.306
    },
    {
      "epoch": 0.5952060328575276,
      "grad_norm": 1.1210656588236518,
      "learning_rate": 7.969728987883734e-06,
      "loss": 0.4295613765716553,
      "num_input_tokens_seen": 144770000,
      "step": 8840,
      "train_runtime": 16999.037,
      "train_tokens_per_second": 8516.365
    },
    {
      "epoch": 0.5955426878534877,
      "grad_norm": 1.1475624279122951,
      "learning_rate": 7.967601395275624e-06,
      "loss": 0.4351513862609863,
      "num_input_tokens_seen": 144851920,
      "step": 8845,
      "train_runtime": 17008.642,
      "train_tokens_per_second": 8516.372
    },
    {
      "epoch": 0.5958793428494479,
      "grad_norm": 0.914150356252195,
      "learning_rate": 7.965472972787247e-06,
      "loss": 0.3699086904525757,
      "num_input_tokens_seen": 144933840,
      "step": 8850,
      "train_runtime": 17018.2481,
      "train_tokens_per_second": 8516.378
    },
    {
      "epoch": 0.596215997845408,
      "grad_norm": 0.9126605794908684,
      "learning_rate": 7.963343721013808e-06,
      "loss": 0.36887919902801514,
      "num_input_tokens_seen": 145015760,
      "step": 8855,
      "train_runtime": 17027.7453,
      "train_tokens_per_second": 8516.439
    },
    {
      "epoch": 0.5965526528413682,
      "grad_norm": 1.1049083147278573,
      "learning_rate": 7.961213640550748e-06,
      "loss": 0.41682796478271483,
      "num_input_tokens_seen": 145097680,
      "step": 8860,
      "train_runtime": 17037.3806,
      "train_tokens_per_second": 8516.431
    },
    {
      "epoch": 0.5968893078373283,
      "grad_norm": 0.9290367923593199,
      "learning_rate": 7.959082731993736e-06,
      "loss": 0.3539085388183594,
      "num_input_tokens_seen": 145179600,
      "step": 8865,
      "train_runtime": 17046.9929,
      "train_tokens_per_second": 8516.435
    },
    {
      "epoch": 0.5972259628332884,
      "grad_norm": 1.1708612233982072,
      "learning_rate": 7.956950995938673e-06,
      "loss": 0.38853328227996825,
      "num_input_tokens_seen": 145261520,
      "step": 8870,
      "train_runtime": 17056.5411,
      "train_tokens_per_second": 8516.47
    },
    {
      "epoch": 0.5975626178292486,
      "grad_norm": 1.1367272579033052,
      "learning_rate": 7.954818432981694e-06,
      "loss": 0.42478394508361816,
      "num_input_tokens_seen": 145343440,
      "step": 8875,
      "train_runtime": 17066.1047,
      "train_tokens_per_second": 8516.498
    },
    {
      "epoch": 0.5978992728252087,
      "grad_norm": 1.038798013991465,
      "learning_rate": 7.952685043719162e-06,
      "loss": 0.417207145690918,
      "num_input_tokens_seen": 145425360,
      "step": 8880,
      "train_runtime": 17075.6796,
      "train_tokens_per_second": 8516.52
    },
    {
      "epoch": 0.5982359278211689,
      "grad_norm": 1.2766367911471153,
      "learning_rate": 7.950550828747672e-06,
      "loss": 0.41807870864868163,
      "num_input_tokens_seen": 145507280,
      "step": 8885,
      "train_runtime": 17085.2213,
      "train_tokens_per_second": 8516.558
    },
    {
      "epoch": 0.598572582817129,
      "grad_norm": 1.1374981977329683,
      "learning_rate": 7.948415788664052e-06,
      "loss": 0.4302704334259033,
      "num_input_tokens_seen": 145589200,
      "step": 8890,
      "train_runtime": 17094.7917,
      "train_tokens_per_second": 8516.582
    },
    {
      "epoch": 0.5989092378130891,
      "grad_norm": 1.1614944146411927,
      "learning_rate": 7.946279924065359e-06,
      "loss": 0.42769107818603513,
      "num_input_tokens_seen": 145671120,
      "step": 8895,
      "train_runtime": 17104.4067,
      "train_tokens_per_second": 8516.584
    },
    {
      "epoch": 0.5992458928090493,
      "grad_norm": 1.229089463345625,
      "learning_rate": 7.94414323554888e-06,
      "loss": 0.40616888999938966,
      "num_input_tokens_seen": 145753040,
      "step": 8900,
      "train_runtime": 17113.9456,
      "train_tokens_per_second": 8516.624
    },
    {
      "epoch": 0.5995825478050094,
      "grad_norm": 1.1810158254271932,
      "learning_rate": 7.942005723712135e-06,
      "loss": 0.4004673004150391,
      "num_input_tokens_seen": 145834960,
      "step": 8905,
      "train_runtime": 17123.4617,
      "train_tokens_per_second": 8516.675
    },
    {
      "epoch": 0.5999192028009696,
      "grad_norm": 1.018492364573803,
      "learning_rate": 7.939867389152868e-06,
      "loss": 0.38183271884918213,
      "num_input_tokens_seen": 145916880,
      "step": 8910,
      "train_runtime": 17133.0705,
      "train_tokens_per_second": 8516.68
    },
    {
      "epoch": 0.6002558577969297,
      "grad_norm": 1.1448585063478829,
      "learning_rate": 7.937728232469062e-06,
      "loss": 0.4183457374572754,
      "num_input_tokens_seen": 145998800,
      "step": 8915,
      "train_runtime": 17142.6849,
      "train_tokens_per_second": 8516.682
    },
    {
      "epoch": 0.6005925127928898,
      "grad_norm": 0.9792335396252364,
      "learning_rate": 7.935588254258921e-06,
      "loss": 0.44341201782226564,
      "num_input_tokens_seen": 146080720,
      "step": 8920,
      "train_runtime": 17152.1879,
      "train_tokens_per_second": 8516.74
    },
    {
      "epoch": 0.60092916778885,
      "grad_norm": 1.1772652920700142,
      "learning_rate": 7.933447455120889e-06,
      "loss": 0.4071674346923828,
      "num_input_tokens_seen": 146162376,
      "step": 8925,
      "train_runtime": 17162.237,
      "train_tokens_per_second": 8516.511
    },
    {
      "epoch": 0.6012658227848101,
      "grad_norm": 1.1675520460930418,
      "learning_rate": 7.931305835653629e-06,
      "loss": 0.39338738918304444,
      "num_input_tokens_seen": 146244296,
      "step": 8930,
      "train_runtime": 17171.8602,
      "train_tokens_per_second": 8516.509
    },
    {
      "epoch": 0.6016024777807703,
      "grad_norm": 1.0112071471272197,
      "learning_rate": 7.929163396456038e-06,
      "loss": 0.4171781063079834,
      "num_input_tokens_seen": 146326216,
      "step": 8935,
      "train_runtime": 17181.3642,
      "train_tokens_per_second": 8516.566
    },
    {
      "epoch": 0.6019391327767304,
      "grad_norm": 1.1592279843526958,
      "learning_rate": 7.927020138127245e-06,
      "loss": 0.4183844566345215,
      "num_input_tokens_seen": 146408136,
      "step": 8940,
      "train_runtime": 17190.9916,
      "train_tokens_per_second": 8516.561
    },
    {
      "epoch": 0.6022757877726905,
      "grad_norm": 1.1305185406629434,
      "learning_rate": 7.924876061266603e-06,
      "loss": 0.423386287689209,
      "num_input_tokens_seen": 146490056,
      "step": 8945,
      "train_runtime": 17200.6001,
      "train_tokens_per_second": 8516.567
    },
    {
      "epoch": 0.6026124427686507,
      "grad_norm": 1.4353622289971242,
      "learning_rate": 7.922731166473698e-06,
      "loss": 0.3871948003768921,
      "num_input_tokens_seen": 146571976,
      "step": 8950,
      "train_runtime": 17210.1444,
      "train_tokens_per_second": 8516.603
    },
    {
      "epoch": 0.6029490977646108,
      "grad_norm": 0.9349556818727319,
      "learning_rate": 7.920585454348341e-06,
      "loss": 0.3567344188690186,
      "num_input_tokens_seen": 146653896,
      "step": 8955,
      "train_runtime": 17219.723,
      "train_tokens_per_second": 8516.623
    },
    {
      "epoch": 0.603285752760571,
      "grad_norm": 0.9425973240239333,
      "learning_rate": 7.918438925490578e-06,
      "loss": 0.38050055503845215,
      "num_input_tokens_seen": 146735816,
      "step": 8960,
      "train_runtime": 17229.3312,
      "train_tokens_per_second": 8516.629
    },
    {
      "epoch": 0.6036224077565311,
      "grad_norm": 1.0627382798659675,
      "learning_rate": 7.916291580500672e-06,
      "loss": 0.4114658355712891,
      "num_input_tokens_seen": 146817736,
      "step": 8965,
      "train_runtime": 17238.8875,
      "train_tokens_per_second": 8516.66
    },
    {
      "epoch": 0.6039590627524912,
      "grad_norm": 1.2191076109820849,
      "learning_rate": 7.914143419979127e-06,
      "loss": 0.4145554542541504,
      "num_input_tokens_seen": 146899656,
      "step": 8970,
      "train_runtime": 17248.4643,
      "train_tokens_per_second": 8516.68
    },
    {
      "epoch": 0.6042957177484514,
      "grad_norm": 0.9520917844387417,
      "learning_rate": 7.911994444526667e-06,
      "loss": 0.3882467269897461,
      "num_input_tokens_seen": 146981576,
      "step": 8975,
      "train_runtime": 17258.0918,
      "train_tokens_per_second": 8516.676
    },
    {
      "epoch": 0.6046323727444115,
      "grad_norm": 0.9400793943868261,
      "learning_rate": 7.909844654744246e-06,
      "loss": 0.3423570156097412,
      "num_input_tokens_seen": 147063496,
      "step": 8980,
      "train_runtime": 17267.7189,
      "train_tokens_per_second": 8516.672
    },
    {
      "epoch": 0.6049690277403716,
      "grad_norm": 1.102629505668985,
      "learning_rate": 7.907694051233044e-06,
      "loss": 0.4534603118896484,
      "num_input_tokens_seen": 147144672,
      "step": 8985,
      "train_runtime": 17277.432,
      "train_tokens_per_second": 8516.582
    },
    {
      "epoch": 0.6053056827363318,
      "grad_norm": 0.9784920654738665,
      "learning_rate": 7.905542634594476e-06,
      "loss": 0.3812037706375122,
      "num_input_tokens_seen": 147226592,
      "step": 8990,
      "train_runtime": 17287.0087,
      "train_tokens_per_second": 8516.603
    },
    {
      "epoch": 0.6056423377322919,
      "grad_norm": 1.1122402679177101,
      "learning_rate": 7.903390405430174e-06,
      "loss": 0.3935607194900513,
      "num_input_tokens_seen": 147308512,
      "step": 8995,
      "train_runtime": 17296.6337,
      "train_tokens_per_second": 8516.6
    },
    {
      "epoch": 0.6059789927282521,
      "grad_norm": 1.1348232774284217,
      "learning_rate": 7.901237364342003e-06,
      "loss": 0.394484543800354,
      "num_input_tokens_seen": 147390432,
      "step": 9000,
      "train_runtime": 17306.143,
      "train_tokens_per_second": 8516.654
    },
    {
      "epoch": 0.6063156477242122,
      "grad_norm": 1.2586024314893387,
      "learning_rate": 7.899083511932053e-06,
      "loss": 0.4022980690002441,
      "num_input_tokens_seen": 147472352,
      "step": 9005,
      "train_runtime": 17315.7159,
      "train_tokens_per_second": 8516.677
    },
    {
      "epoch": 0.6066523027201723,
      "grad_norm": 1.111554945152821,
      "learning_rate": 7.896928848802645e-06,
      "loss": 0.4110873222351074,
      "num_input_tokens_seen": 147554272,
      "step": 9010,
      "train_runtime": 17325.2859,
      "train_tokens_per_second": 8516.701
    },
    {
      "epoch": 0.6069889577161325,
      "grad_norm": 1.0915007511274317,
      "learning_rate": 7.894773375556322e-06,
      "loss": 0.410502290725708,
      "num_input_tokens_seen": 147636192,
      "step": 9015,
      "train_runtime": 17334.7952,
      "train_tokens_per_second": 8516.754
    },
    {
      "epoch": 0.6073256127120926,
      "grad_norm": 1.0012565733139578,
      "learning_rate": 7.892617092795855e-06,
      "loss": 0.42524051666259766,
      "num_input_tokens_seen": 147718112,
      "step": 9020,
      "train_runtime": 17344.4089,
      "train_tokens_per_second": 8516.757
    },
    {
      "epoch": 0.6076622677080528,
      "grad_norm": 1.041656176741022,
      "learning_rate": 7.890460001124242e-06,
      "loss": 0.3956109285354614,
      "num_input_tokens_seen": 147800032,
      "step": 9025,
      "train_runtime": 17354.0224,
      "train_tokens_per_second": 8516.759
    },
    {
      "epoch": 0.6079989227040129,
      "grad_norm": 1.1723199631376697,
      "learning_rate": 7.888302101144707e-06,
      "loss": 0.43264083862304686,
      "num_input_tokens_seen": 147881760,
      "step": 9030,
      "train_runtime": 17364.0657,
      "train_tokens_per_second": 8516.54
    },
    {
      "epoch": 0.608335577699973,
      "grad_norm": 1.0604130902258377,
      "learning_rate": 7.886143393460699e-06,
      "loss": 0.3938114166259766,
      "num_input_tokens_seen": 147963680,
      "step": 9035,
      "train_runtime": 17373.646,
      "train_tokens_per_second": 8516.559
    },
    {
      "epoch": 0.6086722326959332,
      "grad_norm": 0.9875594774033503,
      "learning_rate": 7.883983878675895e-06,
      "loss": 0.4085233688354492,
      "num_input_tokens_seen": 148045600,
      "step": 9040,
      "train_runtime": 17383.2757,
      "train_tokens_per_second": 8516.554
    },
    {
      "epoch": 0.6090088876918933,
      "grad_norm": 1.118442017758606,
      "learning_rate": 7.881823557394194e-06,
      "loss": 0.4521836280822754,
      "num_input_tokens_seen": 148127520,
      "step": 9045,
      "train_runtime": 17392.8251,
      "train_tokens_per_second": 8516.588
    },
    {
      "epoch": 0.6093455426878535,
      "grad_norm": 0.9697823746594351,
      "learning_rate": 7.879662430219728e-06,
      "loss": 0.3711784839630127,
      "num_input_tokens_seen": 148209440,
      "step": 9050,
      "train_runtime": 17402.3905,
      "train_tokens_per_second": 8516.614
    },
    {
      "epoch": 0.6096821976838136,
      "grad_norm": 1.107555777543162,
      "learning_rate": 7.877500497756843e-06,
      "loss": 0.42658653259277346,
      "num_input_tokens_seen": 148291360,
      "step": 9055,
      "train_runtime": 17412.0148,
      "train_tokens_per_second": 8516.611
    },
    {
      "epoch": 0.6100188526797737,
      "grad_norm": 1.3261394190294302,
      "learning_rate": 7.87533776061012e-06,
      "loss": 0.3768439769744873,
      "num_input_tokens_seen": 148373280,
      "step": 9060,
      "train_runtime": 17421.58,
      "train_tokens_per_second": 8516.637
    },
    {
      "epoch": 0.6103555076757339,
      "grad_norm": 1.1467211782124616,
      "learning_rate": 7.873174219384362e-06,
      "loss": 0.443784761428833,
      "num_input_tokens_seen": 148455200,
      "step": 9065,
      "train_runtime": 17431.1609,
      "train_tokens_per_second": 8516.656
    },
    {
      "epoch": 0.610692162671694,
      "grad_norm": 1.2544634924943416,
      "learning_rate": 7.871009874684595e-06,
      "loss": 0.43168983459472654,
      "num_input_tokens_seen": 148537120,
      "step": 9070,
      "train_runtime": 17440.7757,
      "train_tokens_per_second": 8516.658
    },
    {
      "epoch": 0.6110288176676542,
      "grad_norm": 1.306677665757552,
      "learning_rate": 7.868844727116072e-06,
      "loss": 0.4441550254821777,
      "num_input_tokens_seen": 148617664,
      "step": 9075,
      "train_runtime": 17450.454,
      "train_tokens_per_second": 8516.55
    },
    {
      "epoch": 0.6113654726636143,
      "grad_norm": 0.9943533828610129,
      "learning_rate": 7.866678777284267e-06,
      "loss": 0.4087955951690674,
      "num_input_tokens_seen": 148699584,
      "step": 9080,
      "train_runtime": 17460.0376,
      "train_tokens_per_second": 8516.567
    },
    {
      "epoch": 0.6117021276595744,
      "grad_norm": 1.1037218570565088,
      "learning_rate": 7.864512025794886e-06,
      "loss": 0.4202601432800293,
      "num_input_tokens_seen": 148781504,
      "step": 9085,
      "train_runtime": 17469.6491,
      "train_tokens_per_second": 8516.571
    },
    {
      "epoch": 0.6120387826555346,
      "grad_norm": 1.030828014450681,
      "learning_rate": 7.86234447325385e-06,
      "loss": 0.37147841453552244,
      "num_input_tokens_seen": 148863424,
      "step": 9090,
      "train_runtime": 17479.2628,
      "train_tokens_per_second": 8516.573
    },
    {
      "epoch": 0.6123754376514947,
      "grad_norm": 1.0925394387628096,
      "learning_rate": 7.860176120267309e-06,
      "loss": 0.4597188949584961,
      "num_input_tokens_seen": 148945344,
      "step": 9095,
      "train_runtime": 17488.772,
      "train_tokens_per_second": 8516.627
    },
    {
      "epoch": 0.6127120926474549,
      "grad_norm": 0.9717739539622525,
      "learning_rate": 7.858006967441637e-06,
      "loss": 0.41214141845703123,
      "num_input_tokens_seen": 149027264,
      "step": 9100,
      "train_runtime": 17498.3867,
      "train_tokens_per_second": 8516.629
    },
    {
      "epoch": 0.613048747643415,
      "grad_norm": 1.185420782737996,
      "learning_rate": 7.85583701538343e-06,
      "loss": 0.4528674125671387,
      "num_input_tokens_seen": 149109184,
      "step": 9105,
      "train_runtime": 17507.9839,
      "train_tokens_per_second": 8516.639
    },
    {
      "epoch": 0.6133854026393751,
      "grad_norm": 1.1456450846046762,
      "learning_rate": 7.853666264699506e-06,
      "loss": 0.39894838333129884,
      "num_input_tokens_seen": 149191104,
      "step": 9110,
      "train_runtime": 17517.4849,
      "train_tokens_per_second": 8516.697
    },
    {
      "epoch": 0.6137220576353353,
      "grad_norm": 1.188855571591462,
      "learning_rate": 7.851494715996914e-06,
      "loss": 0.4338830471038818,
      "num_input_tokens_seen": 149273024,
      "step": 9115,
      "train_runtime": 17527.6241,
      "train_tokens_per_second": 8516.444
    },
    {
      "epoch": 0.6140587126312954,
      "grad_norm": 1.0399853658134395,
      "learning_rate": 7.849322369882915e-06,
      "loss": 0.39254772663116455,
      "num_input_tokens_seen": 149354944,
      "step": 9120,
      "train_runtime": 17537.2347,
      "train_tokens_per_second": 8516.448
    },
    {
      "epoch": 0.6143953676272556,
      "grad_norm": 1.1343143807609875,
      "learning_rate": 7.847149226965003e-06,
      "loss": 0.40407791137695315,
      "num_input_tokens_seen": 149436864,
      "step": 9125,
      "train_runtime": 17546.786,
      "train_tokens_per_second": 8516.481
    },
    {
      "epoch": 0.6147320226232157,
      "grad_norm": 1.0902093933367927,
      "learning_rate": 7.844975287850886e-06,
      "loss": 0.34553213119506837,
      "num_input_tokens_seen": 149518784,
      "step": 9130,
      "train_runtime": 17556.3641,
      "train_tokens_per_second": 8516.501
    },
    {
      "epoch": 0.6150686776191758,
      "grad_norm": 1.2461089156181326,
      "learning_rate": 7.842800553148506e-06,
      "loss": 0.43389148712158204,
      "num_input_tokens_seen": 149600704,
      "step": 9135,
      "train_runtime": 17565.9718,
      "train_tokens_per_second": 8516.506
    },
    {
      "epoch": 0.615405332615136,
      "grad_norm": 0.8986082374298828,
      "learning_rate": 7.840625023466014e-06,
      "loss": 0.3965892791748047,
      "num_input_tokens_seen": 149682624,
      "step": 9140,
      "train_runtime": 17575.5202,
      "train_tokens_per_second": 8516.54
    },
    {
      "epoch": 0.6157419876110961,
      "grad_norm": 1.096363544278267,
      "learning_rate": 7.838448699411792e-06,
      "loss": 0.3813614368438721,
      "num_input_tokens_seen": 149764544,
      "step": 9145,
      "train_runtime": 17585.0853,
      "train_tokens_per_second": 8516.566
    },
    {
      "epoch": 0.6160786426070562,
      "grad_norm": 1.4091953288858565,
      "learning_rate": 7.836271581594442e-06,
      "loss": 0.4375908851623535,
      "num_input_tokens_seen": 149846464,
      "step": 9150,
      "train_runtime": 17594.7223,
      "train_tokens_per_second": 8516.557
    },
    {
      "epoch": 0.6164152976030164,
      "grad_norm": 1.231347864620381,
      "learning_rate": 7.834093670622789e-06,
      "loss": 0.4664903163909912,
      "num_input_tokens_seen": 149928384,
      "step": 9155,
      "train_runtime": 17604.2631,
      "train_tokens_per_second": 8516.595
    },
    {
      "epoch": 0.6167519525989765,
      "grad_norm": 1.0707152158875135,
      "learning_rate": 7.831914967105878e-06,
      "loss": 0.4460482597351074,
      "num_input_tokens_seen": 150010304,
      "step": 9160,
      "train_runtime": 17613.8237,
      "train_tokens_per_second": 8516.623
    },
    {
      "epoch": 0.6170886075949367,
      "grad_norm": 1.0163464841512795,
      "learning_rate": 7.829735471652978e-06,
      "loss": 0.3833907604217529,
      "num_input_tokens_seen": 150092224,
      "step": 9165,
      "train_runtime": 17623.4392,
      "train_tokens_per_second": 8516.625
    },
    {
      "epoch": 0.6174252625908968,
      "grad_norm": 1.1050638728541433,
      "learning_rate": 7.827555184873575e-06,
      "loss": 0.38476583957672117,
      "num_input_tokens_seen": 150174144,
      "step": 9170,
      "train_runtime": 17633.0637,
      "train_tokens_per_second": 8516.622
    },
    {
      "epoch": 0.617761917586857,
      "grad_norm": 0.9959644095045128,
      "learning_rate": 7.825374107377381e-06,
      "loss": 0.3778055191040039,
      "num_input_tokens_seen": 150256064,
      "step": 9175,
      "train_runtime": 17642.5722,
      "train_tokens_per_second": 8516.676
    },
    {
      "epoch": 0.6180985725828171,
      "grad_norm": 1.2762902409502115,
      "learning_rate": 7.823192239774327e-06,
      "loss": 0.4246524333953857,
      "num_input_tokens_seen": 150337984,
      "step": 9180,
      "train_runtime": 17652.1278,
      "train_tokens_per_second": 8516.706
    },
    {
      "epoch": 0.6184352275787772,
      "grad_norm": 0.9997321536484403,
      "learning_rate": 7.821009582674567e-06,
      "loss": 0.4072460174560547,
      "num_input_tokens_seen": 150419904,
      "step": 9185,
      "train_runtime": 17661.745,
      "train_tokens_per_second": 8516.707
    },
    {
      "epoch": 0.6187718825747374,
      "grad_norm": 1.323401219858831,
      "learning_rate": 7.818826136688473e-06,
      "loss": 0.4190455436706543,
      "num_input_tokens_seen": 150501824,
      "step": 9190,
      "train_runtime": 17671.2493,
      "train_tokens_per_second": 8516.762
    },
    {
      "epoch": 0.6191085375706975,
      "grad_norm": 1.1441272068648525,
      "learning_rate": 7.816641902426638e-06,
      "loss": 0.4555508613586426,
      "num_input_tokens_seen": 150583592,
      "step": 9195,
      "train_runtime": 17681.3746,
      "train_tokens_per_second": 8516.509
    },
    {
      "epoch": 0.6194451925666576,
      "grad_norm": 1.1289728518608724,
      "learning_rate": 7.814456880499877e-06,
      "loss": 0.4126096248626709,
      "num_input_tokens_seen": 150665512,
      "step": 9200,
      "train_runtime": 17690.9937,
      "train_tokens_per_second": 8516.509
    },
    {
      "epoch": 0.6197818475626178,
      "grad_norm": 1.1626404033366027,
      "learning_rate": 7.812271071519222e-06,
      "loss": 0.4123558521270752,
      "num_input_tokens_seen": 150747432,
      "step": 9205,
      "train_runtime": 17700.4982,
      "train_tokens_per_second": 8516.564
    },
    {
      "epoch": 0.6201185025585779,
      "grad_norm": 1.0770082918114245,
      "learning_rate": 7.810084476095932e-06,
      "loss": 0.4001502990722656,
      "num_input_tokens_seen": 150829352,
      "step": 9210,
      "train_runtime": 17710.1225,
      "train_tokens_per_second": 8516.562
    },
    {
      "epoch": 0.6204551575545381,
      "grad_norm": 1.0643737072990638,
      "learning_rate": 7.807897094841476e-06,
      "loss": 0.3789778232574463,
      "num_input_tokens_seen": 150911272,
      "step": 9215,
      "train_runtime": 17719.7268,
      "train_tokens_per_second": 8516.569
    },
    {
      "epoch": 0.6207918125504982,
      "grad_norm": 1.0312685618528032,
      "learning_rate": 7.805708928367555e-06,
      "loss": 0.37386040687561034,
      "num_input_tokens_seen": 150992920,
      "step": 9220,
      "train_runtime": 17729.6954,
      "train_tokens_per_second": 8516.385
    },
    {
      "epoch": 0.6211284675464583,
      "grad_norm": 1.2219709838066952,
      "learning_rate": 7.803519977286075e-06,
      "loss": 0.42103891372680663,
      "num_input_tokens_seen": 151074296,
      "step": 9225,
      "train_runtime": 17739.5813,
      "train_tokens_per_second": 8516.227
    },
    {
      "epoch": 0.6214651225424185,
      "grad_norm": 1.053539157575372,
      "learning_rate": 7.801330242209176e-06,
      "loss": 0.3968287706375122,
      "num_input_tokens_seen": 151156216,
      "step": 9230,
      "train_runtime": 17749.1838,
      "train_tokens_per_second": 8516.235
    },
    {
      "epoch": 0.6218017775383786,
      "grad_norm": 1.2093991145784777,
      "learning_rate": 7.799139723749207e-06,
      "loss": 0.4207425117492676,
      "num_input_tokens_seen": 151238136,
      "step": 9235,
      "train_runtime": 17758.7995,
      "train_tokens_per_second": 8516.236
    },
    {
      "epoch": 0.6221384325343388,
      "grad_norm": 1.083487206639776,
      "learning_rate": 7.79694842251874e-06,
      "loss": 0.44968671798706056,
      "num_input_tokens_seen": 151320056,
      "step": 9240,
      "train_runtime": 17768.3038,
      "train_tokens_per_second": 8516.292
    },
    {
      "epoch": 0.6224750875302989,
      "grad_norm": 1.0187149370253636,
      "learning_rate": 7.794756339130565e-06,
      "loss": 0.39505729675292967,
      "num_input_tokens_seen": 151401976,
      "step": 9245,
      "train_runtime": 17777.9126,
      "train_tokens_per_second": 8516.297
    },
    {
      "epoch": 0.622811742526259,
      "grad_norm": 0.9748356218254983,
      "learning_rate": 7.792563474197692e-06,
      "loss": 0.39171295166015624,
      "num_input_tokens_seen": 151483896,
      "step": 9250,
      "train_runtime": 17787.5438,
      "train_tokens_per_second": 8516.291
    },
    {
      "epoch": 0.6231483975222192,
      "grad_norm": 1.0952813430274677,
      "learning_rate": 7.79036982833335e-06,
      "loss": 0.4544875144958496,
      "num_input_tokens_seen": 151565816,
      "step": 9255,
      "train_runtime": 17797.0476,
      "train_tokens_per_second": 8516.346
    },
    {
      "epoch": 0.6234850525181793,
      "grad_norm": 1.0394446923151508,
      "learning_rate": 7.78817540215098e-06,
      "loss": 0.42662763595581055,
      "num_input_tokens_seen": 151647736,
      "step": 9260,
      "train_runtime": 17806.6558,
      "train_tokens_per_second": 8516.351
    },
    {
      "epoch": 0.6238217075141395,
      "grad_norm": 0.877629015849481,
      "learning_rate": 7.78598019626425e-06,
      "loss": 0.38350696563720704,
      "num_input_tokens_seen": 151729656,
      "step": 9265,
      "train_runtime": 17816.2768,
      "train_tokens_per_second": 8516.35
    },
    {
      "epoch": 0.6241583625100996,
      "grad_norm": 1.0440221825323934,
      "learning_rate": 7.783784211287044e-06,
      "loss": 0.408212947845459,
      "num_input_tokens_seen": 151811048,
      "step": 9270,
      "train_runtime": 17826.1453,
      "train_tokens_per_second": 8516.202
    },
    {
      "epoch": 0.6244950175060597,
      "grad_norm": 1.063889999206071,
      "learning_rate": 7.78158744783346e-06,
      "loss": 0.4311057090759277,
      "num_input_tokens_seen": 151892968,
      "step": 9275,
      "train_runtime": 17835.6486,
      "train_tokens_per_second": 8516.257
    },
    {
      "epoch": 0.6248316725020199,
      "grad_norm": 1.19265062566187,
      "learning_rate": 7.779389906517813e-06,
      "loss": 0.36039080619812014,
      "num_input_tokens_seen": 151974888,
      "step": 9280,
      "train_runtime": 17845.2643,
      "train_tokens_per_second": 8516.259
    },
    {
      "epoch": 0.6251683274979801,
      "grad_norm": 1.2496532061295844,
      "learning_rate": 7.777191587954645e-06,
      "loss": 0.3891784191131592,
      "num_input_tokens_seen": 152056080,
      "step": 9285,
      "train_runtime": 17855.0224,
      "train_tokens_per_second": 8516.152
    },
    {
      "epoch": 0.6255049824939403,
      "grad_norm": 1.0494986475816006,
      "learning_rate": 7.774992492758704e-06,
      "loss": 0.37932262420654295,
      "num_input_tokens_seen": 152138000,
      "step": 9290,
      "train_runtime": 17864.5939,
      "train_tokens_per_second": 8516.175
    },
    {
      "epoch": 0.6258416374899004,
      "grad_norm": 1.018263559537958,
      "learning_rate": 7.77279262154496e-06,
      "loss": 0.41188697814941405,
      "num_input_tokens_seen": 152219920,
      "step": 9295,
      "train_runtime": 17874.1931,
      "train_tokens_per_second": 8516.184
    },
    {
      "epoch": 0.6261782924858605,
      "grad_norm": 1.0944757483756982,
      "learning_rate": 7.770591974928604e-06,
      "loss": 0.46198062896728515,
      "num_input_tokens_seen": 152301840,
      "step": 9300,
      "train_runtime": 17883.8029,
      "train_tokens_per_second": 8516.189
    },
    {
      "epoch": 0.6265149474818207,
      "grad_norm": 1.074896425774843,
      "learning_rate": 7.768390553525034e-06,
      "loss": 0.3764986991882324,
      "num_input_tokens_seen": 152383760,
      "step": 9305,
      "train_runtime": 17893.306,
      "train_tokens_per_second": 8516.244
    },
    {
      "epoch": 0.6268516024777808,
      "grad_norm": 1.0256171368215212,
      "learning_rate": 7.766188357949875e-06,
      "loss": 0.3668057441711426,
      "num_input_tokens_seen": 152465680,
      "step": 9310,
      "train_runtime": 17902.8672,
      "train_tokens_per_second": 8516.272
    },
    {
      "epoch": 0.627188257473741,
      "grad_norm": 0.9198606785270225,
      "learning_rate": 7.763985388818963e-06,
      "loss": 0.38221235275268556,
      "num_input_tokens_seen": 152547600,
      "step": 9315,
      "train_runtime": 17912.4832,
      "train_tokens_per_second": 8516.273
    },
    {
      "epoch": 0.6275249124697011,
      "grad_norm": 1.0397090130627937,
      "learning_rate": 7.76178164674835e-06,
      "loss": 0.34249629974365237,
      "num_input_tokens_seen": 152629520,
      "step": 9320,
      "train_runtime": 17921.9876,
      "train_tokens_per_second": 8516.328
    },
    {
      "epoch": 0.6278615674656612,
      "grad_norm": 1.1909284533345306,
      "learning_rate": 7.759577132354305e-06,
      "loss": 0.4116672515869141,
      "num_input_tokens_seen": 152711440,
      "step": 9325,
      "train_runtime": 17931.6194,
      "train_tokens_per_second": 8516.322
    },
    {
      "epoch": 0.6281982224616214,
      "grad_norm": 1.1341379657542192,
      "learning_rate": 7.757371846253315e-06,
      "loss": 0.39189910888671875,
      "num_input_tokens_seen": 152793256,
      "step": 9330,
      "train_runtime": 17941.7407,
      "train_tokens_per_second": 8516.078
    },
    {
      "epoch": 0.6285348774575815,
      "grad_norm": 0.8993023060141472,
      "learning_rate": 7.755165789062083e-06,
      "loss": 0.39326493740081786,
      "num_input_tokens_seen": 152875176,
      "step": 9335,
      "train_runtime": 17951.2437,
      "train_tokens_per_second": 8516.133
    },
    {
      "epoch": 0.6288715324535417,
      "grad_norm": 1.083486471639662,
      "learning_rate": 7.752958961397522e-06,
      "loss": 0.4194042682647705,
      "num_input_tokens_seen": 152957096,
      "step": 9340,
      "train_runtime": 17960.8548,
      "train_tokens_per_second": 8516.137
    },
    {
      "epoch": 0.6292081874495018,
      "grad_norm": 1.0075370516942888,
      "learning_rate": 7.750751363876765e-06,
      "loss": 0.3857820749282837,
      "num_input_tokens_seen": 153039016,
      "step": 9345,
      "train_runtime": 17970.4815,
      "train_tokens_per_second": 8516.133
    },
    {
      "epoch": 0.6295448424454619,
      "grad_norm": 1.1406936435343482,
      "learning_rate": 7.748542997117162e-06,
      "loss": 0.4285097122192383,
      "num_input_tokens_seen": 153120936,
      "step": 9350,
      "train_runtime": 17979.9842,
      "train_tokens_per_second": 8516.189
    },
    {
      "epoch": 0.6298814974414221,
      "grad_norm": 0.9327386443845574,
      "learning_rate": 7.746333861736275e-06,
      "loss": 0.36500561237335205,
      "num_input_tokens_seen": 153202856,
      "step": 9355,
      "train_runtime": 17989.6014,
      "train_tokens_per_second": 8516.19
    },
    {
      "epoch": 0.6302181524373822,
      "grad_norm": 1.2497384089175085,
      "learning_rate": 7.74412395835188e-06,
      "loss": 0.4108287811279297,
      "num_input_tokens_seen": 153284776,
      "step": 9360,
      "train_runtime": 17999.2237,
      "train_tokens_per_second": 8516.188
    },
    {
      "epoch": 0.6305548074333424,
      "grad_norm": 1.1631167488782035,
      "learning_rate": 7.74191328758197e-06,
      "loss": 0.39518916606903076,
      "num_input_tokens_seen": 153366696,
      "step": 9365,
      "train_runtime": 18008.7685,
      "train_tokens_per_second": 8516.223
    },
    {
      "epoch": 0.6308914624293025,
      "grad_norm": 1.1594218531902865,
      "learning_rate": 7.739701850044752e-06,
      "loss": 0.4469768047332764,
      "num_input_tokens_seen": 153448616,
      "step": 9370,
      "train_runtime": 18018.3487,
      "train_tokens_per_second": 8516.242
    },
    {
      "epoch": 0.6312281174252626,
      "grad_norm": 1.056320730165815,
      "learning_rate": 7.73748964635865e-06,
      "loss": 0.3871025562286377,
      "num_input_tokens_seen": 153530536,
      "step": 9375,
      "train_runtime": 18027.9927,
      "train_tokens_per_second": 8516.23
    },
    {
      "epoch": 0.6315647724212228,
      "grad_norm": 0.967319995265898,
      "learning_rate": 7.735276677142297e-06,
      "loss": 0.3640017509460449,
      "num_input_tokens_seen": 153612432,
      "step": 9380,
      "train_runtime": 18037.5438,
      "train_tokens_per_second": 8516.261
    },
    {
      "epoch": 0.6319014274171829,
      "grad_norm": 1.1535882482796063,
      "learning_rate": 7.733062943014542e-06,
      "loss": 0.4282517910003662,
      "num_input_tokens_seen": 153694352,
      "step": 9385,
      "train_runtime": 18047.1222,
      "train_tokens_per_second": 8516.28
    },
    {
      "epoch": 0.632238082413143,
      "grad_norm": 1.0111813728563364,
      "learning_rate": 7.730848444594452e-06,
      "loss": 0.36345796585083007,
      "num_input_tokens_seen": 153776272,
      "step": 9390,
      "train_runtime": 18056.7414,
      "train_tokens_per_second": 8516.28
    },
    {
      "epoch": 0.6325747374091032,
      "grad_norm": 0.9166917448048076,
      "learning_rate": 7.7286331825013e-06,
      "loss": 0.3752426624298096,
      "num_input_tokens_seen": 153858192,
      "step": 9395,
      "train_runtime": 18066.3278,
      "train_tokens_per_second": 8516.296
    },
    {
      "epoch": 0.6329113924050633,
      "grad_norm": 1.196863082026833,
      "learning_rate": 7.72641715735458e-06,
      "loss": 0.3655829906463623,
      "num_input_tokens_seen": 153940112,
      "step": 9400,
      "train_runtime": 18075.8815,
      "train_tokens_per_second": 8516.327
    },
    {
      "epoch": 0.6332480474010235,
      "grad_norm": 1.1497932663368033,
      "learning_rate": 7.724200369773998e-06,
      "loss": 0.4303262710571289,
      "num_input_tokens_seen": 154022032,
      "step": 9405,
      "train_runtime": 18085.4466,
      "train_tokens_per_second": 8516.352
    },
    {
      "epoch": 0.6335847023969836,
      "grad_norm": 1.0442522149207758,
      "learning_rate": 7.721982820379467e-06,
      "loss": 0.3783085346221924,
      "num_input_tokens_seen": 154103952,
      "step": 9410,
      "train_runtime": 18095.0583,
      "train_tokens_per_second": 8516.356
    },
    {
      "epoch": 0.6339213573929438,
      "grad_norm": 1.0480790282873271,
      "learning_rate": 7.719764509791123e-06,
      "loss": 0.38660488128662107,
      "num_input_tokens_seen": 154185208,
      "step": 9415,
      "train_runtime": 18104.881,
      "train_tokens_per_second": 8516.223
    },
    {
      "epoch": 0.6342580123889039,
      "grad_norm": 1.2831908996599184,
      "learning_rate": 7.717545438629303e-06,
      "loss": 0.4758640766143799,
      "num_input_tokens_seen": 154267128,
      "step": 9420,
      "train_runtime": 18114.5342,
      "train_tokens_per_second": 8516.207
    },
    {
      "epoch": 0.634594667384864,
      "grad_norm": 1.008494207291059,
      "learning_rate": 7.71532560751457e-06,
      "loss": 0.36896114349365233,
      "num_input_tokens_seen": 154349048,
      "step": 9425,
      "train_runtime": 18124.1757,
      "train_tokens_per_second": 8516.197
    },
    {
      "epoch": 0.6349313223808242,
      "grad_norm": 1.2388606362584664,
      "learning_rate": 7.713105017067686e-06,
      "loss": 0.44253387451171877,
      "num_input_tokens_seen": 154430968,
      "step": 9430,
      "train_runtime": 18133.7332,
      "train_tokens_per_second": 8516.226
    },
    {
      "epoch": 0.6352679773767843,
      "grad_norm": 1.071172874769954,
      "learning_rate": 7.710883667909639e-06,
      "loss": 0.413133430480957,
      "num_input_tokens_seen": 154512888,
      "step": 9435,
      "train_runtime": 18143.3268,
      "train_tokens_per_second": 8516.238
    },
    {
      "epoch": 0.6356046323727444,
      "grad_norm": 1.0208768915217314,
      "learning_rate": 7.708661560661617e-06,
      "loss": 0.4196889877319336,
      "num_input_tokens_seen": 154594808,
      "step": 9440,
      "train_runtime": 18152.8385,
      "train_tokens_per_second": 8516.288
    },
    {
      "epoch": 0.6359412873687046,
      "grad_norm": 1.1324360943416445,
      "learning_rate": 7.706438695945026e-06,
      "loss": 0.40938506126403806,
      "num_input_tokens_seen": 154676728,
      "step": 9445,
      "train_runtime": 18162.4629,
      "train_tokens_per_second": 8516.286
    },
    {
      "epoch": 0.6362779423646647,
      "grad_norm": 1.0377954753581717,
      "learning_rate": 7.704215074381484e-06,
      "loss": 0.3799943447113037,
      "num_input_tokens_seen": 154758560,
      "step": 9450,
      "train_runtime": 18172.5384,
      "train_tokens_per_second": 8516.067
    },
    {
      "epoch": 0.6366145973606249,
      "grad_norm": 1.3031438976365568,
      "learning_rate": 7.70199069659282e-06,
      "loss": 0.423862886428833,
      "num_input_tokens_seen": 154840480,
      "step": 9455,
      "train_runtime": 18182.1392,
      "train_tokens_per_second": 8516.076
    },
    {
      "epoch": 0.636951252356585,
      "grad_norm": 1.0570110190326276,
      "learning_rate": 7.699765563201075e-06,
      "loss": 0.4290806770324707,
      "num_input_tokens_seen": 154922400,
      "step": 9460,
      "train_runtime": 18191.6343,
      "train_tokens_per_second": 8516.134
    },
    {
      "epoch": 0.6372879073525451,
      "grad_norm": 1.054754107243349,
      "learning_rate": 7.697539674828498e-06,
      "loss": 0.3895771026611328,
      "num_input_tokens_seen": 155004320,
      "step": 9465,
      "train_runtime": 18201.2421,
      "train_tokens_per_second": 8516.14
    },
    {
      "epoch": 0.6376245623485053,
      "grad_norm": 0.9437841798129473,
      "learning_rate": 7.695313032097554e-06,
      "loss": 0.3700826168060303,
      "num_input_tokens_seen": 155086240,
      "step": 9470,
      "train_runtime": 18210.8691,
      "train_tokens_per_second": 8516.136
    },
    {
      "epoch": 0.6379612173444654,
      "grad_norm": 1.0204219349927928,
      "learning_rate": 7.693085635630913e-06,
      "loss": 0.39543638229370115,
      "num_input_tokens_seen": 155168160,
      "step": 9475,
      "train_runtime": 18220.4963,
      "train_tokens_per_second": 8516.132
    },
    {
      "epoch": 0.6382978723404256,
      "grad_norm": 1.3451205374751678,
      "learning_rate": 7.690857486051463e-06,
      "loss": 0.3514960527420044,
      "num_input_tokens_seen": 155250080,
      "step": 9480,
      "train_runtime": 18230.0002,
      "train_tokens_per_second": 8516.186
    },
    {
      "epoch": 0.6386345273363857,
      "grad_norm": 1.1281806012935907,
      "learning_rate": 7.688628583982298e-06,
      "loss": 0.3633317470550537,
      "num_input_tokens_seen": 155331768,
      "step": 9485,
      "train_runtime": 18239.6356,
      "train_tokens_per_second": 8516.166
    },
    {
      "epoch": 0.6389711823323458,
      "grad_norm": 1.0160948706218333,
      "learning_rate": 7.68639893004672e-06,
      "loss": 0.3717019081115723,
      "num_input_tokens_seen": 155413688,
      "step": 9490,
      "train_runtime": 18249.2645,
      "train_tokens_per_second": 8516.162
    },
    {
      "epoch": 0.639307837328306,
      "grad_norm": 0.9506793164484822,
      "learning_rate": 7.684168524868253e-06,
      "loss": 0.38943655490875245,
      "num_input_tokens_seen": 155495608,
      "step": 9495,
      "train_runtime": 18258.8724,
      "train_tokens_per_second": 8516.167
    },
    {
      "epoch": 0.6396444923242661,
      "grad_norm": 1.2380270826157487,
      "learning_rate": 7.681937369070612e-06,
      "loss": 0.4027376651763916,
      "num_input_tokens_seen": 155577120,
      "step": 9500,
      "train_runtime": 18268.8564,
      "train_tokens_per_second": 8515.975
    },
    {
      "epoch": 0.6399811473202263,
      "grad_norm": 0.9964409630075057,
      "learning_rate": 7.67970546327774e-06,
      "loss": 0.3849789142608643,
      "num_input_tokens_seen": 155659040,
      "step": 9505,
      "train_runtime": 18278.3526,
      "train_tokens_per_second": 8516.032
    },
    {
      "epoch": 0.6403178023161864,
      "grad_norm": 0.9756527723207312,
      "learning_rate": 7.67747280811378e-06,
      "loss": 0.3420548439025879,
      "num_input_tokens_seen": 155740960,
      "step": 9510,
      "train_runtime": 18287.8651,
      "train_tokens_per_second": 8516.082
    },
    {
      "epoch": 0.6406544573121465,
      "grad_norm": 1.0010384190934445,
      "learning_rate": 7.675239404203088e-06,
      "loss": 0.36903891563415525,
      "num_input_tokens_seen": 155822880,
      "step": 9515,
      "train_runtime": 18297.5669,
      "train_tokens_per_second": 8516.044
    },
    {
      "epoch": 0.6409911123081067,
      "grad_norm": 1.3447082884897494,
      "learning_rate": 7.673005252170226e-06,
      "loss": 0.4460052490234375,
      "num_input_tokens_seen": 155904800,
      "step": 9520,
      "train_runtime": 18307.196,
      "train_tokens_per_second": 8516.039
    },
    {
      "epoch": 0.6413277673040668,
      "grad_norm": 1.1513477307639224,
      "learning_rate": 7.670770352639968e-06,
      "loss": 0.38859148025512696,
      "num_input_tokens_seen": 155986720,
      "step": 9525,
      "train_runtime": 18316.8129,
      "train_tokens_per_second": 8516.04
    },
    {
      "epoch": 0.641664422300027,
      "grad_norm": 1.026336386633488,
      "learning_rate": 7.668534706237299e-06,
      "loss": 0.3940721035003662,
      "num_input_tokens_seen": 156068640,
      "step": 9530,
      "train_runtime": 18326.3223,
      "train_tokens_per_second": 8516.092
    },
    {
      "epoch": 0.6420010772959871,
      "grad_norm": 1.3595954267101626,
      "learning_rate": 7.666298313587409e-06,
      "loss": 0.4049412727355957,
      "num_input_tokens_seen": 156150560,
      "step": 9535,
      "train_runtime": 18335.9482,
      "train_tokens_per_second": 8516.089
    },
    {
      "epoch": 0.6423377322919472,
      "grad_norm": 1.0754747912417157,
      "learning_rate": 7.664061175315696e-06,
      "loss": 0.4020057201385498,
      "num_input_tokens_seen": 156232480,
      "step": 9540,
      "train_runtime": 18345.5565,
      "train_tokens_per_second": 8516.094
    },
    {
      "epoch": 0.6426743872879074,
      "grad_norm": 1.0749316515472216,
      "learning_rate": 7.661823292047771e-06,
      "loss": 0.41860275268554686,
      "num_input_tokens_seen": 156314152,
      "step": 9545,
      "train_runtime": 18355.6885,
      "train_tokens_per_second": 8515.842
    },
    {
      "epoch": 0.6430110422838675,
      "grad_norm": 1.1553090723065882,
      "learning_rate": 7.659584664409452e-06,
      "loss": 0.4082381248474121,
      "num_input_tokens_seen": 156396072,
      "step": 9550,
      "train_runtime": 18365.3132,
      "train_tokens_per_second": 8515.84
    },
    {
      "epoch": 0.6433476972798277,
      "grad_norm": 0.9262058241622351,
      "learning_rate": 7.65734529302676e-06,
      "loss": 0.3428227424621582,
      "num_input_tokens_seen": 156477992,
      "step": 9555,
      "train_runtime": 18374.8634,
      "train_tokens_per_second": 8515.872
    },
    {
      "epoch": 0.6436843522757878,
      "grad_norm": 1.1619417999876804,
      "learning_rate": 7.655105178525932e-06,
      "loss": 0.3573341131210327,
      "num_input_tokens_seen": 156559912,
      "step": 9560,
      "train_runtime": 18384.4354,
      "train_tokens_per_second": 8515.894
    },
    {
      "epoch": 0.6440210072717479,
      "grad_norm": 1.1272069610732751,
      "learning_rate": 7.652864321533406e-06,
      "loss": 0.417873477935791,
      "num_input_tokens_seen": 156641832,
      "step": 9565,
      "train_runtime": 18394.0589,
      "train_tokens_per_second": 8515.893
    },
    {
      "epoch": 0.6443576622677081,
      "grad_norm": 0.9770014471710045,
      "learning_rate": 7.650622722675833e-06,
      "loss": 0.42095365524291994,
      "num_input_tokens_seen": 156723752,
      "step": 9570,
      "train_runtime": 18403.6231,
      "train_tokens_per_second": 8515.918
    },
    {
      "epoch": 0.6446943172636682,
      "grad_norm": 1.1385977734185821,
      "learning_rate": 7.648380382580067e-06,
      "loss": 0.3930374622344971,
      "num_input_tokens_seen": 156805672,
      "step": 9575,
      "train_runtime": 18413.1792,
      "train_tokens_per_second": 8515.948
    },
    {
      "epoch": 0.6450309722596284,
      "grad_norm": 1.0994404053713425,
      "learning_rate": 7.646137301873173e-06,
      "loss": 0.4031050205230713,
      "num_input_tokens_seen": 156887592,
      "step": 9580,
      "train_runtime": 18422.7981,
      "train_tokens_per_second": 8515.948
    },
    {
      "epoch": 0.6453676272555885,
      "grad_norm": 1.0436977551341817,
      "learning_rate": 7.643893481182421e-06,
      "loss": 0.37503063678741455,
      "num_input_tokens_seen": 156969512,
      "step": 9585,
      "train_runtime": 18432.4098,
      "train_tokens_per_second": 8515.952
    },
    {
      "epoch": 0.6457042822515486,
      "grad_norm": 1.2248016230987313,
      "learning_rate": 7.641648921135287e-06,
      "loss": 0.45342607498168946,
      "num_input_tokens_seen": 157051432,
      "step": 9590,
      "train_runtime": 18441.9242,
      "train_tokens_per_second": 8516.0
    },
    {
      "epoch": 0.6460409372475088,
      "grad_norm": 1.3245139113377522,
      "learning_rate": 7.639403622359455e-06,
      "loss": 0.4449787139892578,
      "num_input_tokens_seen": 157133352,
      "step": 9595,
      "train_runtime": 18451.5271,
      "train_tokens_per_second": 8516.008
    },
    {
      "epoch": 0.6463775922434689,
      "grad_norm": 1.2481045243875524,
      "learning_rate": 7.637157585482817e-06,
      "loss": 0.3938270092010498,
      "num_input_tokens_seen": 157215272,
      "step": 9600,
      "train_runtime": 18461.1302,
      "train_tokens_per_second": 8516.016
    },
    {
      "epoch": 0.646714247239429,
      "grad_norm": 1.1776271650762442,
      "learning_rate": 7.63491081113347e-06,
      "loss": 0.4156344890594482,
      "num_input_tokens_seen": 157297192,
      "step": 9605,
      "train_runtime": 18470.6356,
      "train_tokens_per_second": 8516.068
    },
    {
      "epoch": 0.6470509022353892,
      "grad_norm": 1.0260960798672805,
      "learning_rate": 7.632663299939718e-06,
      "loss": 0.4437431812286377,
      "num_input_tokens_seen": 157379112,
      "step": 9610,
      "train_runtime": 18480.2459,
      "train_tokens_per_second": 8516.072
    },
    {
      "epoch": 0.6473875572313493,
      "grad_norm": 1.1345411507267529,
      "learning_rate": 7.630415052530066e-06,
      "loss": 0.41757974624633787,
      "num_input_tokens_seen": 157461032,
      "step": 9615,
      "train_runtime": 18489.8733,
      "train_tokens_per_second": 8516.069
    },
    {
      "epoch": 0.6477242122273095,
      "grad_norm": 1.0642518216299535,
      "learning_rate": 7.628166069533235e-06,
      "loss": 0.43201661109924316,
      "num_input_tokens_seen": 157542952,
      "step": 9620,
      "train_runtime": 18499.3788,
      "train_tokens_per_second": 8516.121
    },
    {
      "epoch": 0.6480608672232696,
      "grad_norm": 1.1379969054011267,
      "learning_rate": 7.625916351578141e-06,
      "loss": 0.43998088836669924,
      "num_input_tokens_seen": 157624872,
      "step": 9625,
      "train_runtime": 18508.9877,
      "train_tokens_per_second": 8516.126
    },
    {
      "epoch": 0.6483975222192297,
      "grad_norm": 0.9769525826230402,
      "learning_rate": 7.623665899293914e-06,
      "loss": 0.38914778232574465,
      "num_input_tokens_seen": 157706792,
      "step": 9630,
      "train_runtime": 18518.6039,
      "train_tokens_per_second": 8516.128
    },
    {
      "epoch": 0.6487341772151899,
      "grad_norm": 0.9608250007943671,
      "learning_rate": 7.621414713309885e-06,
      "loss": 0.394182825088501,
      "num_input_tokens_seen": 157788712,
      "step": 9635,
      "train_runtime": 18528.1037,
      "train_tokens_per_second": 8516.182
    },
    {
      "epoch": 0.64907083221115,
      "grad_norm": 1.0314140109566232,
      "learning_rate": 7.61916279425559e-06,
      "loss": 0.40618209838867186,
      "num_input_tokens_seen": 157870632,
      "step": 9640,
      "train_runtime": 18537.6887,
      "train_tokens_per_second": 8516.198
    },
    {
      "epoch": 0.6494074872071102,
      "grad_norm": 1.0716346228065239,
      "learning_rate": 7.6169101427607716e-06,
      "loss": 0.43238515853881837,
      "num_input_tokens_seen": 157952552,
      "step": 9645,
      "train_runtime": 18547.3014,
      "train_tokens_per_second": 8516.201
    },
    {
      "epoch": 0.6497441422030703,
      "grad_norm": 1.111423204995783,
      "learning_rate": 7.614656759455375e-06,
      "loss": 0.3832079887390137,
      "num_input_tokens_seen": 158034472,
      "step": 9650,
      "train_runtime": 18556.806,
      "train_tokens_per_second": 8516.254
    },
    {
      "epoch": 0.6500807971990304,
      "grad_norm": 1.0388405246293877,
      "learning_rate": 7.612402644969555e-06,
      "loss": 0.40001716613769533,
      "num_input_tokens_seen": 158116392,
      "step": 9655,
      "train_runtime": 18566.413,
      "train_tokens_per_second": 8516.26
    },
    {
      "epoch": 0.6504174521949906,
      "grad_norm": 1.2470041644931518,
      "learning_rate": 7.6101477999336646e-06,
      "loss": 0.442031192779541,
      "num_input_tokens_seen": 158198312,
      "step": 9660,
      "train_runtime": 18576.0199,
      "train_tokens_per_second": 8516.265
    },
    {
      "epoch": 0.6507541071909507,
      "grad_norm": 1.1802674783666658,
      "learning_rate": 7.607892224978266e-06,
      "loss": 0.425701904296875,
      "num_input_tokens_seen": 158280232,
      "step": 9665,
      "train_runtime": 18585.525,
      "train_tokens_per_second": 8516.318
    },
    {
      "epoch": 0.6510907621869109,
      "grad_norm": 1.0952025567762906,
      "learning_rate": 7.605635920734122e-06,
      "loss": 0.41988396644592285,
      "num_input_tokens_seen": 158362040,
      "step": 9670,
      "train_runtime": 18595.6641,
      "train_tokens_per_second": 8516.073
    },
    {
      "epoch": 0.651427417182871,
      "grad_norm": 1.1987972483459197,
      "learning_rate": 7.603378887832202e-06,
      "loss": 0.43268723487854005,
      "num_input_tokens_seen": 158443960,
      "step": 9675,
      "train_runtime": 18605.2932,
      "train_tokens_per_second": 8516.069
    },
    {
      "epoch": 0.6517640721788311,
      "grad_norm": 1.0277086605708639,
      "learning_rate": 7.601121126903676e-06,
      "loss": 0.4035487174987793,
      "num_input_tokens_seen": 158525880,
      "step": 9680,
      "train_runtime": 18614.8359,
      "train_tokens_per_second": 8516.104
    },
    {
      "epoch": 0.6521007271747913,
      "grad_norm": 1.6197257638350149,
      "learning_rate": 7.598862638579922e-06,
      "loss": 0.4325672149658203,
      "num_input_tokens_seen": 158607800,
      "step": 9685,
      "train_runtime": 18624.3899,
      "train_tokens_per_second": 8516.134
    },
    {
      "epoch": 0.6524373821707514,
      "grad_norm": 1.031465849926772,
      "learning_rate": 7.5966034234925205e-06,
      "loss": 0.3931559085845947,
      "num_input_tokens_seen": 158689720,
      "step": 9690,
      "train_runtime": 18633.9914,
      "train_tokens_per_second": 8516.142
    },
    {
      "epoch": 0.6527740371667116,
      "grad_norm": 1.1529801638833168,
      "learning_rate": 7.59434348227325e-06,
      "loss": 0.44016275405883787,
      "num_input_tokens_seen": 158771560,
      "step": 9695,
      "train_runtime": 18644.1638,
      "train_tokens_per_second": 8515.885
    },
    {
      "epoch": 0.6531106921626717,
      "grad_norm": 1.1072468775282605,
      "learning_rate": 7.592082815554098e-06,
      "loss": 0.4218442916870117,
      "num_input_tokens_seen": 158853480,
      "step": 9700,
      "train_runtime": 18653.6625,
      "train_tokens_per_second": 8515.94
    },
    {
      "epoch": 0.6534473471586318,
      "grad_norm": 1.2324933080743068,
      "learning_rate": 7.58982142396725e-06,
      "loss": 0.4465160369873047,
      "num_input_tokens_seen": 158935400,
      "step": 9705,
      "train_runtime": 18663.2991,
      "train_tokens_per_second": 8515.933
    },
    {
      "epoch": 0.653784002154592,
      "grad_norm": 1.0826124333650304,
      "learning_rate": 7.5875593081451e-06,
      "loss": 0.389110803604126,
      "num_input_tokens_seen": 159017320,
      "step": 9710,
      "train_runtime": 18672.9017,
      "train_tokens_per_second": 8515.941
    },
    {
      "epoch": 0.6541206571505521,
      "grad_norm": 1.1369390914629667,
      "learning_rate": 7.585296468720243e-06,
      "loss": 0.394818115234375,
      "num_input_tokens_seen": 159099240,
      "step": 9715,
      "train_runtime": 18682.4048,
      "train_tokens_per_second": 8515.994
    },
    {
      "epoch": 0.6544573121465123,
      "grad_norm": 1.7772548721375259,
      "learning_rate": 7.5830329063254705e-06,
      "loss": 0.4297314643859863,
      "num_input_tokens_seen": 159181136,
      "step": 9720,
      "train_runtime": 18692.5554,
      "train_tokens_per_second": 8515.75
    },
    {
      "epoch": 0.6547939671424724,
      "grad_norm": 1.0476953864018645,
      "learning_rate": 7.5807686215937835e-06,
      "loss": 0.4442605018615723,
      "num_input_tokens_seen": 159263056,
      "step": 9725,
      "train_runtime": 18702.1643,
      "train_tokens_per_second": 8515.755
    },
    {
      "epoch": 0.6551306221384325,
      "grad_norm": 0.9160272593827412,
      "learning_rate": 7.578503615158379e-06,
      "loss": 0.3605373382568359,
      "num_input_tokens_seen": 159344976,
      "step": 9730,
      "train_runtime": 18711.6722,
      "train_tokens_per_second": 8515.806
    },
    {
      "epoch": 0.6554672771343927,
      "grad_norm": 1.073676307824627,
      "learning_rate": 7.576237887652662e-06,
      "loss": 0.40599522590637205,
      "num_input_tokens_seen": 159426896,
      "step": 9735,
      "train_runtime": 18721.2926,
      "train_tokens_per_second": 8515.806
    },
    {
      "epoch": 0.6558039321303528,
      "grad_norm": 1.0305189197143523,
      "learning_rate": 7.573971439710235e-06,
      "loss": 0.3936635971069336,
      "num_input_tokens_seen": 159508816,
      "step": 9740,
      "train_runtime": 18730.9045,
      "train_tokens_per_second": 8515.81
    },
    {
      "epoch": 0.656140587126313,
      "grad_norm": 1.1886679959550375,
      "learning_rate": 7.571704271964902e-06,
      "loss": 0.43999757766723635,
      "num_input_tokens_seen": 159590736,
      "step": 9745,
      "train_runtime": 18740.4496,
      "train_tokens_per_second": 8515.844
    },
    {
      "epoch": 0.6564772421222731,
      "grad_norm": 1.1369924832472698,
      "learning_rate": 7.56943638505067e-06,
      "loss": 0.3513887882232666,
      "num_input_tokens_seen": 159672656,
      "step": 9750,
      "train_runtime": 18750.0131,
      "train_tokens_per_second": 8515.869
    },
    {
      "epoch": 0.6568138971182332,
      "grad_norm": 0.9171913368646601,
      "learning_rate": 7.567167779601746e-06,
      "loss": 0.3932427167892456,
      "num_input_tokens_seen": 159754576,
      "step": 9755,
      "train_runtime": 18759.5714,
      "train_tokens_per_second": 8515.897
    },
    {
      "epoch": 0.6571505521141934,
      "grad_norm": 1.0842887216615646,
      "learning_rate": 7.56489845625254e-06,
      "loss": 0.44795818328857423,
      "num_input_tokens_seen": 159836296,
      "step": 9760,
      "train_runtime": 18769.6062,
      "train_tokens_per_second": 8515.698
    },
    {
      "epoch": 0.6574872071101535,
      "grad_norm": 0.9191607140459656,
      "learning_rate": 7.562628415637659e-06,
      "loss": 0.3902191400527954,
      "num_input_tokens_seen": 159918216,
      "step": 9765,
      "train_runtime": 18779.1624,
      "train_tokens_per_second": 8515.727
    },
    {
      "epoch": 0.6578238621061137,
      "grad_norm": 1.3627670460901666,
      "learning_rate": 7.560357658391913e-06,
      "loss": 0.4478104591369629,
      "num_input_tokens_seen": 160000136,
      "step": 9770,
      "train_runtime": 18788.6844,
      "train_tokens_per_second": 8515.771
    },
    {
      "epoch": 0.6581605171020738,
      "grad_norm": 1.0401037481798228,
      "learning_rate": 7.558086185150311e-06,
      "loss": 0.42306928634643554,
      "num_input_tokens_seen": 160081008,
      "step": 9775,
      "train_runtime": 18798.24,
      "train_tokens_per_second": 8515.744
    },
    {
      "epoch": 0.6584971720980339,
      "grad_norm": 1.0096428015792733,
      "learning_rate": 7.555813996548065e-06,
      "loss": 0.42207536697387693,
      "num_input_tokens_seen": 160162928,
      "step": 9780,
      "train_runtime": 18807.825,
      "train_tokens_per_second": 8515.76
    },
    {
      "epoch": 0.6588338270939941,
      "grad_norm": 1.051382663532145,
      "learning_rate": 7.553541093220586e-06,
      "loss": 0.39834449291229246,
      "num_input_tokens_seen": 160244848,
      "step": 9785,
      "train_runtime": 18817.4445,
      "train_tokens_per_second": 8515.76
    },
    {
      "epoch": 0.6591704820899542,
      "grad_norm": 1.0584601357126766,
      "learning_rate": 7.551267475803481e-06,
      "loss": 0.36908607482910155,
      "num_input_tokens_seen": 160326768,
      "step": 9790,
      "train_runtime": 18826.9927,
      "train_tokens_per_second": 8515.793
    },
    {
      "epoch": 0.6595071370859144,
      "grad_norm": 1.1673332518668678,
      "learning_rate": 7.5489931449325635e-06,
      "loss": 0.4411196708679199,
      "num_input_tokens_seen": 160408688,
      "step": 9795,
      "train_runtime": 18836.5684,
      "train_tokens_per_second": 8515.813
    },
    {
      "epoch": 0.6598437920818745,
      "grad_norm": 1.461436923866458,
      "learning_rate": 7.546718101243842e-06,
      "loss": 0.42180609703063965,
      "num_input_tokens_seen": 160490608,
      "step": 9800,
      "train_runtime": 18846.1781,
      "train_tokens_per_second": 8515.817
    },
    {
      "epoch": 0.6601804470778346,
      "grad_norm": 1.0934491283610872,
      "learning_rate": 7.544442345373522e-06,
      "loss": 0.37937636375427247,
      "num_input_tokens_seen": 160572528,
      "step": 9805,
      "train_runtime": 18855.7257,
      "train_tokens_per_second": 8515.85
    },
    {
      "epoch": 0.6605171020737948,
      "grad_norm": 0.9786761497446109,
      "learning_rate": 7.542165877958014e-06,
      "loss": 0.4082324981689453,
      "num_input_tokens_seen": 160654448,
      "step": 9810,
      "train_runtime": 18865.2973,
      "train_tokens_per_second": 8515.872
    },
    {
      "epoch": 0.6608537570697549,
      "grad_norm": 1.2179343084860563,
      "learning_rate": 7.539888699633926e-06,
      "loss": 0.3949910640716553,
      "num_input_tokens_seen": 160736368,
      "step": 9815,
      "train_runtime": 18874.918,
      "train_tokens_per_second": 8515.871
    },
    {
      "epoch": 0.661190412065715,
      "grad_norm": 1.1130936788323582,
      "learning_rate": 7.5376108110380606e-06,
      "loss": 0.3846081256866455,
      "num_input_tokens_seen": 160818288,
      "step": 9820,
      "train_runtime": 18884.5194,
      "train_tokens_per_second": 8515.879
    },
    {
      "epoch": 0.6615270670616752,
      "grad_norm": 0.9148344805122284,
      "learning_rate": 7.535332212807426e-06,
      "loss": 0.35684542655944823,
      "num_input_tokens_seen": 160900208,
      "step": 9825,
      "train_runtime": 18894.0243,
      "train_tokens_per_second": 8515.931
    },
    {
      "epoch": 0.6618637220576353,
      "grad_norm": 1.344716006597158,
      "learning_rate": 7.53305290557922e-06,
      "loss": 0.45001850128173826,
      "num_input_tokens_seen": 160982128,
      "step": 9830,
      "train_runtime": 18903.6398,
      "train_tokens_per_second": 8515.933
    },
    {
      "epoch": 0.6622003770535955,
      "grad_norm": 1.1258598219990568,
      "learning_rate": 7.530772889990848e-06,
      "loss": 0.390595817565918,
      "num_input_tokens_seen": 161064048,
      "step": 9835,
      "train_runtime": 18913.2459,
      "train_tokens_per_second": 8515.939
    },
    {
      "epoch": 0.6625370320495556,
      "grad_norm": 0.9028553388243641,
      "learning_rate": 7.528492166679907e-06,
      "loss": 0.36255745887756347,
      "num_input_tokens_seen": 161145968,
      "step": 9840,
      "train_runtime": 18922.7425,
      "train_tokens_per_second": 8515.994
    },
    {
      "epoch": 0.6628736870455157,
      "grad_norm": 0.8843091840680736,
      "learning_rate": 7.526210736284194e-06,
      "loss": 0.40423145294189455,
      "num_input_tokens_seen": 161227888,
      "step": 9845,
      "train_runtime": 18932.3488,
      "train_tokens_per_second": 8516.0
    },
    {
      "epoch": 0.6632103420414759,
      "grad_norm": 1.041242063915254,
      "learning_rate": 7.523928599441706e-06,
      "loss": 0.4086163520812988,
      "num_input_tokens_seen": 161309808,
      "step": 9850,
      "train_runtime": 18941.9636,
      "train_tokens_per_second": 8516.002
    },
    {
      "epoch": 0.663546997037436,
      "grad_norm": 1.0496380493220838,
      "learning_rate": 7.5216457567906295e-06,
      "loss": 0.4180124282836914,
      "num_input_tokens_seen": 161391728,
      "step": 9855,
      "train_runtime": 18951.4625,
      "train_tokens_per_second": 8516.057
    },
    {
      "epoch": 0.6638836520333962,
      "grad_norm": 1.1777398428553694,
      "learning_rate": 7.519362208969362e-06,
      "loss": 0.4102897644042969,
      "num_input_tokens_seen": 161473648,
      "step": 9860,
      "train_runtime": 18961.0667,
      "train_tokens_per_second": 8516.063
    },
    {
      "epoch": 0.6642203070293563,
      "grad_norm": 0.9449477634092558,
      "learning_rate": 7.5170779566164855e-06,
      "loss": 0.40523743629455566,
      "num_input_tokens_seen": 161555568,
      "step": 9865,
      "train_runtime": 18970.6605,
      "train_tokens_per_second": 8516.075
    },
    {
      "epoch": 0.6645569620253164,
      "grad_norm": 1.1225100958037508,
      "learning_rate": 7.514793000370783e-06,
      "loss": 0.44009833335876464,
      "num_input_tokens_seen": 161637488,
      "step": 9870,
      "train_runtime": 18980.1551,
      "train_tokens_per_second": 8516.131
    },
    {
      "epoch": 0.6648936170212766,
      "grad_norm": 1.0850136463718596,
      "learning_rate": 7.512507340871239e-06,
      "loss": 0.37497453689575194,
      "num_input_tokens_seen": 161719408,
      "step": 9875,
      "train_runtime": 18989.7677,
      "train_tokens_per_second": 8516.134
    },
    {
      "epoch": 0.6652302720172367,
      "grad_norm": 1.2272564474575534,
      "learning_rate": 7.510220978757028e-06,
      "loss": 0.42433929443359375,
      "num_input_tokens_seen": 161801328,
      "step": 9880,
      "train_runtime": 18999.3786,
      "train_tokens_per_second": 8516.138
    },
    {
      "epoch": 0.6655669270131969,
      "grad_norm": 1.136750033895194,
      "learning_rate": 7.507933914667524e-06,
      "loss": 0.4533599853515625,
      "num_input_tokens_seen": 161883248,
      "step": 9885,
      "train_runtime": 19008.8733,
      "train_tokens_per_second": 8516.194
    },
    {
      "epoch": 0.665903582009157,
      "grad_norm": 1.1460985567014694,
      "learning_rate": 7.505646149242299e-06,
      "loss": 0.38213381767272947,
      "num_input_tokens_seen": 161965168,
      "step": 9890,
      "train_runtime": 19018.5162,
      "train_tokens_per_second": 8516.183
    },
    {
      "epoch": 0.6662402370051171,
      "grad_norm": 0.8384950541477503,
      "learning_rate": 7.503357683121119e-06,
      "loss": 0.37324838638305663,
      "num_input_tokens_seen": 162047088,
      "step": 9895,
      "train_runtime": 19028.1388,
      "train_tokens_per_second": 8516.182
    },
    {
      "epoch": 0.6665768920010773,
      "grad_norm": 1.0594561824366993,
      "learning_rate": 7.501068516943944e-06,
      "loss": 0.322078013420105,
      "num_input_tokens_seen": 162129008,
      "step": 9900,
      "train_runtime": 19037.6875,
      "train_tokens_per_second": 8516.213
    },
    {
      "epoch": 0.6669135469970374,
      "grad_norm": 1.0607007787843254,
      "learning_rate": 7.498778651350934e-06,
      "loss": 0.4259199142456055,
      "num_input_tokens_seen": 162210928,
      "step": 9905,
      "train_runtime": 19047.2406,
      "train_tokens_per_second": 8516.243
    },
    {
      "epoch": 0.6672502019929976,
      "grad_norm": 1.125379911485343,
      "learning_rate": 7.496488086982443e-06,
      "loss": 0.3987056970596313,
      "num_input_tokens_seen": 162292848,
      "step": 9910,
      "train_runtime": 19056.8281,
      "train_tokens_per_second": 8516.257
    },
    {
      "epoch": 0.6675868569889577,
      "grad_norm": 1.0809736692928809,
      "learning_rate": 7.494196824479019e-06,
      "loss": 0.38409583568572997,
      "num_input_tokens_seen": 162374768,
      "step": 9915,
      "train_runtime": 19066.3747,
      "train_tokens_per_second": 8516.29
    },
    {
      "epoch": 0.6679235119849178,
      "grad_norm": 1.2697839796545671,
      "learning_rate": 7.491904864481406e-06,
      "loss": 0.4193130016326904,
      "num_input_tokens_seen": 162456688,
      "step": 9920,
      "train_runtime": 19075.936,
      "train_tokens_per_second": 8516.315
    },
    {
      "epoch": 0.668260166980878,
      "grad_norm": 0.9905276874635967,
      "learning_rate": 7.489612207630543e-06,
      "loss": 0.36633591651916503,
      "num_input_tokens_seen": 162538608,
      "step": 9925,
      "train_runtime": 19085.555,
      "train_tokens_per_second": 8516.316
    },
    {
      "epoch": 0.6685968219768381,
      "grad_norm": 0.9533707400078845,
      "learning_rate": 7.487318854567566e-06,
      "loss": 0.416455602645874,
      "num_input_tokens_seen": 162620528,
      "step": 9930,
      "train_runtime": 19095.0996,
      "train_tokens_per_second": 8516.349
    },
    {
      "epoch": 0.6689334769727983,
      "grad_norm": 0.9749932489289406,
      "learning_rate": 7.485024805933803e-06,
      "loss": 0.37098050117492676,
      "num_input_tokens_seen": 162702448,
      "step": 9935,
      "train_runtime": 19104.6617,
      "train_tokens_per_second": 8516.374
    },
    {
      "epoch": 0.6692701319687584,
      "grad_norm": 0.9163163568165569,
      "learning_rate": 7.482730062370776e-06,
      "loss": 0.4047552108764648,
      "num_input_tokens_seen": 162784368,
      "step": 9940,
      "train_runtime": 19114.2586,
      "train_tokens_per_second": 8516.384
    },
    {
      "epoch": 0.6696067869647185,
      "grad_norm": 1.1291451671353538,
      "learning_rate": 7.480434624520205e-06,
      "loss": 0.3750454902648926,
      "num_input_tokens_seen": 162866288,
      "step": 9945,
      "train_runtime": 19123.7963,
      "train_tokens_per_second": 8516.42
    },
    {
      "epoch": 0.6699434419606787,
      "grad_norm": 0.9678967400466107,
      "learning_rate": 7.478138493024001e-06,
      "loss": 0.4617617607116699,
      "num_input_tokens_seen": 162948208,
      "step": 9950,
      "train_runtime": 19133.2925,
      "train_tokens_per_second": 8516.475
    },
    {
      "epoch": 0.6702800969566388,
      "grad_norm": 1.1228654945839975,
      "learning_rate": 7.475841668524268e-06,
      "loss": 0.38391952514648436,
      "num_input_tokens_seen": 163030128,
      "step": 9955,
      "train_runtime": 19142.8916,
      "train_tokens_per_second": 8516.484
    },
    {
      "epoch": 0.670616751952599,
      "grad_norm": 1.0944261232190984,
      "learning_rate": 7.473544151663309e-06,
      "loss": 0.4308178424835205,
      "num_input_tokens_seen": 163111568,
      "step": 9960,
      "train_runtime": 19152.813,
      "train_tokens_per_second": 8516.324
    },
    {
      "epoch": 0.6709534069485591,
      "grad_norm": 1.2032313058890394,
      "learning_rate": 7.471245943083615e-06,
      "loss": 0.4002055644989014,
      "num_input_tokens_seen": 163193488,
      "step": 9965,
      "train_runtime": 19162.3175,
      "train_tokens_per_second": 8516.375
    },
    {
      "epoch": 0.6712900619445192,
      "grad_norm": 1.0152131847348091,
      "learning_rate": 7.468947043427873e-06,
      "loss": 0.36348609924316405,
      "num_input_tokens_seen": 163275408,
      "step": 9970,
      "train_runtime": 19171.9167,
      "train_tokens_per_second": 8516.384
    },
    {
      "epoch": 0.6716267169404794,
      "grad_norm": 0.9957826281425283,
      "learning_rate": 7.466647453338965e-06,
      "loss": 0.4356955051422119,
      "num_input_tokens_seen": 163357328,
      "step": 9975,
      "train_runtime": 19181.553,
      "train_tokens_per_second": 8516.377
    },
    {
      "epoch": 0.6719633719364395,
      "grad_norm": 1.0795205586590546,
      "learning_rate": 7.464347173459958e-06,
      "loss": 0.41832795143127444,
      "num_input_tokens_seen": 163439248,
      "step": 9980,
      "train_runtime": 19191.0588,
      "train_tokens_per_second": 8516.427
    },
    {
      "epoch": 0.6723000269323997,
      "grad_norm": 1.1127985111656755,
      "learning_rate": 7.462046204434126e-06,
      "loss": 0.4127674579620361,
      "num_input_tokens_seen": 163521168,
      "step": 9985,
      "train_runtime": 19200.6668,
      "train_tokens_per_second": 8516.432
    },
    {
      "epoch": 0.6726366819283598,
      "grad_norm": 1.072913425344034,
      "learning_rate": 7.459744546904922e-06,
      "loss": 0.40815162658691406,
      "num_input_tokens_seen": 163603088,
      "step": 9990,
      "train_runtime": 19210.2743,
      "train_tokens_per_second": 8516.437
    },
    {
      "epoch": 0.6729733369243199,
      "grad_norm": 0.8580157854752923,
      "learning_rate": 7.457442201516001e-06,
      "loss": 0.3594981670379639,
      "num_input_tokens_seen": 163685008,
      "step": 9995,
      "train_runtime": 19219.7885,
      "train_tokens_per_second": 8516.483
    },
    {
      "epoch": 0.6733099919202801,
      "grad_norm": 1.0652105178094264,
      "learning_rate": 7.455139168911203e-06,
      "loss": 0.391219425201416,
      "num_input_tokens_seen": 163766752,
      "step": 10000,
      "train_runtime": 19229.8459,
      "train_tokens_per_second": 8516.28
    },
    {
      "epoch": 0.6736466469162402,
      "grad_norm": 1.052093896253516,
      "learning_rate": 7.452835449734564e-06,
      "loss": 0.36627931594848634,
      "num_input_tokens_seen": 163848272,
      "step": 10005,
      "train_runtime": 19239.7815,
      "train_tokens_per_second": 8516.119
    },
    {
      "epoch": 0.6739833019122003,
      "grad_norm": 1.1181106539872676,
      "learning_rate": 7.450531044630315e-06,
      "loss": 0.37569713592529297,
      "num_input_tokens_seen": 163930192,
      "step": 10010,
      "train_runtime": 19249.3362,
      "train_tokens_per_second": 8516.148
    },
    {
      "epoch": 0.6743199569081605,
      "grad_norm": 1.2707983702339278,
      "learning_rate": 7.448225954242874e-06,
      "loss": 0.4048656463623047,
      "num_input_tokens_seen": 164012112,
      "step": 10015,
      "train_runtime": 19258.9178,
      "train_tokens_per_second": 8516.165
    },
    {
      "epoch": 0.6746566119041206,
      "grad_norm": 1.2227726300046098,
      "learning_rate": 7.44592017921685e-06,
      "loss": 0.4059778690338135,
      "num_input_tokens_seen": 164094032,
      "step": 10020,
      "train_runtime": 19268.5298,
      "train_tokens_per_second": 8516.168
    },
    {
      "epoch": 0.6749932669000808,
      "grad_norm": 1.115867781339054,
      "learning_rate": 7.443613720197048e-06,
      "loss": 0.4180468559265137,
      "num_input_tokens_seen": 164175224,
      "step": 10025,
      "train_runtime": 19278.3504,
      "train_tokens_per_second": 8516.041
    },
    {
      "epoch": 0.6753299218960409,
      "grad_norm": 0.9976158144506341,
      "learning_rate": 7.44130657782846e-06,
      "loss": 0.3683424472808838,
      "num_input_tokens_seen": 164257144,
      "step": 10030,
      "train_runtime": 19287.851,
      "train_tokens_per_second": 8516.094
    },
    {
      "epoch": 0.675666576892001,
      "grad_norm": 0.9374216158798054,
      "learning_rate": 7.438998752756274e-06,
      "loss": 0.3873204231262207,
      "num_input_tokens_seen": 164339064,
      "step": 10035,
      "train_runtime": 19297.4684,
      "train_tokens_per_second": 8516.094
    },
    {
      "epoch": 0.6760032318879612,
      "grad_norm": 1.106095964065187,
      "learning_rate": 7.436690245625864e-06,
      "loss": 0.3706430196762085,
      "num_input_tokens_seen": 164420984,
      "step": 10040,
      "train_runtime": 19307.0877,
      "train_tokens_per_second": 8516.095
    },
    {
      "epoch": 0.6763398868839213,
      "grad_norm": 1.126773849018499,
      "learning_rate": 7.434381057082797e-06,
      "loss": 0.37948882579803467,
      "num_input_tokens_seen": 164502904,
      "step": 10045,
      "train_runtime": 19316.5834,
      "train_tokens_per_second": 8516.149
    },
    {
      "epoch": 0.6766765418798815,
      "grad_norm": 0.8937405558280218,
      "learning_rate": 7.4320711877728305e-06,
      "loss": 0.42539610862731936,
      "num_input_tokens_seen": 164584824,
      "step": 10050,
      "train_runtime": 19326.1885,
      "train_tokens_per_second": 8516.155
    },
    {
      "epoch": 0.6770131968758416,
      "grad_norm": 1.013579763884177,
      "learning_rate": 7.429760638341909e-06,
      "loss": 0.3869424104690552,
      "num_input_tokens_seen": 164666744,
      "step": 10055,
      "train_runtime": 19335.755,
      "train_tokens_per_second": 8516.179
    },
    {
      "epoch": 0.6773498518718017,
      "grad_norm": 1.1132111804541909,
      "learning_rate": 7.427449409436176e-06,
      "loss": 0.4527591705322266,
      "num_input_tokens_seen": 164748664,
      "step": 10060,
      "train_runtime": 19345.2512,
      "train_tokens_per_second": 8516.233
    },
    {
      "epoch": 0.6776865068677619,
      "grad_norm": 0.8623578631580415,
      "learning_rate": 7.425137501701955e-06,
      "loss": 0.3562300682067871,
      "num_input_tokens_seen": 164830584,
      "step": 10065,
      "train_runtime": 19354.8663,
      "train_tokens_per_second": 8516.235
    },
    {
      "epoch": 0.678023161863722,
      "grad_norm": 1.2562072713593933,
      "learning_rate": 7.422824915785765e-06,
      "loss": 0.40029144287109375,
      "num_input_tokens_seen": 164912504,
      "step": 10070,
      "train_runtime": 19364.5169,
      "train_tokens_per_second": 8516.221
    },
    {
      "epoch": 0.6783598168596822,
      "grad_norm": 1.3185067935868242,
      "learning_rate": 7.420511652334313e-06,
      "loss": 0.4197396755218506,
      "num_input_tokens_seen": 164994424,
      "step": 10075,
      "train_runtime": 19374.0119,
      "train_tokens_per_second": 8516.276
    },
    {
      "epoch": 0.6786964718556423,
      "grad_norm": 0.9447995430607388,
      "learning_rate": 7.418197711994495e-06,
      "loss": 0.372930645942688,
      "num_input_tokens_seen": 165076344,
      "step": 10080,
      "train_runtime": 19383.6276,
      "train_tokens_per_second": 8516.277
    },
    {
      "epoch": 0.6790331268516024,
      "grad_norm": 1.1864998985263153,
      "learning_rate": 7.4158830954133984e-06,
      "loss": 0.4000699043273926,
      "num_input_tokens_seen": 165158264,
      "step": 10085,
      "train_runtime": 19393.2408,
      "train_tokens_per_second": 8516.28
    },
    {
      "epoch": 0.6793697818475626,
      "grad_norm": 0.9005951099125644,
      "learning_rate": 7.413567803238298e-06,
      "loss": 0.3924149990081787,
      "num_input_tokens_seen": 165239976,
      "step": 10090,
      "train_runtime": 19403.2358,
      "train_tokens_per_second": 8516.104
    },
    {
      "epoch": 0.6797064368435227,
      "grad_norm": 1.103552118499848,
      "learning_rate": 7.411251836116658e-06,
      "loss": 0.40140066146850584,
      "num_input_tokens_seen": 165321896,
      "step": 10095,
      "train_runtime": 19412.7719,
      "train_tokens_per_second": 8516.141
    },
    {
      "epoch": 0.6800430918394829,
      "grad_norm": 0.9822044831666279,
      "learning_rate": 7.408935194696132e-06,
      "loss": 0.40674667358398436,
      "num_input_tokens_seen": 165403816,
      "step": 10100,
      "train_runtime": 19422.378,
      "train_tokens_per_second": 8516.146
    },
    {
      "epoch": 0.680379746835443,
      "grad_norm": 1.2697531914158262,
      "learning_rate": 7.406617879624558e-06,
      "loss": 0.4354417324066162,
      "num_input_tokens_seen": 165485736,
      "step": 10105,
      "train_runtime": 19431.9279,
      "train_tokens_per_second": 8516.177
    },
    {
      "epoch": 0.6807164018314031,
      "grad_norm": 1.0480041506016458,
      "learning_rate": 7.40429989154997e-06,
      "loss": 0.36368584632873535,
      "num_input_tokens_seen": 165567656,
      "step": 10110,
      "train_runtime": 19441.5031,
      "train_tokens_per_second": 8516.196
    },
    {
      "epoch": 0.6810530568273633,
      "grad_norm": 1.0850390552218316,
      "learning_rate": 7.4019812311205826e-06,
      "loss": 0.4043704986572266,
      "num_input_tokens_seen": 165649576,
      "step": 10115,
      "train_runtime": 19451.1102,
      "train_tokens_per_second": 8516.202
    },
    {
      "epoch": 0.6813897118233234,
      "grad_norm": 1.1142331573295354,
      "learning_rate": 7.399661898984805e-06,
      "loss": 0.3943622589111328,
      "num_input_tokens_seen": 165731496,
      "step": 10120,
      "train_runtime": 19460.6593,
      "train_tokens_per_second": 8516.232
    },
    {
      "epoch": 0.6817263668192836,
      "grad_norm": 1.2893236059614748,
      "learning_rate": 7.397341895791228e-06,
      "loss": 0.41268281936645507,
      "num_input_tokens_seen": 165813416,
      "step": 10125,
      "train_runtime": 19470.2338,
      "train_tokens_per_second": 8516.252
    },
    {
      "epoch": 0.6820630218152437,
      "grad_norm": 1.1145016208422915,
      "learning_rate": 7.395021222188634e-06,
      "loss": 0.38657734394073484,
      "num_input_tokens_seen": 165895336,
      "step": 10130,
      "train_runtime": 19479.8742,
      "train_tokens_per_second": 8516.243
    },
    {
      "epoch": 0.6823996768112038,
      "grad_norm": 1.0352799552652159,
      "learning_rate": 7.392699878825993e-06,
      "loss": 0.4016095161437988,
      "num_input_tokens_seen": 165977256,
      "step": 10135,
      "train_runtime": 19489.4208,
      "train_tokens_per_second": 8516.274
    },
    {
      "epoch": 0.682736331807164,
      "grad_norm": 0.9382912590117471,
      "learning_rate": 7.390377866352462e-06,
      "loss": 0.4074245452880859,
      "num_input_tokens_seen": 166058288,
      "step": 10140,
      "train_runtime": 19499.1599,
      "train_tokens_per_second": 8516.177
    },
    {
      "epoch": 0.6830729868031241,
      "grad_norm": 1.3973573544993054,
      "learning_rate": 7.388055185417382e-06,
      "loss": 0.43050189018249513,
      "num_input_tokens_seen": 166140208,
      "step": 10145,
      "train_runtime": 19508.7481,
      "train_tokens_per_second": 8516.19
    },
    {
      "epoch": 0.6834096417990843,
      "grad_norm": 1.138412477796185,
      "learning_rate": 7.3857318366702844e-06,
      "loss": 0.3422327995300293,
      "num_input_tokens_seen": 166222128,
      "step": 10150,
      "train_runtime": 19518.3122,
      "train_tokens_per_second": 8516.214
    },
    {
      "epoch": 0.6837462967950444,
      "grad_norm": 1.061845178637576,
      "learning_rate": 7.383407820760885e-06,
      "loss": 0.394743013381958,
      "num_input_tokens_seen": 166304048,
      "step": 10155,
      "train_runtime": 19527.8755,
      "train_tokens_per_second": 8516.239
    },
    {
      "epoch": 0.6840829517910045,
      "grad_norm": 0.9228103892254267,
      "learning_rate": 7.381083138339089e-06,
      "loss": 0.384051251411438,
      "num_input_tokens_seen": 166385968,
      "step": 10160,
      "train_runtime": 19537.4861,
      "train_tokens_per_second": 8516.242
    },
    {
      "epoch": 0.6844196067869647,
      "grad_norm": 1.020467942222764,
      "learning_rate": 7.378757790054988e-06,
      "loss": 0.42467217445373534,
      "num_input_tokens_seen": 166467888,
      "step": 10165,
      "train_runtime": 19547.0959,
      "train_tokens_per_second": 8516.247
    },
    {
      "epoch": 0.6847562617829248,
      "grad_norm": 1.218696420144293,
      "learning_rate": 7.376431776558855e-06,
      "loss": 0.38878536224365234,
      "num_input_tokens_seen": 166549808,
      "step": 10170,
      "train_runtime": 19556.5958,
      "train_tokens_per_second": 8516.299
    },
    {
      "epoch": 0.685092916778885,
      "grad_norm": 1.154313580453448,
      "learning_rate": 7.3741050985011546e-06,
      "loss": 0.374937105178833,
      "num_input_tokens_seen": 166631728,
      "step": 10175,
      "train_runtime": 19566.2405,
      "train_tokens_per_second": 8516.287
    },
    {
      "epoch": 0.6854295717748451,
      "grad_norm": 0.996763369183028,
      "learning_rate": 7.371777756532531e-06,
      "loss": 0.36196446418762207,
      "num_input_tokens_seen": 166713648,
      "step": 10180,
      "train_runtime": 19575.847,
      "train_tokens_per_second": 8516.293
    },
    {
      "epoch": 0.6857662267708052,
      "grad_norm": 0.9836828452133732,
      "learning_rate": 7.369449751303823e-06,
      "loss": 0.4155407905578613,
      "num_input_tokens_seen": 166795568,
      "step": 10185,
      "train_runtime": 19585.3576,
      "train_tokens_per_second": 8516.34
    },
    {
      "epoch": 0.6861028817667654,
      "grad_norm": 0.9485012762392321,
      "learning_rate": 7.367121083466048e-06,
      "loss": 0.39232032299041747,
      "num_input_tokens_seen": 166877488,
      "step": 10190,
      "train_runtime": 19594.9619,
      "train_tokens_per_second": 8516.347
    },
    {
      "epoch": 0.6864395367627255,
      "grad_norm": 1.110844113456718,
      "learning_rate": 7.364791753670409e-06,
      "loss": 0.41348419189453123,
      "num_input_tokens_seen": 166959408,
      "step": 10195,
      "train_runtime": 19604.5739,
      "train_tokens_per_second": 8516.35
    },
    {
      "epoch": 0.6867761917586857,
      "grad_norm": 1.1152431634467626,
      "learning_rate": 7.362461762568298e-06,
      "loss": 0.3812551975250244,
      "num_input_tokens_seen": 167041328,
      "step": 10200,
      "train_runtime": 19614.0773,
      "train_tokens_per_second": 8516.4
    },
    {
      "epoch": 0.6871128467546458,
      "grad_norm": 0.9840699392056393,
      "learning_rate": 7.360131110811287e-06,
      "loss": 0.3909309864044189,
      "num_input_tokens_seen": 167123248,
      "step": 10205,
      "train_runtime": 19623.6661,
      "train_tokens_per_second": 8516.413
    },
    {
      "epoch": 0.6874495017506059,
      "grad_norm": 1.009660045922163,
      "learning_rate": 7.357799799051138e-06,
      "loss": 0.3347932815551758,
      "num_input_tokens_seen": 167205168,
      "step": 10210,
      "train_runtime": 19633.2068,
      "train_tokens_per_second": 8516.447
    },
    {
      "epoch": 0.6877861567465661,
      "grad_norm": 1.0449687769804958,
      "learning_rate": 7.355467827939794e-06,
      "loss": 0.4061903476715088,
      "num_input_tokens_seen": 167286704,
      "step": 10215,
      "train_runtime": 19642.6968,
      "train_tokens_per_second": 8516.484
    },
    {
      "epoch": 0.6881228117425262,
      "grad_norm": 1.2095340312043719,
      "learning_rate": 7.353135198129382e-06,
      "loss": 0.3881645441055298,
      "num_input_tokens_seen": 167368432,
      "step": 10220,
      "train_runtime": 19652.7712,
      "train_tokens_per_second": 8516.276
    },
    {
      "epoch": 0.6884594667384863,
      "grad_norm": 1.1071065955243937,
      "learning_rate": 7.350801910272216e-06,
      "loss": 0.40590667724609375,
      "num_input_tokens_seen": 167450352,
      "step": 10225,
      "train_runtime": 19662.4021,
      "train_tokens_per_second": 8516.271
    },
    {
      "epoch": 0.6887961217344465,
      "grad_norm": 0.9354156780490739,
      "learning_rate": 7.348467965020791e-06,
      "loss": 0.3701158046722412,
      "num_input_tokens_seen": 167532272,
      "step": 10230,
      "train_runtime": 19671.9515,
      "train_tokens_per_second": 8516.302
    },
    {
      "epoch": 0.6891327767304066,
      "grad_norm": 1.122095440436704,
      "learning_rate": 7.3461333630277885e-06,
      "loss": 0.4011536121368408,
      "num_input_tokens_seen": 167614192,
      "step": 10235,
      "train_runtime": 19681.5019,
      "train_tokens_per_second": 8516.331
    },
    {
      "epoch": 0.6894694317263668,
      "grad_norm": 1.1012852145743517,
      "learning_rate": 7.343798104946074e-06,
      "loss": 0.37251312732696534,
      "num_input_tokens_seen": 167696112,
      "step": 10240,
      "train_runtime": 19691.1161,
      "train_tokens_per_second": 8516.334
    },
    {
      "epoch": 0.6898060867223269,
      "grad_norm": 1.183500188003326,
      "learning_rate": 7.341462191428692e-06,
      "loss": 0.3923861026763916,
      "num_input_tokens_seen": 167778032,
      "step": 10245,
      "train_runtime": 19700.6629,
      "train_tokens_per_second": 8516.365
    },
    {
      "epoch": 0.690142741718287,
      "grad_norm": 0.8990438514536896,
      "learning_rate": 7.339125623128874e-06,
      "loss": 0.41459059715270996,
      "num_input_tokens_seen": 167859952,
      "step": 10250,
      "train_runtime": 19710.2167,
      "train_tokens_per_second": 8516.393
    },
    {
      "epoch": 0.6904793967142472,
      "grad_norm": 1.0271267058181983,
      "learning_rate": 7.336788400700035e-06,
      "loss": 0.3859578609466553,
      "num_input_tokens_seen": 167941872,
      "step": 10255,
      "train_runtime": 19719.8369,
      "train_tokens_per_second": 8516.393
    },
    {
      "epoch": 0.6908160517102074,
      "grad_norm": 1.3422200754549445,
      "learning_rate": 7.334450524795771e-06,
      "loss": 0.4291349411010742,
      "num_input_tokens_seen": 168023792,
      "step": 10260,
      "train_runtime": 19729.3865,
      "train_tokens_per_second": 8516.423
    },
    {
      "epoch": 0.6911527067061676,
      "grad_norm": 0.971732940681263,
      "learning_rate": 7.332111996069863e-06,
      "loss": 0.3716987371444702,
      "num_input_tokens_seen": 168105712,
      "step": 10265,
      "train_runtime": 19738.9367,
      "train_tokens_per_second": 8516.452
    },
    {
      "epoch": 0.6914893617021277,
      "grad_norm": 1.0631544771182955,
      "learning_rate": 7.32977281517627e-06,
      "loss": 0.43066959381103515,
      "num_input_tokens_seen": 168187632,
      "step": 10270,
      "train_runtime": 19748.5361,
      "train_tokens_per_second": 8516.461
    },
    {
      "epoch": 0.6918260166980879,
      "grad_norm": 1.324927699425938,
      "learning_rate": 7.3274329827691406e-06,
      "loss": 0.4307962417602539,
      "num_input_tokens_seen": 168269344,
      "step": 10275,
      "train_runtime": 19758.5361,
      "train_tokens_per_second": 8516.286
    },
    {
      "epoch": 0.692162671694048,
      "grad_norm": 1.2203430620112625,
      "learning_rate": 7.325092499502796e-06,
      "loss": 0.37196190357208253,
      "num_input_tokens_seen": 168351264,
      "step": 10280,
      "train_runtime": 19768.1096,
      "train_tokens_per_second": 8516.306
    },
    {
      "epoch": 0.6924993266900081,
      "grad_norm": 0.9188856513487929,
      "learning_rate": 7.322751366031751e-06,
      "loss": 0.39542901515960693,
      "num_input_tokens_seen": 168433184,
      "step": 10285,
      "train_runtime": 19777.7174,
      "train_tokens_per_second": 8516.311
    },
    {
      "epoch": 0.6928359816859683,
      "grad_norm": 1.0652962439571974,
      "learning_rate": 7.320409583010691e-06,
      "loss": 0.38469386100769043,
      "num_input_tokens_seen": 168515104,
      "step": 10290,
      "train_runtime": 19787.3193,
      "train_tokens_per_second": 8516.318
    },
    {
      "epoch": 0.6931726366819284,
      "grad_norm": 1.0586731910668012,
      "learning_rate": 7.318067151094493e-06,
      "loss": 0.435040807723999,
      "num_input_tokens_seen": 168597024,
      "step": 10295,
      "train_runtime": 19796.822,
      "train_tokens_per_second": 8516.368
    },
    {
      "epoch": 0.6935092916778886,
      "grad_norm": 1.1780234692678095,
      "learning_rate": 7.3157240709382046e-06,
      "loss": 0.466918420791626,
      "num_input_tokens_seen": 168678944,
      "step": 10300,
      "train_runtime": 19806.4488,
      "train_tokens_per_second": 8516.365
    },
    {
      "epoch": 0.6938459466738487,
      "grad_norm": 1.2232596826246003,
      "learning_rate": 7.313380343197066e-06,
      "loss": 0.3635729789733887,
      "num_input_tokens_seen": 168760864,
      "step": 10305,
      "train_runtime": 19816.0605,
      "train_tokens_per_second": 8516.368
    },
    {
      "epoch": 0.6941826016698088,
      "grad_norm": 1.133143571699816,
      "learning_rate": 7.311035968526492e-06,
      "loss": 0.4084640979766846,
      "num_input_tokens_seen": 168842784,
      "step": 10310,
      "train_runtime": 19825.5535,
      "train_tokens_per_second": 8516.422
    },
    {
      "epoch": 0.694519256665769,
      "grad_norm": 0.9438961030374136,
      "learning_rate": 7.308690947582078e-06,
      "loss": 0.43308367729187014,
      "num_input_tokens_seen": 168924704,
      "step": 10315,
      "train_runtime": 19835.1518,
      "train_tokens_per_second": 8516.431
    },
    {
      "epoch": 0.6948559116617291,
      "grad_norm": 0.9550791021429924,
      "learning_rate": 7.306345281019602e-06,
      "loss": 0.3674100399017334,
      "num_input_tokens_seen": 169006624,
      "step": 10320,
      "train_runtime": 19844.7682,
      "train_tokens_per_second": 8516.432
    },
    {
      "epoch": 0.6951925666576892,
      "grad_norm": 1.2673496674155966,
      "learning_rate": 7.303998969495023e-06,
      "loss": 0.4885863780975342,
      "num_input_tokens_seen": 169088544,
      "step": 10325,
      "train_runtime": 19854.2616,
      "train_tokens_per_second": 8516.486
    },
    {
      "epoch": 0.6955292216536494,
      "grad_norm": 1.152432816359767,
      "learning_rate": 7.301652013664479e-06,
      "loss": 0.38690648078918455,
      "num_input_tokens_seen": 169170464,
      "step": 10330,
      "train_runtime": 19863.8822,
      "train_tokens_per_second": 8516.485
    },
    {
      "epoch": 0.6958658766496095,
      "grad_norm": 1.197433908886238,
      "learning_rate": 7.299304414184288e-06,
      "loss": 0.4580343723297119,
      "num_input_tokens_seen": 169252384,
      "step": 10335,
      "train_runtime": 19873.481,
      "train_tokens_per_second": 8516.494
    },
    {
      "epoch": 0.6962025316455697,
      "grad_norm": 1.2015197284044408,
      "learning_rate": 7.296956171710951e-06,
      "loss": 0.37716946601867674,
      "num_input_tokens_seen": 169334304,
      "step": 10340,
      "train_runtime": 19883.0316,
      "train_tokens_per_second": 8516.523
    },
    {
      "epoch": 0.6965391866415298,
      "grad_norm": 1.283407773216707,
      "learning_rate": 7.294607286901146e-06,
      "loss": 0.3880154132843018,
      "num_input_tokens_seen": 169416224,
      "step": 10345,
      "train_runtime": 19892.5887,
      "train_tokens_per_second": 8516.55
    },
    {
      "epoch": 0.69687584163749,
      "grad_norm": 1.055208237155117,
      "learning_rate": 7.292257760411726e-06,
      "loss": 0.4277467727661133,
      "num_input_tokens_seen": 169498144,
      "step": 10350,
      "train_runtime": 19902.215,
      "train_tokens_per_second": 8516.547
    },
    {
      "epoch": 0.6972124966334501,
      "grad_norm": 1.329731587084388,
      "learning_rate": 7.289907592899737e-06,
      "loss": 0.395269775390625,
      "num_input_tokens_seen": 169580064,
      "step": 10355,
      "train_runtime": 19911.7598,
      "train_tokens_per_second": 8516.578
    },
    {
      "epoch": 0.6975491516294102,
      "grad_norm": 1.0465888496635023,
      "learning_rate": 7.287556785022389e-06,
      "loss": 0.4389288902282715,
      "num_input_tokens_seen": 169661984,
      "step": 10360,
      "train_runtime": 19921.3165,
      "train_tokens_per_second": 8516.605
    },
    {
      "epoch": 0.6978858066253704,
      "grad_norm": 1.086133772413678,
      "learning_rate": 7.285205337437082e-06,
      "loss": 0.4118372917175293,
      "num_input_tokens_seen": 169742968,
      "step": 10365,
      "train_runtime": 19931.0867,
      "train_tokens_per_second": 8516.493
    },
    {
      "epoch": 0.6982224616213305,
      "grad_norm": 1.1138772135220185,
      "learning_rate": 7.2828532508013895e-06,
      "loss": 0.40930819511413574,
      "num_input_tokens_seen": 169824888,
      "step": 10370,
      "train_runtime": 19940.6832,
      "train_tokens_per_second": 8516.503
    },
    {
      "epoch": 0.6985591166172906,
      "grad_norm": 1.2764095174853578,
      "learning_rate": 7.2805005257730625e-06,
      "loss": 0.4242231369018555,
      "num_input_tokens_seen": 169906808,
      "step": 10375,
      "train_runtime": 19950.1845,
      "train_tokens_per_second": 8516.553
    },
    {
      "epoch": 0.6988957716132508,
      "grad_norm": 1.14764357697604,
      "learning_rate": 7.278147163010037e-06,
      "loss": 0.4038609504699707,
      "num_input_tokens_seen": 169988728,
      "step": 10380,
      "train_runtime": 19959.8011,
      "train_tokens_per_second": 8516.554
    },
    {
      "epoch": 0.6992324266092109,
      "grad_norm": 1.1302388807604031,
      "learning_rate": 7.275793163170423e-06,
      "loss": 0.42490148544311523,
      "num_input_tokens_seen": 170070648,
      "step": 10385,
      "train_runtime": 19969.3921,
      "train_tokens_per_second": 8516.566
    },
    {
      "epoch": 0.6995690816051711,
      "grad_norm": 1.0466948184100682,
      "learning_rate": 7.273438526912506e-06,
      "loss": 0.4248520851135254,
      "num_input_tokens_seen": 170152568,
      "step": 10390,
      "train_runtime": 19978.8942,
      "train_tokens_per_second": 8516.616
    },
    {
      "epoch": 0.6999057366011312,
      "grad_norm": 0.9798054344815414,
      "learning_rate": 7.271083254894757e-06,
      "loss": 0.35831375122070314,
      "num_input_tokens_seen": 170234488,
      "step": 10395,
      "train_runtime": 19988.439,
      "train_tokens_per_second": 8516.647
    },
    {
      "epoch": 0.7002423915970913,
      "grad_norm": 1.1354585959133892,
      "learning_rate": 7.268727347775815e-06,
      "loss": 0.4088297367095947,
      "num_input_tokens_seen": 170316408,
      "step": 10400,
      "train_runtime": 19998.0449,
      "train_tokens_per_second": 8516.653
    },
    {
      "epoch": 0.7005790465930515,
      "grad_norm": 1.0730180307339765,
      "learning_rate": 7.266370806214506e-06,
      "loss": 0.3371293544769287,
      "num_input_tokens_seen": 170398328,
      "step": 10405,
      "train_runtime": 20007.5463,
      "train_tokens_per_second": 8516.703
    },
    {
      "epoch": 0.7009157015890116,
      "grad_norm": 1.1255821453656438,
      "learning_rate": 7.264013630869827e-06,
      "loss": 0.4306171894073486,
      "num_input_tokens_seen": 170480248,
      "step": 10410,
      "train_runtime": 20017.132,
      "train_tokens_per_second": 8516.717
    },
    {
      "epoch": 0.7012523565849718,
      "grad_norm": 1.1051199588485394,
      "learning_rate": 7.261655822400957e-06,
      "loss": 0.47879829406738283,
      "num_input_tokens_seen": 170562168,
      "step": 10415,
      "train_runtime": 20026.7309,
      "train_tokens_per_second": 8516.725
    },
    {
      "epoch": 0.7015890115809319,
      "grad_norm": 0.9893416131153692,
      "learning_rate": 7.259297381467248e-06,
      "loss": 0.413361120223999,
      "num_input_tokens_seen": 170644088,
      "step": 10420,
      "train_runtime": 20036.2287,
      "train_tokens_per_second": 8516.777
    },
    {
      "epoch": 0.701925666576892,
      "grad_norm": 0.9334519144146592,
      "learning_rate": 7.2569383087282285e-06,
      "loss": 0.35714058876037597,
      "num_input_tokens_seen": 170726008,
      "step": 10425,
      "train_runtime": 20045.8317,
      "train_tokens_per_second": 8516.783
    },
    {
      "epoch": 0.7022623215728522,
      "grad_norm": 1.0635592464754817,
      "learning_rate": 7.25457860484361e-06,
      "loss": 0.3914948463439941,
      "num_input_tokens_seen": 170807928,
      "step": 10430,
      "train_runtime": 20055.4518,
      "train_tokens_per_second": 8516.783
    },
    {
      "epoch": 0.7025989765688123,
      "grad_norm": 1.0181702872624279,
      "learning_rate": 7.252218270473274e-06,
      "loss": 0.4110713958740234,
      "num_input_tokens_seen": 170889848,
      "step": 10435,
      "train_runtime": 20064.9538,
      "train_tokens_per_second": 8516.832
    },
    {
      "epoch": 0.7029356315647725,
      "grad_norm": 1.372040327941894,
      "learning_rate": 7.249857306277279e-06,
      "loss": 0.42636818885803224,
      "num_input_tokens_seen": 170971768,
      "step": 10440,
      "train_runtime": 20074.5568,
      "train_tokens_per_second": 8516.839
    },
    {
      "epoch": 0.7032722865607326,
      "grad_norm": 1.0683567617128988,
      "learning_rate": 7.247495712915864e-06,
      "loss": 0.3851033926010132,
      "num_input_tokens_seen": 171053064,
      "step": 10445,
      "train_runtime": 20084.4364,
      "train_tokens_per_second": 8516.697
    },
    {
      "epoch": 0.7036089415566927,
      "grad_norm": 0.9965401623648756,
      "learning_rate": 7.245133491049439e-06,
      "loss": 0.3999875783920288,
      "num_input_tokens_seen": 171134984,
      "step": 10450,
      "train_runtime": 20093.9937,
      "train_tokens_per_second": 8516.723
    },
    {
      "epoch": 0.7039455965526529,
      "grad_norm": 1.07425004329933,
      "learning_rate": 7.242770641338592e-06,
      "loss": 0.43955841064453127,
      "num_input_tokens_seen": 171216904,
      "step": 10455,
      "train_runtime": 20103.5533,
      "train_tokens_per_second": 8516.748
    },
    {
      "epoch": 0.704282251548613,
      "grad_norm": 0.9311409817063568,
      "learning_rate": 7.240407164444088e-06,
      "loss": 0.3758403301239014,
      "num_input_tokens_seen": 171298824,
      "step": 10460,
      "train_runtime": 20113.0806,
      "train_tokens_per_second": 8516.787
    },
    {
      "epoch": 0.7046189065445732,
      "grad_norm": 1.0850579309931914,
      "learning_rate": 7.238043061026862e-06,
      "loss": 0.36962459087371824,
      "num_input_tokens_seen": 171380744,
      "step": 10465,
      "train_runtime": 20122.6099,
      "train_tokens_per_second": 8516.825
    },
    {
      "epoch": 0.7049555615405333,
      "grad_norm": 0.894197383141231,
      "learning_rate": 7.235678331748034e-06,
      "loss": 0.36409995555877683,
      "num_input_tokens_seen": 171462664,
      "step": 10470,
      "train_runtime": 20132.161,
      "train_tokens_per_second": 8516.853
    },
    {
      "epoch": 0.7052922165364934,
      "grad_norm": 1.0841228325091468,
      "learning_rate": 7.233312977268887e-06,
      "loss": 0.4329547882080078,
      "num_input_tokens_seen": 171544176,
      "step": 10475,
      "train_runtime": 20142.0974,
      "train_tokens_per_second": 8516.699
    },
    {
      "epoch": 0.7056288715324536,
      "grad_norm": 1.0130751925447141,
      "learning_rate": 7.230946998250889e-06,
      "loss": 0.39334104061126707,
      "num_input_tokens_seen": 171626096,
      "step": 10480,
      "train_runtime": 20151.7122,
      "train_tokens_per_second": 8516.7
    },
    {
      "epoch": 0.7059655265284137,
      "grad_norm": 0.9384843481817308,
      "learning_rate": 7.228580395355676e-06,
      "loss": 0.38215012550354005,
      "num_input_tokens_seen": 171708016,
      "step": 10485,
      "train_runtime": 20161.2153,
      "train_tokens_per_second": 8516.749
    },
    {
      "epoch": 0.7063021815243739,
      "grad_norm": 1.2714237312641397,
      "learning_rate": 7.226213169245062e-06,
      "loss": 0.41020913124084474,
      "num_input_tokens_seen": 171789936,
      "step": 10490,
      "train_runtime": 20170.829,
      "train_tokens_per_second": 8516.751
    },
    {
      "epoch": 0.706638836520334,
      "grad_norm": 1.26424653566781,
      "learning_rate": 7.223845320581035e-06,
      "loss": 0.4051228046417236,
      "num_input_tokens_seen": 171871856,
      "step": 10495,
      "train_runtime": 20180.4292,
      "train_tokens_per_second": 8516.759
    },
    {
      "epoch": 0.7069754915162941,
      "grad_norm": 0.9886674928825187,
      "learning_rate": 7.221476850025754e-06,
      "loss": 0.3619550704956055,
      "num_input_tokens_seen": 171953776,
      "step": 10500,
      "train_runtime": 20189.9298,
      "train_tokens_per_second": 8516.809
    },
    {
      "epoch": 0.7073121465122543,
      "grad_norm": 1.069503958515919,
      "learning_rate": 7.219107758241557e-06,
      "loss": 0.4381204128265381,
      "num_input_tokens_seen": 172035696,
      "step": 10505,
      "train_runtime": 20199.8512,
      "train_tokens_per_second": 8516.681
    },
    {
      "epoch": 0.7076488015082144,
      "grad_norm": 1.0966619698514795,
      "learning_rate": 7.2167380458909505e-06,
      "loss": 0.40926761627197267,
      "num_input_tokens_seen": 172117616,
      "step": 10510,
      "train_runtime": 20209.4796,
      "train_tokens_per_second": 8516.677
    },
    {
      "epoch": 0.7079854565041745,
      "grad_norm": 1.1445938192097442,
      "learning_rate": 7.214367713636619e-06,
      "loss": 0.3834277629852295,
      "num_input_tokens_seen": 172199536,
      "step": 10515,
      "train_runtime": 20219.2731,
      "train_tokens_per_second": 8516.604
    },
    {
      "epoch": 0.7083221115001347,
      "grad_norm": 0.8477589394674185,
      "learning_rate": 7.211996762141418e-06,
      "loss": 0.42458286285400393,
      "num_input_tokens_seen": 172281456,
      "step": 10520,
      "train_runtime": 20228.8554,
      "train_tokens_per_second": 8516.619
    },
    {
      "epoch": 0.7086587664960948,
      "grad_norm": 1.0704402722384947,
      "learning_rate": 7.209625192068374e-06,
      "loss": 0.4312246322631836,
      "num_input_tokens_seen": 172363376,
      "step": 10525,
      "train_runtime": 20238.456,
      "train_tokens_per_second": 8516.627
    },
    {
      "epoch": 0.708995421492055,
      "grad_norm": 1.0175109860417115,
      "learning_rate": 7.207253004080692e-06,
      "loss": 0.4084312438964844,
      "num_input_tokens_seen": 172445296,
      "step": 10530,
      "train_runtime": 20247.9778,
      "train_tokens_per_second": 8516.668
    },
    {
      "epoch": 0.7093320764880151,
      "grad_norm": 1.0824818580199291,
      "learning_rate": 7.204880198841746e-06,
      "loss": 0.3840348243713379,
      "num_input_tokens_seen": 172527216,
      "step": 10535,
      "train_runtime": 20257.8673,
      "train_tokens_per_second": 8516.554
    },
    {
      "epoch": 0.7096687314839752,
      "grad_norm": 1.0188940132284618,
      "learning_rate": 7.2025067770150835e-06,
      "loss": 0.4272745609283447,
      "num_input_tokens_seen": 172609136,
      "step": 10540,
      "train_runtime": 20267.4252,
      "train_tokens_per_second": 8516.579
    },
    {
      "epoch": 0.7100053864799354,
      "grad_norm": 1.0275628237695795,
      "learning_rate": 7.2001327392644235e-06,
      "loss": 0.4127820014953613,
      "num_input_tokens_seen": 172691056,
      "step": 10545,
      "train_runtime": 20276.978,
      "train_tokens_per_second": 8516.607
    },
    {
      "epoch": 0.7103420414758955,
      "grad_norm": 1.2413442573827527,
      "learning_rate": 7.197758086253659e-06,
      "loss": 0.4059319496154785,
      "num_input_tokens_seen": 172772976,
      "step": 10550,
      "train_runtime": 20286.535,
      "train_tokens_per_second": 8516.633
    },
    {
      "epoch": 0.7106786964718557,
      "grad_norm": 1.0150808851592195,
      "learning_rate": 7.195382818646854e-06,
      "loss": 0.40938501358032225,
      "num_input_tokens_seen": 172854896,
      "step": 10555,
      "train_runtime": 20296.1537,
      "train_tokens_per_second": 8516.633
    },
    {
      "epoch": 0.7110153514678158,
      "grad_norm": 1.0173422237284926,
      "learning_rate": 7.1930069371082445e-06,
      "loss": 0.37403597831726076,
      "num_input_tokens_seen": 172936816,
      "step": 10560,
      "train_runtime": 20305.698,
      "train_tokens_per_second": 8516.664
    },
    {
      "epoch": 0.7113520064637759,
      "grad_norm": 1.0813039004503675,
      "learning_rate": 7.190630442302238e-06,
      "loss": 0.41136760711669923,
      "num_input_tokens_seen": 173018736,
      "step": 10565,
      "train_runtime": 20315.2596,
      "train_tokens_per_second": 8516.688
    },
    {
      "epoch": 0.7116886614597361,
      "grad_norm": 0.9774617767819177,
      "learning_rate": 7.188253334893414e-06,
      "loss": 0.4100478172302246,
      "num_input_tokens_seen": 173100656,
      "step": 10570,
      "train_runtime": 20324.8185,
      "train_tokens_per_second": 8516.713
    },
    {
      "epoch": 0.7120253164556962,
      "grad_norm": 1.3912908932489256,
      "learning_rate": 7.18587561554652e-06,
      "loss": 0.44077239036560056,
      "num_input_tokens_seen": 173182576,
      "step": 10575,
      "train_runtime": 20334.635,
      "train_tokens_per_second": 8516.631
    },
    {
      "epoch": 0.7123619714516564,
      "grad_norm": 2.2564882440499434,
      "learning_rate": 7.183497284926483e-06,
      "loss": 0.38530967235565183,
      "num_input_tokens_seen": 173264496,
      "step": 10580,
      "train_runtime": 20344.2056,
      "train_tokens_per_second": 8516.651
    },
    {
      "epoch": 0.7126986264476165,
      "grad_norm": 1.0144784191209293,
      "learning_rate": 7.181118343698392e-06,
      "loss": 0.38458685874938964,
      "num_input_tokens_seen": 173346416,
      "step": 10585,
      "train_runtime": 20353.808,
      "train_tokens_per_second": 8516.658
    },
    {
      "epoch": 0.7130352814435766,
      "grad_norm": 1.1664482858193264,
      "learning_rate": 7.178738792527512e-06,
      "loss": 0.4041862964630127,
      "num_input_tokens_seen": 173428336,
      "step": 10590,
      "train_runtime": 20363.3601,
      "train_tokens_per_second": 8516.686
    },
    {
      "epoch": 0.7133719364395368,
      "grad_norm": 1.0161765560596214,
      "learning_rate": 7.176358632079277e-06,
      "loss": 0.3736846446990967,
      "num_input_tokens_seen": 173510256,
      "step": 10595,
      "train_runtime": 20372.9422,
      "train_tokens_per_second": 8516.701
    },
    {
      "epoch": 0.7137085914354969,
      "grad_norm": 1.089577343394882,
      "learning_rate": 7.173977863019288e-06,
      "loss": 0.39903898239135743,
      "num_input_tokens_seen": 173592176,
      "step": 10600,
      "train_runtime": 20382.5695,
      "train_tokens_per_second": 8516.697
    },
    {
      "epoch": 0.7140452464314571,
      "grad_norm": 1.125226238569262,
      "learning_rate": 7.171596486013323e-06,
      "loss": 0.3775098562240601,
      "num_input_tokens_seen": 173674096,
      "step": 10605,
      "train_runtime": 20392.1857,
      "train_tokens_per_second": 8516.698
    },
    {
      "epoch": 0.7143819014274172,
      "grad_norm": 1.2390389852373205,
      "learning_rate": 7.169214501727326e-06,
      "loss": 0.429394006729126,
      "num_input_tokens_seen": 173755464,
      "step": 10610,
      "train_runtime": 20401.9975,
      "train_tokens_per_second": 8516.591
    },
    {
      "epoch": 0.7147185564233773,
      "grad_norm": 0.9755395567146443,
      "learning_rate": 7.16683191082741e-06,
      "loss": 0.40158977508544924,
      "num_input_tokens_seen": 173837384,
      "step": 10615,
      "train_runtime": 20411.6153,
      "train_tokens_per_second": 8516.591
    },
    {
      "epoch": 0.7150552114193375,
      "grad_norm": 1.1345738039975493,
      "learning_rate": 7.16444871397986e-06,
      "loss": 0.39020648002624514,
      "num_input_tokens_seen": 173919304,
      "step": 10620,
      "train_runtime": 20421.2552,
      "train_tokens_per_second": 8516.582
    },
    {
      "epoch": 0.7153918664152976,
      "grad_norm": 1.0018821450264075,
      "learning_rate": 7.162064911851129e-06,
      "loss": 0.3795687437057495,
      "num_input_tokens_seen": 174001224,
      "step": 10625,
      "train_runtime": 20430.7635,
      "train_tokens_per_second": 8516.629
    },
    {
      "epoch": 0.7157285214112578,
      "grad_norm": 0.988526738812784,
      "learning_rate": 7.15968050510784e-06,
      "loss": 0.44234523773193357,
      "num_input_tokens_seen": 174083144,
      "step": 10630,
      "train_runtime": 20440.3328,
      "train_tokens_per_second": 8516.649
    },
    {
      "epoch": 0.7160651764072179,
      "grad_norm": 1.2888096625225616,
      "learning_rate": 7.1572954944167834e-06,
      "loss": 0.39835391044616697,
      "num_input_tokens_seen": 174163536,
      "step": 10635,
      "train_runtime": 20450.0189,
      "train_tokens_per_second": 8516.546
    },
    {
      "epoch": 0.716401831403178,
      "grad_norm": 1.0658240852658636,
      "learning_rate": 7.154909880444922e-06,
      "loss": 0.41911516189575193,
      "num_input_tokens_seen": 174245456,
      "step": 10640,
      "train_runtime": 20459.5727,
      "train_tokens_per_second": 8516.574
    },
    {
      "epoch": 0.7167384863991382,
      "grad_norm": 1.18516956060817,
      "learning_rate": 7.152523663859384e-06,
      "loss": 0.436143684387207,
      "num_input_tokens_seen": 174327376,
      "step": 10645,
      "train_runtime": 20469.1447,
      "train_tokens_per_second": 8516.593
    },
    {
      "epoch": 0.7170751413950983,
      "grad_norm": 1.0029481338689075,
      "learning_rate": 7.150136845327466e-06,
      "loss": 0.38344640731811525,
      "num_input_tokens_seen": 174409296,
      "step": 10650,
      "train_runtime": 20478.7506,
      "train_tokens_per_second": 8516.598
    },
    {
      "epoch": 0.7174117963910585,
      "grad_norm": 1.0462792981828561,
      "learning_rate": 7.147749425516637e-06,
      "loss": 0.3888838291168213,
      "num_input_tokens_seen": 174491216,
      "step": 10655,
      "train_runtime": 20488.3002,
      "train_tokens_per_second": 8516.627
    },
    {
      "epoch": 0.7177484513870186,
      "grad_norm": 1.119434659614455,
      "learning_rate": 7.145361405094529e-06,
      "loss": 0.36071295738220216,
      "num_input_tokens_seen": 174573136,
      "step": 10660,
      "train_runtime": 20497.8678,
      "train_tokens_per_second": 8516.649
    },
    {
      "epoch": 0.7180851063829787,
      "grad_norm": 1.0569325473881566,
      "learning_rate": 7.142972784728945e-06,
      "loss": 0.3468487739562988,
      "num_input_tokens_seen": 174655056,
      "step": 10665,
      "train_runtime": 20507.5124,
      "train_tokens_per_second": 8516.638
    },
    {
      "epoch": 0.7184217613789389,
      "grad_norm": 1.0672647963230213,
      "learning_rate": 7.140583565087856e-06,
      "loss": 0.3403507471084595,
      "num_input_tokens_seen": 174736976,
      "step": 10670,
      "train_runtime": 20517.0724,
      "train_tokens_per_second": 8516.662
    },
    {
      "epoch": 0.718758416374899,
      "grad_norm": 1.0671692996468243,
      "learning_rate": 7.138193746839399e-06,
      "loss": 0.43944511413574217,
      "num_input_tokens_seen": 174818896,
      "step": 10675,
      "train_runtime": 20526.6426,
      "train_tokens_per_second": 8516.682
    },
    {
      "epoch": 0.7190950713708592,
      "grad_norm": 1.4507741008574153,
      "learning_rate": 7.135803330651878e-06,
      "loss": 0.40560355186462405,
      "num_input_tokens_seen": 174900816,
      "step": 10680,
      "train_runtime": 20536.2455,
      "train_tokens_per_second": 8516.689
    },
    {
      "epoch": 0.7194317263668193,
      "grad_norm": 1.0993286120477532,
      "learning_rate": 7.133412317193766e-06,
      "loss": 0.3760250091552734,
      "num_input_tokens_seen": 174982736,
      "step": 10685,
      "train_runtime": 20545.8726,
      "train_tokens_per_second": 8516.686
    },
    {
      "epoch": 0.7197683813627794,
      "grad_norm": 1.2344155179558376,
      "learning_rate": 7.131020707133703e-06,
      "loss": 0.369489860534668,
      "num_input_tokens_seen": 175064384,
      "step": 10690,
      "train_runtime": 20555.8006,
      "train_tokens_per_second": 8516.544
    },
    {
      "epoch": 0.7201050363587396,
      "grad_norm": 1.006711623377276,
      "learning_rate": 7.128628501140495e-06,
      "loss": 0.4130581855773926,
      "num_input_tokens_seen": 175146304,
      "step": 10695,
      "train_runtime": 20565.4229,
      "train_tokens_per_second": 8516.543
    },
    {
      "epoch": 0.7204416913546997,
      "grad_norm": 0.8759623163857184,
      "learning_rate": 7.126235699883115e-06,
      "loss": 0.3552953004837036,
      "num_input_tokens_seen": 175228224,
      "step": 10700,
      "train_runtime": 20575.0348,
      "train_tokens_per_second": 8516.546
    },
    {
      "epoch": 0.7207783463506598,
      "grad_norm": 1.0320349038137662,
      "learning_rate": 7.1238423040306995e-06,
      "loss": 0.3743264198303223,
      "num_input_tokens_seen": 175310144,
      "step": 10705,
      "train_runtime": 20584.5336,
      "train_tokens_per_second": 8516.595
    },
    {
      "epoch": 0.72111500134662,
      "grad_norm": 1.074651305101698,
      "learning_rate": 7.121448314252557e-06,
      "loss": 0.33684673309326174,
      "num_input_tokens_seen": 175392064,
      "step": 10710,
      "train_runtime": 20594.1768,
      "train_tokens_per_second": 8516.585
    },
    {
      "epoch": 0.7214516563425801,
      "grad_norm": 1.0731280938370533,
      "learning_rate": 7.1190537312181586e-06,
      "loss": 0.4172016143798828,
      "num_input_tokens_seen": 175473976,
      "step": 10715,
      "train_runtime": 20603.7716,
      "train_tokens_per_second": 8516.595
    },
    {
      "epoch": 0.7217883113385403,
      "grad_norm": 1.1655825226886842,
      "learning_rate": 7.116658555597141e-06,
      "loss": 0.37480745315551756,
      "num_input_tokens_seen": 175555896,
      "step": 10720,
      "train_runtime": 20613.2774,
      "train_tokens_per_second": 8516.642
    },
    {
      "epoch": 0.7221249663345004,
      "grad_norm": 0.9780827611098063,
      "learning_rate": 7.114262788059308e-06,
      "loss": 0.3979971885681152,
      "num_input_tokens_seen": 175637816,
      "step": 10725,
      "train_runtime": 20622.8714,
      "train_tokens_per_second": 8516.652
    },
    {
      "epoch": 0.7224616213304605,
      "grad_norm": 1.047918639269207,
      "learning_rate": 7.111866429274627e-06,
      "loss": 0.4022343635559082,
      "num_input_tokens_seen": 175719320,
      "step": 10730,
      "train_runtime": 20632.8464,
      "train_tokens_per_second": 8516.485
    },
    {
      "epoch": 0.7227982763264207,
      "grad_norm": 1.190874858668769,
      "learning_rate": 7.1094694799132356e-06,
      "loss": 0.4546461582183838,
      "num_input_tokens_seen": 175801240,
      "step": 10735,
      "train_runtime": 20642.3447,
      "train_tokens_per_second": 8516.534
    },
    {
      "epoch": 0.7231349313223808,
      "grad_norm": 1.353126819882113,
      "learning_rate": 7.10707194064543e-06,
      "loss": 0.37216806411743164,
      "num_input_tokens_seen": 175883160,
      "step": 10740,
      "train_runtime": 20651.9717,
      "train_tokens_per_second": 8516.531
    },
    {
      "epoch": 0.723471586318341,
      "grad_norm": 1.1493588516754887,
      "learning_rate": 7.104673812141676e-06,
      "loss": 0.4297636985778809,
      "num_input_tokens_seen": 175965080,
      "step": 10745,
      "train_runtime": 20661.5713,
      "train_tokens_per_second": 8516.539
    },
    {
      "epoch": 0.7238082413143011,
      "grad_norm": 1.223683884949917,
      "learning_rate": 7.102275095072601e-06,
      "loss": 0.41290645599365233,
      "num_input_tokens_seen": 176047000,
      "step": 10750,
      "train_runtime": 20671.0722,
      "train_tokens_per_second": 8516.588
    },
    {
      "epoch": 0.7241448963102612,
      "grad_norm": 1.0148811089990637,
      "learning_rate": 7.099875790108998e-06,
      "loss": 0.37735011577606203,
      "num_input_tokens_seen": 176128920,
      "step": 10755,
      "train_runtime": 20680.7067,
      "train_tokens_per_second": 8516.581
    },
    {
      "epoch": 0.7244815513062214,
      "grad_norm": 0.9862641675288043,
      "learning_rate": 7.097475897921827e-06,
      "loss": 0.39106340408325196,
      "num_input_tokens_seen": 176210840,
      "step": 10760,
      "train_runtime": 20690.319,
      "train_tokens_per_second": 8516.584
    },
    {
      "epoch": 0.7248182063021815,
      "grad_norm": 0.9064052766034553,
      "learning_rate": 7.09507541918221e-06,
      "loss": 0.427521800994873,
      "num_input_tokens_seen": 176292760,
      "step": 10765,
      "train_runtime": 20699.8625,
      "train_tokens_per_second": 8516.615
    },
    {
      "epoch": 0.7251548612981417,
      "grad_norm": 0.8455500035470638,
      "learning_rate": 7.092674354561433e-06,
      "loss": 0.427744197845459,
      "num_input_tokens_seen": 176374680,
      "step": 10770,
      "train_runtime": 20709.4311,
      "train_tokens_per_second": 8516.636
    },
    {
      "epoch": 0.7254915162941018,
      "grad_norm": 1.0480778176267576,
      "learning_rate": 7.090272704730945e-06,
      "loss": 0.372967267036438,
      "num_input_tokens_seen": 176456600,
      "step": 10775,
      "train_runtime": 20719.0348,
      "train_tokens_per_second": 8516.642
    },
    {
      "epoch": 0.7258281712900619,
      "grad_norm": 1.0761961301468481,
      "learning_rate": 7.08787047036236e-06,
      "loss": 0.3981212615966797,
      "num_input_tokens_seen": 176538520,
      "step": 10780,
      "train_runtime": 20728.5501,
      "train_tokens_per_second": 8516.684
    },
    {
      "epoch": 0.7261648262860221,
      "grad_norm": 1.0290301407557079,
      "learning_rate": 7.085467652127457e-06,
      "loss": 0.4061272144317627,
      "num_input_tokens_seen": 176620440,
      "step": 10785,
      "train_runtime": 20738.1059,
      "train_tokens_per_second": 8516.71
    },
    {
      "epoch": 0.7265014812819822,
      "grad_norm": 1.1670600670602052,
      "learning_rate": 7.083064250698175e-06,
      "loss": 0.37153077125549316,
      "num_input_tokens_seen": 176702360,
      "step": 10790,
      "train_runtime": 20747.7171,
      "train_tokens_per_second": 8516.713
    },
    {
      "epoch": 0.7268381362779424,
      "grad_norm": 1.052667098961124,
      "learning_rate": 7.0806602667466175e-06,
      "loss": 0.44872140884399414,
      "num_input_tokens_seen": 176784280,
      "step": 10795,
      "train_runtime": 20757.2677,
      "train_tokens_per_second": 8516.741
    },
    {
      "epoch": 0.7271747912739025,
      "grad_norm": 1.1478618286538798,
      "learning_rate": 7.078255700945051e-06,
      "loss": 0.42076454162597654,
      "num_input_tokens_seen": 176864912,
      "step": 10800,
      "train_runtime": 20766.9568,
      "train_tokens_per_second": 8516.65
    },
    {
      "epoch": 0.7275114462698626,
      "grad_norm": 1.2566990602829138,
      "learning_rate": 7.075850553965904e-06,
      "loss": 0.3993297338485718,
      "num_input_tokens_seen": 176946640,
      "step": 10805,
      "train_runtime": 20777.063,
      "train_tokens_per_second": 8516.441
    },
    {
      "epoch": 0.7278481012658228,
      "grad_norm": 1.0549586462452294,
      "learning_rate": 7.07344482648177e-06,
      "loss": 0.41957731246948243,
      "num_input_tokens_seen": 177028560,
      "step": 10810,
      "train_runtime": 20786.6637,
      "train_tokens_per_second": 8516.449
    },
    {
      "epoch": 0.7281847562617829,
      "grad_norm": 1.104456448068623,
      "learning_rate": 7.071038519165402e-06,
      "loss": 0.4161380290985107,
      "num_input_tokens_seen": 177110480,
      "step": 10815,
      "train_runtime": 20796.1693,
      "train_tokens_per_second": 8516.495
    },
    {
      "epoch": 0.7285214112577431,
      "grad_norm": 1.059503951418979,
      "learning_rate": 7.068631632689716e-06,
      "loss": 0.4034508228302002,
      "num_input_tokens_seen": 177192400,
      "step": 10820,
      "train_runtime": 20805.789,
      "train_tokens_per_second": 8516.495
    },
    {
      "epoch": 0.7288580662537032,
      "grad_norm": 1.1482220523633626,
      "learning_rate": 7.066224167727791e-06,
      "loss": 0.40539236068725587,
      "num_input_tokens_seen": 177274320,
      "step": 10825,
      "train_runtime": 20815.4179,
      "train_tokens_per_second": 8516.491
    },
    {
      "epoch": 0.7291947212496633,
      "grad_norm": 1.0516293481668888,
      "learning_rate": 7.063816124952865e-06,
      "loss": 0.3762645721435547,
      "num_input_tokens_seen": 177356240,
      "step": 10830,
      "train_runtime": 20824.9211,
      "train_tokens_per_second": 8516.538
    },
    {
      "epoch": 0.7295313762456235,
      "grad_norm": 0.9575637370149406,
      "learning_rate": 7.061407505038341e-06,
      "loss": 0.39408111572265625,
      "num_input_tokens_seen": 177438160,
      "step": 10835,
      "train_runtime": 20834.5327,
      "train_tokens_per_second": 8516.541
    },
    {
      "epoch": 0.7298680312415836,
      "grad_norm": 1.162650171189095,
      "learning_rate": 7.058998308657782e-06,
      "loss": 0.43257627487182615,
      "num_input_tokens_seen": 177520080,
      "step": 10840,
      "train_runtime": 20844.1404,
      "train_tokens_per_second": 8516.546
    },
    {
      "epoch": 0.7302046862375438,
      "grad_norm": 0.9295042795444296,
      "learning_rate": 7.056588536484912e-06,
      "loss": 0.34402098655700686,
      "num_input_tokens_seen": 177600808,
      "step": 10845,
      "train_runtime": 20853.7652,
      "train_tokens_per_second": 8516.486
    },
    {
      "epoch": 0.7305413412335039,
      "grad_norm": 0.9311203933091854,
      "learning_rate": 7.0541781891936145e-06,
      "loss": 0.4106177806854248,
      "num_input_tokens_seen": 177682728,
      "step": 10850,
      "train_runtime": 20863.3366,
      "train_tokens_per_second": 8516.506
    },
    {
      "epoch": 0.730877996229464,
      "grad_norm": 0.9743305120187018,
      "learning_rate": 7.0517672674579364e-06,
      "loss": 0.40253429412841796,
      "num_input_tokens_seen": 177764648,
      "step": 10855,
      "train_runtime": 20872.9506,
      "train_tokens_per_second": 8516.508
    },
    {
      "epoch": 0.7312146512254242,
      "grad_norm": 1.0536700586321752,
      "learning_rate": 7.049355771952087e-06,
      "loss": 0.4018121242523193,
      "num_input_tokens_seen": 177846568,
      "step": 10860,
      "train_runtime": 20882.4518,
      "train_tokens_per_second": 8516.556
    },
    {
      "epoch": 0.7315513062213843,
      "grad_norm": 1.0909960759814543,
      "learning_rate": 7.046943703350429e-06,
      "loss": 0.4162801742553711,
      "num_input_tokens_seen": 177928488,
      "step": 10865,
      "train_runtime": 20892.0777,
      "train_tokens_per_second": 8516.553
    },
    {
      "epoch": 0.7318879612173445,
      "grad_norm": 1.1693874811616123,
      "learning_rate": 7.044531062327492e-06,
      "loss": 0.39851644039154055,
      "num_input_tokens_seen": 178010408,
      "step": 10870,
      "train_runtime": 20901.6996,
      "train_tokens_per_second": 8516.552
    },
    {
      "epoch": 0.7322246162133046,
      "grad_norm": 0.9574135092947466,
      "learning_rate": 7.042117849557963e-06,
      "loss": 0.40190773010253905,
      "num_input_tokens_seen": 178092328,
      "step": 10875,
      "train_runtime": 20911.2061,
      "train_tokens_per_second": 8516.598
    },
    {
      "epoch": 0.7325612712092647,
      "grad_norm": 0.8503834763541102,
      "learning_rate": 7.0397040657166885e-06,
      "loss": 0.40671329498291015,
      "num_input_tokens_seen": 178174248,
      "step": 10880,
      "train_runtime": 20920.8111,
      "train_tokens_per_second": 8516.603
    },
    {
      "epoch": 0.7328979262052249,
      "grad_norm": 1.1018587319805981,
      "learning_rate": 7.037289711478676e-06,
      "loss": 0.3685450077056885,
      "num_input_tokens_seen": 178256168,
      "step": 10885,
      "train_runtime": 20930.4301,
      "train_tokens_per_second": 8516.603
    },
    {
      "epoch": 0.733234581201185,
      "grad_norm": 1.0152937042857217,
      "learning_rate": 7.034874787519095e-06,
      "loss": 0.39684937000274656,
      "num_input_tokens_seen": 178338088,
      "step": 10890,
      "train_runtime": 20939.9251,
      "train_tokens_per_second": 8516.654
    },
    {
      "epoch": 0.7335712361971451,
      "grad_norm": 1.0720561478399686,
      "learning_rate": 7.0324592945132655e-06,
      "loss": 0.42624850273132325,
      "num_input_tokens_seen": 178420008,
      "step": 10895,
      "train_runtime": 20949.5583,
      "train_tokens_per_second": 8516.648
    },
    {
      "epoch": 0.7339078911931053,
      "grad_norm": 1.1155727499070318,
      "learning_rate": 7.0300432331366764e-06,
      "loss": 0.40452003479003906,
      "num_input_tokens_seen": 178501928,
      "step": 10900,
      "train_runtime": 20959.1894,
      "train_tokens_per_second": 8516.643
    },
    {
      "epoch": 0.7342445461890654,
      "grad_norm": 1.0125728326703876,
      "learning_rate": 7.02762660406497e-06,
      "loss": 0.38294358253479005,
      "num_input_tokens_seen": 178583848,
      "step": 10905,
      "train_runtime": 20968.7507,
      "train_tokens_per_second": 8516.666
    },
    {
      "epoch": 0.7345812011850256,
      "grad_norm": 1.1233878652516085,
      "learning_rate": 7.02520940797395e-06,
      "loss": 0.41587486267089846,
      "num_input_tokens_seen": 178665768,
      "step": 10910,
      "train_runtime": 20978.3192,
      "train_tokens_per_second": 8516.686
    },
    {
      "epoch": 0.7349178561809857,
      "grad_norm": 1.1516346206328725,
      "learning_rate": 7.022791645539576e-06,
      "loss": 0.39545483589172364,
      "num_input_tokens_seen": 178747688,
      "step": 10915,
      "train_runtime": 20987.9404,
      "train_tokens_per_second": 8516.686
    },
    {
      "epoch": 0.7352545111769458,
      "grad_norm": 1.151693664659927,
      "learning_rate": 7.02037331743797e-06,
      "loss": 0.41542668342590333,
      "num_input_tokens_seen": 178829608,
      "step": 10920,
      "train_runtime": 20997.4751,
      "train_tokens_per_second": 8516.72
    },
    {
      "epoch": 0.735591166172906,
      "grad_norm": 1.1509162371729313,
      "learning_rate": 7.0179544243454076e-06,
      "loss": 0.38487093448638915,
      "num_input_tokens_seen": 178911528,
      "step": 10925,
      "train_runtime": 21007.0556,
      "train_tokens_per_second": 8516.735
    },
    {
      "epoch": 0.7359278211688661,
      "grad_norm": 1.0400063081966264,
      "learning_rate": 7.0155349669383245e-06,
      "loss": 0.4073920249938965,
      "num_input_tokens_seen": 178993448,
      "step": 10930,
      "train_runtime": 21016.6797,
      "train_tokens_per_second": 8516.733
    },
    {
      "epoch": 0.7362644761648263,
      "grad_norm": 1.0354500637621131,
      "learning_rate": 7.013114945893316e-06,
      "loss": 0.38936572074890136,
      "num_input_tokens_seen": 179075368,
      "step": 10935,
      "train_runtime": 21026.2505,
      "train_tokens_per_second": 8516.752
    },
    {
      "epoch": 0.7366011311607864,
      "grad_norm": 0.9295277762295584,
      "learning_rate": 7.010694361887131e-06,
      "loss": 0.351425838470459,
      "num_input_tokens_seen": 179157288,
      "step": 10940,
      "train_runtime": 21035.8332,
      "train_tokens_per_second": 8516.767
    },
    {
      "epoch": 0.7369377861567465,
      "grad_norm": 0.9766085213373289,
      "learning_rate": 7.008273215596681e-06,
      "loss": 0.38548195362091064,
      "num_input_tokens_seen": 179239208,
      "step": 10945,
      "train_runtime": 21045.4393,
      "train_tokens_per_second": 8516.772
    },
    {
      "epoch": 0.7372744411527067,
      "grad_norm": 0.9579994115994903,
      "learning_rate": 7.005851507699029e-06,
      "loss": 0.386813759803772,
      "num_input_tokens_seen": 179321128,
      "step": 10950,
      "train_runtime": 21055.0371,
      "train_tokens_per_second": 8516.78
    },
    {
      "epoch": 0.7376110961486668,
      "grad_norm": 1.0017813992371605,
      "learning_rate": 7.003429238871399e-06,
      "loss": 0.3709810018539429,
      "num_input_tokens_seen": 179403048,
      "step": 10955,
      "train_runtime": 21064.5389,
      "train_tokens_per_second": 8516.828
    },
    {
      "epoch": 0.737947751144627,
      "grad_norm": 1.0629593737017942,
      "learning_rate": 7.001006409791171e-06,
      "loss": 0.3967580318450928,
      "num_input_tokens_seen": 179484968,
      "step": 10960,
      "train_runtime": 21074.1722,
      "train_tokens_per_second": 8516.822
    },
    {
      "epoch": 0.7382844061405871,
      "grad_norm": 1.1066089809880353,
      "learning_rate": 6.998583021135882e-06,
      "loss": 0.409555721282959,
      "num_input_tokens_seen": 179566888,
      "step": 10965,
      "train_runtime": 21083.8012,
      "train_tokens_per_second": 8516.818
    },
    {
      "epoch": 0.7386210611365472,
      "grad_norm": 1.1499199960831825,
      "learning_rate": 6.996159073583224e-06,
      "loss": 0.4165364742279053,
      "num_input_tokens_seen": 179648808,
      "step": 10970,
      "train_runtime": 21093.308,
      "train_tokens_per_second": 8516.863
    },
    {
      "epoch": 0.7389577161325074,
      "grad_norm": 1.0549499799421806,
      "learning_rate": 6.993734567811046e-06,
      "loss": 0.3677542209625244,
      "num_input_tokens_seen": 179730728,
      "step": 10975,
      "train_runtime": 21102.91,
      "train_tokens_per_second": 8516.869
    },
    {
      "epoch": 0.7392943711284675,
      "grad_norm": 1.04377854423962,
      "learning_rate": 6.991309504497352e-06,
      "loss": 0.39058537483215333,
      "num_input_tokens_seen": 179812648,
      "step": 10980,
      "train_runtime": 21112.5688,
      "train_tokens_per_second": 8516.853
    },
    {
      "epoch": 0.7396310261244277,
      "grad_norm": 1.1943721698443648,
      "learning_rate": 6.988883884320307e-06,
      "loss": 0.38251638412475586,
      "num_input_tokens_seen": 179894568,
      "step": 10985,
      "train_runtime": 21122.0742,
      "train_tokens_per_second": 8516.899
    },
    {
      "epoch": 0.7399676811203878,
      "grad_norm": 1.0422918612554235,
      "learning_rate": 6.9864577079582254e-06,
      "loss": 0.3862327575683594,
      "num_input_tokens_seen": 179976488,
      "step": 10990,
      "train_runtime": 21131.7013,
      "train_tokens_per_second": 8516.895
    },
    {
      "epoch": 0.7403043361163479,
      "grad_norm": 0.9084879058354597,
      "learning_rate": 6.984030976089581e-06,
      "loss": 0.3672579288482666,
      "num_input_tokens_seen": 180058408,
      "step": 10995,
      "train_runtime": 21141.3331,
      "train_tokens_per_second": 8516.89
    },
    {
      "epoch": 0.7406409911123081,
      "grad_norm": 1.1111331703720608,
      "learning_rate": 6.981603689392999e-06,
      "loss": 0.41779794692993166,
      "num_input_tokens_seen": 180140328,
      "step": 11000,
      "train_runtime": 21150.8479,
      "train_tokens_per_second": 8516.932
    },
    {
      "epoch": 0.7409776461082682,
      "grad_norm": 0.8976392905929486,
      "learning_rate": 6.979175848547263e-06,
      "loss": 0.3706032752990723,
      "num_input_tokens_seen": 180222248,
      "step": 11005,
      "train_runtime": 21160.4353,
      "train_tokens_per_second": 8516.944
    },
    {
      "epoch": 0.7413143011042284,
      "grad_norm": 1.1204236464922965,
      "learning_rate": 6.976747454231316e-06,
      "loss": 0.36915740966796873,
      "num_input_tokens_seen": 180304168,
      "step": 11010,
      "train_runtime": 21170.0522,
      "train_tokens_per_second": 8516.945
    },
    {
      "epoch": 0.7416509561001885,
      "grad_norm": 0.9989916254889086,
      "learning_rate": 6.974318507124245e-06,
      "loss": 0.41425180435180664,
      "num_input_tokens_seen": 180386088,
      "step": 11015,
      "train_runtime": 21179.6808,
      "train_tokens_per_second": 8516.941
    },
    {
      "epoch": 0.7419876110961486,
      "grad_norm": 1.0401753229498705,
      "learning_rate": 6.971889007905299e-06,
      "loss": 0.402665901184082,
      "num_input_tokens_seen": 180468008,
      "step": 11020,
      "train_runtime": 21189.2249,
      "train_tokens_per_second": 8516.971
    },
    {
      "epoch": 0.7423242660921088,
      "grad_norm": 0.9896771305155188,
      "learning_rate": 6.9694589572538795e-06,
      "loss": 0.413340425491333,
      "num_input_tokens_seen": 180549928,
      "step": 11025,
      "train_runtime": 21198.751,
      "train_tokens_per_second": 8517.008
    },
    {
      "epoch": 0.7426609210880689,
      "grad_norm": 1.0542172285996991,
      "learning_rate": 6.9670283558495436e-06,
      "loss": 0.4146266460418701,
      "num_input_tokens_seen": 180631848,
      "step": 11030,
      "train_runtime": 21208.3648,
      "train_tokens_per_second": 8517.01
    },
    {
      "epoch": 0.742997576084029,
      "grad_norm": 1.3991561408461426,
      "learning_rate": 6.964597204371999e-06,
      "loss": 0.45803089141845704,
      "num_input_tokens_seen": 180713768,
      "step": 11035,
      "train_runtime": 21217.9689,
      "train_tokens_per_second": 8517.015
    },
    {
      "epoch": 0.7433342310799892,
      "grad_norm": 0.976813006597976,
      "learning_rate": 6.962165503501111e-06,
      "loss": 0.34134883880615235,
      "num_input_tokens_seen": 180795688,
      "step": 11040,
      "train_runtime": 21227.5234,
      "train_tokens_per_second": 8517.041
    },
    {
      "epoch": 0.7436708860759493,
      "grad_norm": 1.2760975466076419,
      "learning_rate": 6.959733253916899e-06,
      "loss": 0.4294559955596924,
      "num_input_tokens_seen": 180877608,
      "step": 11045,
      "train_runtime": 21237.1044,
      "train_tokens_per_second": 8517.056
    },
    {
      "epoch": 0.7440075410719095,
      "grad_norm": 1.0376854925812393,
      "learning_rate": 6.95730045629953e-06,
      "loss": 0.40329694747924805,
      "num_input_tokens_seen": 180959528,
      "step": 11050,
      "train_runtime": 21246.7311,
      "train_tokens_per_second": 8517.053
    },
    {
      "epoch": 0.7443441960678696,
      "grad_norm": 1.0906002595564555,
      "learning_rate": 6.954867111329329e-06,
      "loss": 0.3925994873046875,
      "num_input_tokens_seen": 181041448,
      "step": 11055,
      "train_runtime": 21256.3149,
      "train_tokens_per_second": 8517.067
    },
    {
      "epoch": 0.7446808510638298,
      "grad_norm": 1.1073390832526806,
      "learning_rate": 6.952433219686775e-06,
      "loss": 0.4361220359802246,
      "num_input_tokens_seen": 181123368,
      "step": 11060,
      "train_runtime": 21266.0237,
      "train_tokens_per_second": 8517.03
    },
    {
      "epoch": 0.7450175060597899,
      "grad_norm": 1.0892829821513232,
      "learning_rate": 6.949998782052497e-06,
      "loss": 0.4208925724029541,
      "num_input_tokens_seen": 181205288,
      "step": 11065,
      "train_runtime": 21275.6746,
      "train_tokens_per_second": 8517.017
    },
    {
      "epoch": 0.74535416105575,
      "grad_norm": 1.0436505911215161,
      "learning_rate": 6.947563799107277e-06,
      "loss": 0.39055395126342773,
      "num_input_tokens_seen": 181287176,
      "step": 11070,
      "train_runtime": 21285.8313,
      "train_tokens_per_second": 8516.8
    },
    {
      "epoch": 0.7456908160517102,
      "grad_norm": 1.28771089792248,
      "learning_rate": 6.945128271532052e-06,
      "loss": 0.4011195182800293,
      "num_input_tokens_seen": 181369096,
      "step": 11075,
      "train_runtime": 21295.392,
      "train_tokens_per_second": 8516.824
    },
    {
      "epoch": 0.7460274710476703,
      "grad_norm": 0.943588929276615,
      "learning_rate": 6.942692200007908e-06,
      "loss": 0.38066878318786623,
      "num_input_tokens_seen": 181451016,
      "step": 11080,
      "train_runtime": 21304.9797,
      "train_tokens_per_second": 8516.836
    },
    {
      "epoch": 0.7463641260436304,
      "grad_norm": 0.9625066569880572,
      "learning_rate": 6.9402555852160845e-06,
      "loss": 0.40946102142333984,
      "num_input_tokens_seen": 181532936,
      "step": 11085,
      "train_runtime": 21314.5954,
      "train_tokens_per_second": 8516.837
    },
    {
      "epoch": 0.7467007810395906,
      "grad_norm": 0.8554727731419148,
      "learning_rate": 6.937818427837975e-06,
      "loss": 0.4095050811767578,
      "num_input_tokens_seen": 181614856,
      "step": 11090,
      "train_runtime": 21324.2257,
      "train_tokens_per_second": 8516.832
    },
    {
      "epoch": 0.7470374360355507,
      "grad_norm": 1.1090725955962126,
      "learning_rate": 6.93538072855512e-06,
      "loss": 0.42599711418151853,
      "num_input_tokens_seen": 181696776,
      "step": 11095,
      "train_runtime": 21333.7203,
      "train_tokens_per_second": 8516.882
    },
    {
      "epoch": 0.7473740910315109,
      "grad_norm": 0.8780300717432376,
      "learning_rate": 6.932942488049217e-06,
      "loss": 0.34532723426818845,
      "num_input_tokens_seen": 181778696,
      "step": 11100,
      "train_runtime": 21343.3672,
      "train_tokens_per_second": 8516.871
    },
    {
      "epoch": 0.747710746027471,
      "grad_norm": 1.0122250926643954,
      "learning_rate": 6.930503707002108e-06,
      "loss": 0.4079418659210205,
      "num_input_tokens_seen": 181860616,
      "step": 11105,
      "train_runtime": 21353.0005,
      "train_tokens_per_second": 8516.865
    },
    {
      "epoch": 0.7480474010234311,
      "grad_norm": 1.0545792815308168,
      "learning_rate": 6.928064386095796e-06,
      "loss": 0.3950007438659668,
      "num_input_tokens_seen": 181942536,
      "step": 11110,
      "train_runtime": 21362.6304,
      "train_tokens_per_second": 8516.86
    },
    {
      "epoch": 0.7483840560193913,
      "grad_norm": 1.354068152450215,
      "learning_rate": 6.9256245260124264e-06,
      "loss": 0.4601134300231934,
      "num_input_tokens_seen": 182024456,
      "step": 11115,
      "train_runtime": 21372.2902,
      "train_tokens_per_second": 8516.844
    },
    {
      "epoch": 0.7487207110153514,
      "grad_norm": 1.6194070331574681,
      "learning_rate": 6.923184127434298e-06,
      "loss": 0.4049524784088135,
      "num_input_tokens_seen": 182106376,
      "step": 11120,
      "train_runtime": 21382.0552,
      "train_tokens_per_second": 8516.785
    },
    {
      "epoch": 0.7490573660113116,
      "grad_norm": 1.178622410629862,
      "learning_rate": 6.920743191043861e-06,
      "loss": 0.3680035352706909,
      "num_input_tokens_seen": 182188296,
      "step": 11125,
      "train_runtime": 21391.6094,
      "train_tokens_per_second": 8516.811
    },
    {
      "epoch": 0.7493940210072717,
      "grad_norm": 1.038433850412875,
      "learning_rate": 6.918301717523714e-06,
      "loss": 0.42522263526916504,
      "num_input_tokens_seen": 182270216,
      "step": 11130,
      "train_runtime": 21401.1891,
      "train_tokens_per_second": 8516.827
    },
    {
      "epoch": 0.7497306760032318,
      "grad_norm": 1.117417338385132,
      "learning_rate": 6.915859707556611e-06,
      "loss": 0.4231563568115234,
      "num_input_tokens_seen": 182352136,
      "step": 11135,
      "train_runtime": 21410.8128,
      "train_tokens_per_second": 8516.825
    },
    {
      "epoch": 0.750067330999192,
      "grad_norm": 1.2067302544828153,
      "learning_rate": 6.913417161825449e-06,
      "loss": 0.44112396240234375,
      "num_input_tokens_seen": 182434056,
      "step": 11140,
      "train_runtime": 21420.4339,
      "train_tokens_per_second": 8516.824
    },
    {
      "epoch": 0.7504039859951521,
      "grad_norm": 0.9988660891030222,
      "learning_rate": 6.91097408101328e-06,
      "loss": 0.35783097743988035,
      "num_input_tokens_seen": 182515976,
      "step": 11145,
      "train_runtime": 21429.9352,
      "train_tokens_per_second": 8516.87
    },
    {
      "epoch": 0.7507406409911123,
      "grad_norm": 1.2720856275915753,
      "learning_rate": 6.908530465803302e-06,
      "loss": 0.3908069610595703,
      "num_input_tokens_seen": 182597432,
      "step": 11150,
      "train_runtime": 21439.8949,
      "train_tokens_per_second": 8516.713
    },
    {
      "epoch": 0.7510772959870724,
      "grad_norm": 1.2190008979704332,
      "learning_rate": 6.906086316878865e-06,
      "loss": 0.3914205551147461,
      "num_input_tokens_seen": 182679352,
      "step": 11155,
      "train_runtime": 21449.512,
      "train_tokens_per_second": 8516.714
    },
    {
      "epoch": 0.7514139509830325,
      "grad_norm": 0.9625517809919599,
      "learning_rate": 6.903641634923468e-06,
      "loss": 0.3893484354019165,
      "num_input_tokens_seen": 182761272,
      "step": 11160,
      "train_runtime": 21459.1354,
      "train_tokens_per_second": 8516.712
    },
    {
      "epoch": 0.7517506059789927,
      "grad_norm": 1.0158463724350495,
      "learning_rate": 6.9011964206207585e-06,
      "loss": 0.4280811309814453,
      "num_input_tokens_seen": 182841952,
      "step": 11165,
      "train_runtime": 21468.756,
      "train_tokens_per_second": 8516.653
    },
    {
      "epoch": 0.7520872609749528,
      "grad_norm": 1.0406612307764609,
      "learning_rate": 6.898750674654531e-06,
      "loss": 0.3837984561920166,
      "num_input_tokens_seen": 182923872,
      "step": 11170,
      "train_runtime": 21478.3919,
      "train_tokens_per_second": 8516.647
    },
    {
      "epoch": 0.752423915970913,
      "grad_norm": 0.9546141632117108,
      "learning_rate": 6.896304397708733e-06,
      "loss": 0.3694461822509766,
      "num_input_tokens_seen": 183005792,
      "step": 11175,
      "train_runtime": 21488.0268,
      "train_tokens_per_second": 8516.64
    },
    {
      "epoch": 0.7527605709668731,
      "grad_norm": 1.0024927834125539,
      "learning_rate": 6.893857590467456e-06,
      "loss": 0.37308406829833984,
      "num_input_tokens_seen": 183087496,
      "step": 11180,
      "train_runtime": 21498.1438,
      "train_tokens_per_second": 8516.433
    },
    {
      "epoch": 0.7530972259628332,
      "grad_norm": 1.0897452479522878,
      "learning_rate": 6.891410253614944e-06,
      "loss": 0.3808279991149902,
      "num_input_tokens_seen": 183168896,
      "step": 11185,
      "train_runtime": 21508.0587,
      "train_tokens_per_second": 8516.291
    },
    {
      "epoch": 0.7534338809587934,
      "grad_norm": 1.0596345315259315,
      "learning_rate": 6.888962387835586e-06,
      "loss": 0.3755460500717163,
      "num_input_tokens_seen": 183250816,
      "step": 11190,
      "train_runtime": 21517.6028,
      "train_tokens_per_second": 8516.321
    },
    {
      "epoch": 0.7537705359547535,
      "grad_norm": 1.1127074683819567,
      "learning_rate": 6.886513993813919e-06,
      "loss": 0.3908921480178833,
      "num_input_tokens_seen": 183332736,
      "step": 11195,
      "train_runtime": 21527.1568,
      "train_tokens_per_second": 8516.347
    },
    {
      "epoch": 0.7541071909507137,
      "grad_norm": 1.0235608127400435,
      "learning_rate": 6.88406507223463e-06,
      "loss": 0.3671010255813599,
      "num_input_tokens_seen": 183414656,
      "step": 11200,
      "train_runtime": 21536.7813,
      "train_tokens_per_second": 8516.345
    },
    {
      "epoch": 0.7544438459466738,
      "grad_norm": 1.0310903051129334,
      "learning_rate": 6.881615623782551e-06,
      "loss": 0.38761518001556394,
      "num_input_tokens_seen": 183496576,
      "step": 11205,
      "train_runtime": 21546.3173,
      "train_tokens_per_second": 8516.378
    },
    {
      "epoch": 0.7547805009426339,
      "grad_norm": 1.0074839755815994,
      "learning_rate": 6.879165649142663e-06,
      "loss": 0.3975763559341431,
      "num_input_tokens_seen": 183578496,
      "step": 11210,
      "train_runtime": 21555.8778,
      "train_tokens_per_second": 8516.401
    },
    {
      "epoch": 0.7551171559385941,
      "grad_norm": 1.1623829510744679,
      "learning_rate": 6.876715149000093e-06,
      "loss": 0.39841256141662595,
      "num_input_tokens_seen": 183660416,
      "step": 11215,
      "train_runtime": 21565.9746,
      "train_tokens_per_second": 8516.212
    },
    {
      "epoch": 0.7554538109345542,
      "grad_norm": 0.8528583014540164,
      "learning_rate": 6.874264124040117e-06,
      "loss": 0.3754589080810547,
      "num_input_tokens_seen": 183742336,
      "step": 11220,
      "train_runtime": 21575.5868,
      "train_tokens_per_second": 8516.215
    },
    {
      "epoch": 0.7557904659305144,
      "grad_norm": 1.0676487182258272,
      "learning_rate": 6.871812574948154e-06,
      "loss": 0.4120182991027832,
      "num_input_tokens_seen": 183824256,
      "step": 11225,
      "train_runtime": 21585.1078,
      "train_tokens_per_second": 8516.254
    },
    {
      "epoch": 0.7561271209264745,
      "grad_norm": 1.113217073166621,
      "learning_rate": 6.869360502409773e-06,
      "loss": 0.41984095573425295,
      "num_input_tokens_seen": 183906176,
      "step": 11230,
      "train_runtime": 21594.7213,
      "train_tokens_per_second": 8516.256
    },
    {
      "epoch": 0.7564637759224346,
      "grad_norm": 1.1306426884002458,
      "learning_rate": 6.866907907110689e-06,
      "loss": 0.3791032791137695,
      "num_input_tokens_seen": 183988096,
      "step": 11235,
      "train_runtime": 21604.3299,
      "train_tokens_per_second": 8516.26
    },
    {
      "epoch": 0.7568004309183949,
      "grad_norm": 0.9269840398167913,
      "learning_rate": 6.864454789736763e-06,
      "loss": 0.3543464183807373,
      "num_input_tokens_seen": 184070016,
      "step": 11240,
      "train_runtime": 21613.8298,
      "train_tokens_per_second": 8516.307
    },
    {
      "epoch": 0.757137085914355,
      "grad_norm": 1.1222738801333314,
      "learning_rate": 6.8620011509740005e-06,
      "loss": 0.3652854681015015,
      "num_input_tokens_seen": 184151936,
      "step": 11245,
      "train_runtime": 21623.4227,
      "train_tokens_per_second": 8516.318
    },
    {
      "epoch": 0.7574737409103152,
      "grad_norm": 1.068357101918241,
      "learning_rate": 6.859546991508552e-06,
      "loss": 0.39225521087646487,
      "num_input_tokens_seen": 184233856,
      "step": 11250,
      "train_runtime": 21633.0291,
      "train_tokens_per_second": 8516.323
    },
    {
      "epoch": 0.7578103959062753,
      "grad_norm": 1.0991653341851815,
      "learning_rate": 6.857092312026716e-06,
      "loss": 0.4045055866241455,
      "num_input_tokens_seen": 184315776,
      "step": 11255,
      "train_runtime": 21642.5784,
      "train_tokens_per_second": 8516.35
    },
    {
      "epoch": 0.7581470509022354,
      "grad_norm": 1.0133860567845978,
      "learning_rate": 6.8546371132149395e-06,
      "loss": 0.40831737518310546,
      "num_input_tokens_seen": 184397696,
      "step": 11260,
      "train_runtime": 21652.1551,
      "train_tokens_per_second": 8516.367
    },
    {
      "epoch": 0.7584837058981956,
      "grad_norm": 0.8912797204352549,
      "learning_rate": 6.852181395759807e-06,
      "loss": 0.3826915740966797,
      "num_input_tokens_seen": 184479616,
      "step": 11265,
      "train_runtime": 21661.7685,
      "train_tokens_per_second": 8516.369
    },
    {
      "epoch": 0.7588203608941557,
      "grad_norm": 1.0447294403736076,
      "learning_rate": 6.849725160348054e-06,
      "loss": 0.4066917419433594,
      "num_input_tokens_seen": 184561536,
      "step": 11270,
      "train_runtime": 21671.3265,
      "train_tokens_per_second": 8516.393
    },
    {
      "epoch": 0.7591570158901159,
      "grad_norm": 0.9622051519833472,
      "learning_rate": 6.847268407666557e-06,
      "loss": 0.3794696569442749,
      "num_input_tokens_seen": 184643456,
      "step": 11275,
      "train_runtime": 21680.901,
      "train_tokens_per_second": 8516.411
    },
    {
      "epoch": 0.759493670886076,
      "grad_norm": 1.0986124175148198,
      "learning_rate": 6.844811138402341e-06,
      "loss": 0.39482650756835935,
      "num_input_tokens_seen": 184725376,
      "step": 11280,
      "train_runtime": 21690.5054,
      "train_tokens_per_second": 8516.416
    },
    {
      "epoch": 0.7598303258820361,
      "grad_norm": 0.9789363470133807,
      "learning_rate": 6.842353353242574e-06,
      "loss": 0.3782644510269165,
      "num_input_tokens_seen": 184807296,
      "step": 11285,
      "train_runtime": 21700.1457,
      "train_tokens_per_second": 8516.408
    },
    {
      "epoch": 0.7601669808779963,
      "grad_norm": 1.1621777453975686,
      "learning_rate": 6.839895052874565e-06,
      "loss": 0.4143389701843262,
      "num_input_tokens_seen": 184889216,
      "step": 11290,
      "train_runtime": 21709.6568,
      "train_tokens_per_second": 8516.45
    },
    {
      "epoch": 0.7605036358739564,
      "grad_norm": 1.0042677524742194,
      "learning_rate": 6.837436237985772e-06,
      "loss": 0.37768361568450926,
      "num_input_tokens_seen": 184971136,
      "step": 11295,
      "train_runtime": 21719.2817,
      "train_tokens_per_second": 8516.448
    },
    {
      "epoch": 0.7608402908699166,
      "grad_norm": 1.0697764275932895,
      "learning_rate": 6.834976909263794e-06,
      "loss": 0.3898283004760742,
      "num_input_tokens_seen": 185053056,
      "step": 11300,
      "train_runtime": 21728.9022,
      "train_tokens_per_second": 8516.448
    },
    {
      "epoch": 0.7611769458658767,
      "grad_norm": 1.1522821827253722,
      "learning_rate": 6.832517067396374e-06,
      "loss": 0.38615190982818604,
      "num_input_tokens_seen": 185134976,
      "step": 11305,
      "train_runtime": 21738.4099,
      "train_tokens_per_second": 8516.491
    },
    {
      "epoch": 0.7615136008618368,
      "grad_norm": 1.1837852774402984,
      "learning_rate": 6.8300567130714e-06,
      "loss": 0.4845320701599121,
      "num_input_tokens_seen": 185216896,
      "step": 11310,
      "train_runtime": 21748.0347,
      "train_tokens_per_second": 8516.489
    },
    {
      "epoch": 0.761850255857797,
      "grad_norm": 1.0472607566232612,
      "learning_rate": 6.8275958469769035e-06,
      "loss": 0.43174543380737307,
      "num_input_tokens_seen": 185298816,
      "step": 11315,
      "train_runtime": 21757.6357,
      "train_tokens_per_second": 8516.496
    },
    {
      "epoch": 0.7621869108537571,
      "grad_norm": 1.0733962165528441,
      "learning_rate": 6.825134469801055e-06,
      "loss": 0.3787389755249023,
      "num_input_tokens_seen": 185380736,
      "step": 11320,
      "train_runtime": 21767.1378,
      "train_tokens_per_second": 8516.542
    },
    {
      "epoch": 0.7625235658497173,
      "grad_norm": 1.0066793394257252,
      "learning_rate": 6.8226725822321715e-06,
      "loss": 0.4035038948059082,
      "num_input_tokens_seen": 185462656,
      "step": 11325,
      "train_runtime": 21776.7878,
      "train_tokens_per_second": 8516.53
    },
    {
      "epoch": 0.7628602208456774,
      "grad_norm": 1.1096934635608484,
      "learning_rate": 6.820210184958712e-06,
      "loss": 0.44266843795776367,
      "num_input_tokens_seen": 185544576,
      "step": 11330,
      "train_runtime": 21786.3556,
      "train_tokens_per_second": 8516.55
    },
    {
      "epoch": 0.7631968758416375,
      "grad_norm": 1.0946385107858525,
      "learning_rate": 6.81774727866928e-06,
      "loss": 0.3793335914611816,
      "num_input_tokens_seen": 185626496,
      "step": 11335,
      "train_runtime": 21795.9107,
      "train_tokens_per_second": 8516.574
    },
    {
      "epoch": 0.7635335308375977,
      "grad_norm": 1.0173038059818218,
      "learning_rate": 6.815283864052617e-06,
      "loss": 0.41010313034057616,
      "num_input_tokens_seen": 185708416,
      "step": 11340,
      "train_runtime": 21805.4869,
      "train_tokens_per_second": 8516.591
    },
    {
      "epoch": 0.7638701858335578,
      "grad_norm": 1.2658135635570578,
      "learning_rate": 6.812819941797611e-06,
      "loss": 0.4240370273590088,
      "num_input_tokens_seen": 185790336,
      "step": 11345,
      "train_runtime": 21815.0965,
      "train_tokens_per_second": 8516.595
    },
    {
      "epoch": 0.764206840829518,
      "grad_norm": 1.1635830361455233,
      "learning_rate": 6.810355512593288e-06,
      "loss": 0.3976787328720093,
      "num_input_tokens_seen": 185872256,
      "step": 11350,
      "train_runtime": 21824.6509,
      "train_tokens_per_second": 8516.62
    },
    {
      "epoch": 0.7645434958254781,
      "grad_norm": 1.0226664121563624,
      "learning_rate": 6.807890577128817e-06,
      "loss": 0.3629487991333008,
      "num_input_tokens_seen": 185954176,
      "step": 11355,
      "train_runtime": 21834.2091,
      "train_tokens_per_second": 8516.644
    },
    {
      "epoch": 0.7648801508214382,
      "grad_norm": 1.1015760807465924,
      "learning_rate": 6.805425136093513e-06,
      "loss": 0.41353445053100585,
      "num_input_tokens_seen": 186036096,
      "step": 11360,
      "train_runtime": 21843.8136,
      "train_tokens_per_second": 8516.649
    },
    {
      "epoch": 0.7652168058173984,
      "grad_norm": 1.1386193450420004,
      "learning_rate": 6.802959190176827e-06,
      "loss": 0.41338133811950684,
      "num_input_tokens_seen": 186118016,
      "step": 11365,
      "train_runtime": 21853.4264,
      "train_tokens_per_second": 8516.651
    },
    {
      "epoch": 0.7655534608133585,
      "grad_norm": 1.0248099055076916,
      "learning_rate": 6.800492740068352e-06,
      "loss": 0.401457691192627,
      "num_input_tokens_seen": 186199936,
      "step": 11370,
      "train_runtime": 21862.9374,
      "train_tokens_per_second": 8516.693
    },
    {
      "epoch": 0.7658901158093187,
      "grad_norm": 1.1567154951495418,
      "learning_rate": 6.798025786457825e-06,
      "loss": 0.39359796047210693,
      "num_input_tokens_seen": 186281856,
      "step": 11375,
      "train_runtime": 21872.5458,
      "train_tokens_per_second": 8516.697
    },
    {
      "epoch": 0.7662267708052788,
      "grad_norm": 0.9358976101289447,
      "learning_rate": 6.795558330035118e-06,
      "loss": 0.3445276737213135,
      "num_input_tokens_seen": 186363776,
      "step": 11380,
      "train_runtime": 21882.1107,
      "train_tokens_per_second": 8516.718
    },
    {
      "epoch": 0.7665634258012389,
      "grad_norm": 1.009973747478916,
      "learning_rate": 6.793090371490251e-06,
      "loss": 0.40588970184326173,
      "num_input_tokens_seen": 186445696,
      "step": 11385,
      "train_runtime": 21891.6118,
      "train_tokens_per_second": 8516.764
    },
    {
      "epoch": 0.7669000807971991,
      "grad_norm": 0.9796243318777613,
      "learning_rate": 6.790621911513378e-06,
      "loss": 0.3790421962738037,
      "num_input_tokens_seen": 186527616,
      "step": 11390,
      "train_runtime": 21901.2235,
      "train_tokens_per_second": 8516.767
    },
    {
      "epoch": 0.7672367357931592,
      "grad_norm": 1.2642220365994354,
      "learning_rate": 6.788152950794799e-06,
      "loss": 0.43274688720703125,
      "num_input_tokens_seen": 186609536,
      "step": 11395,
      "train_runtime": 21910.8624,
      "train_tokens_per_second": 8516.759
    },
    {
      "epoch": 0.7675733907891193,
      "grad_norm": 1.19982487314678,
      "learning_rate": 6.7856834900249495e-06,
      "loss": 0.4207283973693848,
      "num_input_tokens_seen": 186691456,
      "step": 11400,
      "train_runtime": 21920.372,
      "train_tokens_per_second": 8516.801
    },
    {
      "epoch": 0.7679100457850795,
      "grad_norm": 1.0306944050507634,
      "learning_rate": 6.783213529894405e-06,
      "loss": 0.3648836612701416,
      "num_input_tokens_seen": 186773376,
      "step": 11405,
      "train_runtime": 21929.9841,
      "train_tokens_per_second": 8516.804
    },
    {
      "epoch": 0.7682467007810396,
      "grad_norm": 1.110560724241591,
      "learning_rate": 6.780743071093883e-06,
      "loss": 0.40535645484924315,
      "num_input_tokens_seen": 186855296,
      "step": 11410,
      "train_runtime": 21939.6122,
      "train_tokens_per_second": 8516.8
    },
    {
      "epoch": 0.7685833557769998,
      "grad_norm": 1.089435446346703,
      "learning_rate": 6.77827211431424e-06,
      "loss": 0.39268195629119873,
      "num_input_tokens_seen": 186937216,
      "step": 11415,
      "train_runtime": 21949.1724,
      "train_tokens_per_second": 8516.823
    },
    {
      "epoch": 0.7689200107729599,
      "grad_norm": 0.9739175928103554,
      "learning_rate": 6.775800660246471e-06,
      "loss": 0.410624361038208,
      "num_input_tokens_seen": 187019104,
      "step": 11420,
      "train_runtime": 21959.28,
      "train_tokens_per_second": 8516.632
    },
    {
      "epoch": 0.76925666576892,
      "grad_norm": 1.0247693364211394,
      "learning_rate": 6.77332870958171e-06,
      "loss": 0.38894200325012207,
      "num_input_tokens_seen": 187101024,
      "step": 11425,
      "train_runtime": 21968.8674,
      "train_tokens_per_second": 8516.644
    },
    {
      "epoch": 0.7695933207648802,
      "grad_norm": 0.8474135960874885,
      "learning_rate": 6.770856263011229e-06,
      "loss": 0.3493107557296753,
      "num_input_tokens_seen": 187182944,
      "step": 11430,
      "train_runtime": 21978.4248,
      "train_tokens_per_second": 8516.668
    },
    {
      "epoch": 0.7699299757608403,
      "grad_norm": 1.139706026625615,
      "learning_rate": 6.768383321226442e-06,
      "loss": 0.3987672567367554,
      "num_input_tokens_seen": 187264864,
      "step": 11435,
      "train_runtime": 21988.0093,
      "train_tokens_per_second": 8516.681
    },
    {
      "epoch": 0.7702666307568005,
      "grad_norm": 1.066629806566119,
      "learning_rate": 6.765909884918898e-06,
      "loss": 0.39685022830963135,
      "num_input_tokens_seen": 187346784,
      "step": 11440,
      "train_runtime": 21997.632,
      "train_tokens_per_second": 8516.68
    },
    {
      "epoch": 0.7706032857527606,
      "grad_norm": 1.122318734150163,
      "learning_rate": 6.763435954780286e-06,
      "loss": 0.4062179088592529,
      "num_input_tokens_seen": 187428704,
      "step": 11445,
      "train_runtime": 22007.254,
      "train_tokens_per_second": 8516.678
    },
    {
      "epoch": 0.7709399407487207,
      "grad_norm": 1.1755186200969812,
      "learning_rate": 6.7609615315024325e-06,
      "loss": 0.41233100891113283,
      "num_input_tokens_seen": 187510624,
      "step": 11450,
      "train_runtime": 22016.7625,
      "train_tokens_per_second": 8516.721
    },
    {
      "epoch": 0.7712765957446809,
      "grad_norm": 1.1029986334086141,
      "learning_rate": 6.758486615777301e-06,
      "loss": 0.3670032978057861,
      "num_input_tokens_seen": 187592544,
      "step": 11455,
      "train_runtime": 22026.3716,
      "train_tokens_per_second": 8516.725
    },
    {
      "epoch": 0.771613250740641,
      "grad_norm": 1.0697662374622685,
      "learning_rate": 6.756011208296995e-06,
      "loss": 0.37116992473602295,
      "num_input_tokens_seen": 187674464,
      "step": 11460,
      "train_runtime": 22036.0313,
      "train_tokens_per_second": 8516.709
    },
    {
      "epoch": 0.7719499057366012,
      "grad_norm": 1.276351412050908,
      "learning_rate": 6.753535309753755e-06,
      "loss": 0.3988738298416138,
      "num_input_tokens_seen": 187756384,
      "step": 11465,
      "train_runtime": 22045.5428,
      "train_tokens_per_second": 8516.75
    },
    {
      "epoch": 0.7722865607325613,
      "grad_norm": 1.1629661370510416,
      "learning_rate": 6.7510589208399565e-06,
      "loss": 0.3888532638549805,
      "num_input_tokens_seen": 187838304,
      "step": 11470,
      "train_runtime": 22055.119,
      "train_tokens_per_second": 8516.767
    },
    {
      "epoch": 0.7726232157285214,
      "grad_norm": 1.3139215801262656,
      "learning_rate": 6.748582042248115e-06,
      "loss": 0.40160493850708007,
      "num_input_tokens_seen": 187920224,
      "step": 11475,
      "train_runtime": 22064.7402,
      "train_tokens_per_second": 8516.766
    },
    {
      "epoch": 0.7729598707244816,
      "grad_norm": 0.9893324782494062,
      "learning_rate": 6.746104674670882e-06,
      "loss": 0.4007710456848145,
      "num_input_tokens_seen": 188002144,
      "step": 11480,
      "train_runtime": 22074.2475,
      "train_tokens_per_second": 8516.809
    },
    {
      "epoch": 0.7732965257204417,
      "grad_norm": 1.032547876319991,
      "learning_rate": 6.743626818801042e-06,
      "loss": 0.44275569915771484,
      "num_input_tokens_seen": 188084064,
      "step": 11485,
      "train_runtime": 22083.8272,
      "train_tokens_per_second": 8516.824
    },
    {
      "epoch": 0.7736331807164019,
      "grad_norm": 0.9684094475469501,
      "learning_rate": 6.741148475331525e-06,
      "loss": 0.4205801963806152,
      "num_input_tokens_seen": 188165984,
      "step": 11490,
      "train_runtime": 22093.4062,
      "train_tokens_per_second": 8516.839
    },
    {
      "epoch": 0.773969835712362,
      "grad_norm": 0.9794568874087392,
      "learning_rate": 6.738669644955388e-06,
      "loss": 0.38278989791870116,
      "num_input_tokens_seen": 188247904,
      "step": 11495,
      "train_runtime": 22102.9635,
      "train_tokens_per_second": 8516.863
    },
    {
      "epoch": 0.7743064907083221,
      "grad_norm": 1.0434821125236429,
      "learning_rate": 6.736190328365828e-06,
      "loss": 0.3784179210662842,
      "num_input_tokens_seen": 188329824,
      "step": 11500,
      "train_runtime": 22112.5362,
      "train_tokens_per_second": 8516.88
    },
    {
      "epoch": 0.7746431457042823,
      "grad_norm": 1.148097121745586,
      "learning_rate": 6.73371052625618e-06,
      "loss": 0.44685983657836914,
      "num_input_tokens_seen": 188411744,
      "step": 11505,
      "train_runtime": 22122.1469,
      "train_tokens_per_second": 8516.883
    },
    {
      "epoch": 0.7749798007002424,
      "grad_norm": 0.9696779605688524,
      "learning_rate": 6.731230239319912e-06,
      "loss": 0.401274585723877,
      "num_input_tokens_seen": 188493664,
      "step": 11510,
      "train_runtime": 22131.7204,
      "train_tokens_per_second": 8516.901
    },
    {
      "epoch": 0.7753164556962026,
      "grad_norm": 1.070685241029627,
      "learning_rate": 6.728749468250627e-06,
      "loss": 0.4292941093444824,
      "num_input_tokens_seen": 188575584,
      "step": 11515,
      "train_runtime": 22141.2915,
      "train_tokens_per_second": 8516.919
    },
    {
      "epoch": 0.7756531106921627,
      "grad_norm": 1.15996602158177,
      "learning_rate": 6.726268213742066e-06,
      "loss": 0.38234767913818357,
      "num_input_tokens_seen": 188657504,
      "step": 11520,
      "train_runtime": 22150.9156,
      "train_tokens_per_second": 8516.917
    },
    {
      "epoch": 0.7759897656881228,
      "grad_norm": 0.9259638102309848,
      "learning_rate": 6.723786476488105e-06,
      "loss": 0.3857060432434082,
      "num_input_tokens_seen": 188739424,
      "step": 11525,
      "train_runtime": 22160.5462,
      "train_tokens_per_second": 8516.912
    },
    {
      "epoch": 0.776326420684083,
      "grad_norm": 0.9572142796070752,
      "learning_rate": 6.72130425718275e-06,
      "loss": 0.3788535356521606,
      "num_input_tokens_seen": 188821344,
      "step": 11530,
      "train_runtime": 22170.0562,
      "train_tokens_per_second": 8516.954
    },
    {
      "epoch": 0.7766630756800431,
      "grad_norm": 1.052371764402304,
      "learning_rate": 6.718821556520151e-06,
      "loss": 0.3949310779571533,
      "num_input_tokens_seen": 188903264,
      "step": 11535,
      "train_runtime": 22179.6815,
      "train_tokens_per_second": 8516.951
    },
    {
      "epoch": 0.7769997306760033,
      "grad_norm": 1.2339898500066606,
      "learning_rate": 6.716338375194583e-06,
      "loss": 0.4365112781524658,
      "num_input_tokens_seen": 188985184,
      "step": 11540,
      "train_runtime": 22189.2885,
      "train_tokens_per_second": 8516.956
    },
    {
      "epoch": 0.7773363856719634,
      "grad_norm": 1.0131628491787408,
      "learning_rate": 6.713854713900463e-06,
      "loss": 0.4106463432312012,
      "num_input_tokens_seen": 189067104,
      "step": 11545,
      "train_runtime": 22198.7897,
      "train_tokens_per_second": 8517.001
    },
    {
      "epoch": 0.7776730406679235,
      "grad_norm": 1.1827628955200502,
      "learning_rate": 6.7113705733323365e-06,
      "loss": 0.42244348526000974,
      "num_input_tokens_seen": 189149024,
      "step": 11550,
      "train_runtime": 22208.4009,
      "train_tokens_per_second": 8517.003
    },
    {
      "epoch": 0.7780096956638837,
      "grad_norm": 1.1899174340160124,
      "learning_rate": 6.708885954184885e-06,
      "loss": 0.3957364559173584,
      "num_input_tokens_seen": 189230944,
      "step": 11555,
      "train_runtime": 22218.0087,
      "train_tokens_per_second": 8517.007
    },
    {
      "epoch": 0.7783463506598438,
      "grad_norm": 1.15142294576868,
      "learning_rate": 6.7064008571529275e-06,
      "loss": 0.39062564373016356,
      "num_input_tokens_seen": 189312864,
      "step": 11560,
      "train_runtime": 22227.566,
      "train_tokens_per_second": 8517.031
    },
    {
      "epoch": 0.778683005655804,
      "grad_norm": 0.8159214419347172,
      "learning_rate": 6.703915282931411e-06,
      "loss": 0.3917693138122559,
      "num_input_tokens_seen": 189394784,
      "step": 11565,
      "train_runtime": 22237.0834,
      "train_tokens_per_second": 8517.069
    },
    {
      "epoch": 0.7790196606517641,
      "grad_norm": 0.9730780495878428,
      "learning_rate": 6.701429232215418e-06,
      "loss": 0.39929192066192626,
      "num_input_tokens_seen": 189476704,
      "step": 11570,
      "train_runtime": 22246.7192,
      "train_tokens_per_second": 8517.063
    },
    {
      "epoch": 0.7793563156477242,
      "grad_norm": 1.266261000943124,
      "learning_rate": 6.6989427057001665e-06,
      "loss": 0.44538702964782717,
      "num_input_tokens_seen": 189558624,
      "step": 11575,
      "train_runtime": 22256.2676,
      "train_tokens_per_second": 8517.09
    },
    {
      "epoch": 0.7796929706436844,
      "grad_norm": 1.1619192519918886,
      "learning_rate": 6.696455704081006e-06,
      "loss": 0.40988569259643554,
      "num_input_tokens_seen": 189640544,
      "step": 11580,
      "train_runtime": 22265.8346,
      "train_tokens_per_second": 8517.109
    },
    {
      "epoch": 0.7800296256396445,
      "grad_norm": 0.9455708769143673,
      "learning_rate": 6.693968228053416e-06,
      "loss": 0.40745134353637696,
      "num_input_tokens_seen": 189722464,
      "step": 11585,
      "train_runtime": 22275.4514,
      "train_tokens_per_second": 8517.11
    },
    {
      "epoch": 0.7803662806356046,
      "grad_norm": 1.0727831039471494,
      "learning_rate": 6.691480278313013e-06,
      "loss": 0.3893634796142578,
      "num_input_tokens_seen": 189804384,
      "step": 11590,
      "train_runtime": 22285.0152,
      "train_tokens_per_second": 8517.131
    },
    {
      "epoch": 0.7807029356315648,
      "grad_norm": 1.04760179265613,
      "learning_rate": 6.688991855555545e-06,
      "loss": 0.37516376972198484,
      "num_input_tokens_seen": 189886304,
      "step": 11595,
      "train_runtime": 22294.6082,
      "train_tokens_per_second": 8517.14
    },
    {
      "epoch": 0.7810395906275249,
      "grad_norm": 1.0902747105138435,
      "learning_rate": 6.686502960476888e-06,
      "loss": 0.40005927085876464,
      "num_input_tokens_seen": 189968224,
      "step": 11600,
      "train_runtime": 22304.2402,
      "train_tokens_per_second": 8517.135
    },
    {
      "epoch": 0.7813762456234851,
      "grad_norm": 1.1326733736547134,
      "learning_rate": 6.684013593773058e-06,
      "loss": 0.36806821823120117,
      "num_input_tokens_seen": 190050144,
      "step": 11605,
      "train_runtime": 22313.8486,
      "train_tokens_per_second": 8517.139
    },
    {
      "epoch": 0.7817129006194452,
      "grad_norm": 1.2395975329276314,
      "learning_rate": 6.6815237561401955e-06,
      "loss": 0.383362340927124,
      "num_input_tokens_seen": 190132064,
      "step": 11610,
      "train_runtime": 22323.3643,
      "train_tokens_per_second": 8517.178
    },
    {
      "epoch": 0.7820495556154053,
      "grad_norm": 1.138872887900833,
      "learning_rate": 6.679033448274577e-06,
      "loss": 0.38904519081115724,
      "num_input_tokens_seen": 190213984,
      "step": 11615,
      "train_runtime": 22332.9867,
      "train_tokens_per_second": 8517.176
    },
    {
      "epoch": 0.7823862106113655,
      "grad_norm": 0.891291923147076,
      "learning_rate": 6.676542670872609e-06,
      "loss": 0.3596473455429077,
      "num_input_tokens_seen": 190295904,
      "step": 11620,
      "train_runtime": 22342.621,
      "train_tokens_per_second": 8517.17
    },
    {
      "epoch": 0.7827228656073256,
      "grad_norm": 1.281963794136922,
      "learning_rate": 6.674051424630829e-06,
      "loss": 0.3949894905090332,
      "num_input_tokens_seen": 190377824,
      "step": 11625,
      "train_runtime": 22352.1247,
      "train_tokens_per_second": 8517.214
    },
    {
      "epoch": 0.7830595206032858,
      "grad_norm": 1.0286225901368458,
      "learning_rate": 6.671559710245905e-06,
      "loss": 0.3540106534957886,
      "num_input_tokens_seen": 190459744,
      "step": 11630,
      "train_runtime": 22361.7412,
      "train_tokens_per_second": 8517.214
    },
    {
      "epoch": 0.7833961755992459,
      "grad_norm": 1.3408339485694707,
      "learning_rate": 6.669067528414639e-06,
      "loss": 0.36615719795227053,
      "num_input_tokens_seen": 190541664,
      "step": 11635,
      "train_runtime": 22371.3674,
      "train_tokens_per_second": 8517.211
    },
    {
      "epoch": 0.783732830595206,
      "grad_norm": 0.9621562727091856,
      "learning_rate": 6.6665748798339606e-06,
      "loss": 0.424742603302002,
      "num_input_tokens_seen": 190623584,
      "step": 11640,
      "train_runtime": 22380.8766,
      "train_tokens_per_second": 8517.253
    },
    {
      "epoch": 0.7840694855911662,
      "grad_norm": 1.0957686976929673,
      "learning_rate": 6.664081765200932e-06,
      "loss": 0.3967721939086914,
      "num_input_tokens_seen": 190705504,
      "step": 11645,
      "train_runtime": 22390.4195,
      "train_tokens_per_second": 8517.281
    },
    {
      "epoch": 0.7844061405871263,
      "grad_norm": 1.0400135909785237,
      "learning_rate": 6.661588185212745e-06,
      "loss": 0.40307369232177737,
      "num_input_tokens_seen": 190787424,
      "step": 11650,
      "train_runtime": 22400.0204,
      "train_tokens_per_second": 8517.288
    },
    {
      "epoch": 0.7847427955830865,
      "grad_norm": 1.0465291804837584,
      "learning_rate": 6.65909414056672e-06,
      "loss": 0.3919831275939941,
      "num_input_tokens_seen": 190869344,
      "step": 11655,
      "train_runtime": 22409.5727,
      "train_tokens_per_second": 8517.313
    },
    {
      "epoch": 0.7850794505790466,
      "grad_norm": 1.1100959975836031,
      "learning_rate": 6.6565996319603075e-06,
      "loss": 0.3899850368499756,
      "num_input_tokens_seen": 190951208,
      "step": 11660,
      "train_runtime": 22419.6729,
      "train_tokens_per_second": 8517.127
    },
    {
      "epoch": 0.7854161055750067,
      "grad_norm": 0.9009725575097663,
      "learning_rate": 6.6541046600910915e-06,
      "loss": 0.354740309715271,
      "num_input_tokens_seen": 191033128,
      "step": 11665,
      "train_runtime": 22429.2991,
      "train_tokens_per_second": 8517.124
    },
    {
      "epoch": 0.7857527605709669,
      "grad_norm": 0.9851992372270004,
      "learning_rate": 6.6516092256567824e-06,
      "loss": 0.37420339584350587,
      "num_input_tokens_seen": 191115048,
      "step": 11670,
      "train_runtime": 22438.8808,
      "train_tokens_per_second": 8517.138
    },
    {
      "epoch": 0.786089415566927,
      "grad_norm": 1.264530649964702,
      "learning_rate": 6.64911332935522e-06,
      "loss": 0.3938652515411377,
      "num_input_tokens_seen": 191196968,
      "step": 11675,
      "train_runtime": 22448.4599,
      "train_tokens_per_second": 8517.153
    },
    {
      "epoch": 0.7864260705628872,
      "grad_norm": 1.0938042453307382,
      "learning_rate": 6.646616971884373e-06,
      "loss": 0.4097557067871094,
      "num_input_tokens_seen": 191278888,
      "step": 11680,
      "train_runtime": 22458.0751,
      "train_tokens_per_second": 8517.154
    },
    {
      "epoch": 0.7867627255588473,
      "grad_norm": 1.0559596350191394,
      "learning_rate": 6.64412015394234e-06,
      "loss": 0.39625279903411864,
      "num_input_tokens_seen": 191359728,
      "step": 11685,
      "train_runtime": 22467.8596,
      "train_tokens_per_second": 8517.043
    },
    {
      "epoch": 0.7870993805548074,
      "grad_norm": 1.2056470439595066,
      "learning_rate": 6.641622876227349e-06,
      "loss": 0.4287084102630615,
      "num_input_tokens_seen": 191441648,
      "step": 11690,
      "train_runtime": 22477.3618,
      "train_tokens_per_second": 8517.087
    },
    {
      "epoch": 0.7874360355507676,
      "grad_norm": 0.8033008505468352,
      "learning_rate": 6.639125139437756e-06,
      "loss": 0.3566891193389893,
      "num_input_tokens_seen": 191523568,
      "step": 11695,
      "train_runtime": 22486.9769,
      "train_tokens_per_second": 8517.088
    },
    {
      "epoch": 0.7877726905467277,
      "grad_norm": 1.1217209854952332,
      "learning_rate": 6.6366269442720446e-06,
      "loss": 0.3967562675476074,
      "num_input_tokens_seen": 191605488,
      "step": 11700,
      "train_runtime": 22496.5838,
      "train_tokens_per_second": 8517.093
    },
    {
      "epoch": 0.7881093455426879,
      "grad_norm": 1.0564997175091584,
      "learning_rate": 6.634128291428827e-06,
      "loss": 0.41399297714233396,
      "num_input_tokens_seen": 191687408,
      "step": 11705,
      "train_runtime": 22506.0941,
      "train_tokens_per_second": 8517.134
    },
    {
      "epoch": 0.788446000538648,
      "grad_norm": 1.00818767363929,
      "learning_rate": 6.631629181606842e-06,
      "loss": 0.3784499645233154,
      "num_input_tokens_seen": 191769328,
      "step": 11710,
      "train_runtime": 22515.7245,
      "train_tokens_per_second": 8517.129
    },
    {
      "epoch": 0.7887826555346081,
      "grad_norm": 1.2486772966499948,
      "learning_rate": 6.629129615504961e-06,
      "loss": 0.40331759452819826,
      "num_input_tokens_seen": 191851248,
      "step": 11715,
      "train_runtime": 22525.3339,
      "train_tokens_per_second": 8517.132
    },
    {
      "epoch": 0.7891193105305683,
      "grad_norm": 1.0903895675212147,
      "learning_rate": 6.626629593822178e-06,
      "loss": 0.3834541320800781,
      "num_input_tokens_seen": 191933168,
      "step": 11720,
      "train_runtime": 22534.8861,
      "train_tokens_per_second": 8517.157
    },
    {
      "epoch": 0.7894559655265284,
      "grad_norm": 1.2280546817546083,
      "learning_rate": 6.624129117257616e-06,
      "loss": 0.4208484172821045,
      "num_input_tokens_seen": 192015088,
      "step": 11725,
      "train_runtime": 22544.4599,
      "train_tokens_per_second": 8517.174
    },
    {
      "epoch": 0.7897926205224886,
      "grad_norm": 1.3916963622156466,
      "learning_rate": 6.621628186510524e-06,
      "loss": 0.4182647705078125,
      "num_input_tokens_seen": 192097008,
      "step": 11730,
      "train_runtime": 22554.0884,
      "train_tokens_per_second": 8517.17
    },
    {
      "epoch": 0.7901292755184487,
      "grad_norm": 0.9472683821032516,
      "learning_rate": 6.619126802280281e-06,
      "loss": 0.3733630657196045,
      "num_input_tokens_seen": 192178928,
      "step": 11735,
      "train_runtime": 22563.6433,
      "train_tokens_per_second": 8517.194
    },
    {
      "epoch": 0.7904659305144088,
      "grad_norm": 1.0469915600132524,
      "learning_rate": 6.616624965266392e-06,
      "loss": 0.4225646495819092,
      "num_input_tokens_seen": 192260848,
      "step": 11740,
      "train_runtime": 22573.2084,
      "train_tokens_per_second": 8517.214
    },
    {
      "epoch": 0.790802585510369,
      "grad_norm": 1.1859993976829566,
      "learning_rate": 6.614122676168486e-06,
      "loss": 0.4205310821533203,
      "num_input_tokens_seen": 192342768,
      "step": 11745,
      "train_runtime": 22582.8272,
      "train_tokens_per_second": 8517.214
    },
    {
      "epoch": 0.7911392405063291,
      "grad_norm": 1.0453350375815478,
      "learning_rate": 6.611619935686321e-06,
      "loss": 0.37412779331207274,
      "num_input_tokens_seen": 192424392,
      "step": 11750,
      "train_runtime": 22592.8471,
      "train_tokens_per_second": 8517.049
    },
    {
      "epoch": 0.7914758955022893,
      "grad_norm": 1.0337634800841258,
      "learning_rate": 6.609116744519779e-06,
      "loss": 0.41260547637939454,
      "num_input_tokens_seen": 192506312,
      "step": 11755,
      "train_runtime": 22602.352,
      "train_tokens_per_second": 8517.092
    },
    {
      "epoch": 0.7918125504982494,
      "grad_norm": 1.15744349090679,
      "learning_rate": 6.60661310336887e-06,
      "loss": 0.33180060386657717,
      "num_input_tokens_seen": 192588232,
      "step": 11760,
      "train_runtime": 22611.9369,
      "train_tokens_per_second": 8517.105
    },
    {
      "epoch": 0.7921492054942095,
      "grad_norm": 1.0855532139045414,
      "learning_rate": 6.604109012933732e-06,
      "loss": 0.4022209644317627,
      "num_input_tokens_seen": 192669208,
      "step": 11765,
      "train_runtime": 22621.724,
      "train_tokens_per_second": 8516.999
    },
    {
      "epoch": 0.7924858604901697,
      "grad_norm": 0.9605899205702353,
      "learning_rate": 6.601604473914622e-06,
      "loss": 0.4034572601318359,
      "num_input_tokens_seen": 192751128,
      "step": 11770,
      "train_runtime": 22631.2261,
      "train_tokens_per_second": 8517.043
    },
    {
      "epoch": 0.7928225154861298,
      "grad_norm": 0.9896516287322786,
      "learning_rate": 6.599099487011928e-06,
      "loss": 0.39771437644958496,
      "num_input_tokens_seen": 192833048,
      "step": 11775,
      "train_runtime": 22640.8396,
      "train_tokens_per_second": 8517.045
    },
    {
      "epoch": 0.79315917048209,
      "grad_norm": 1.0795308062825766,
      "learning_rate": 6.596594052926163e-06,
      "loss": 0.4026334285736084,
      "num_input_tokens_seen": 192914968,
      "step": 11780,
      "train_runtime": 22650.4565,
      "train_tokens_per_second": 8517.045
    },
    {
      "epoch": 0.7934958254780501,
      "grad_norm": 0.9925933531595927,
      "learning_rate": 6.5940881723579595e-06,
      "loss": 0.3612679958343506,
      "num_input_tokens_seen": 192996888,
      "step": 11785,
      "train_runtime": 22659.9691,
      "train_tokens_per_second": 8517.085
    },
    {
      "epoch": 0.7938324804740102,
      "grad_norm": 1.0612735708321657,
      "learning_rate": 6.591581846008085e-06,
      "loss": 0.3588229179382324,
      "num_input_tokens_seen": 193078808,
      "step": 11790,
      "train_runtime": 22669.584,
      "train_tokens_per_second": 8517.087
    },
    {
      "epoch": 0.7941691354699704,
      "grad_norm": 0.9388009927557244,
      "learning_rate": 6.589075074577421e-06,
      "loss": 0.36874845027923586,
      "num_input_tokens_seen": 193160728,
      "step": 11795,
      "train_runtime": 22679.1976,
      "train_tokens_per_second": 8517.088
    },
    {
      "epoch": 0.7945057904659305,
      "grad_norm": 1.0215865908209893,
      "learning_rate": 6.5865678587669805e-06,
      "loss": 0.4016460418701172,
      "num_input_tokens_seen": 193242648,
      "step": 11800,
      "train_runtime": 22688.7603,
      "train_tokens_per_second": 8517.109
    },
    {
      "epoch": 0.7948424454618906,
      "grad_norm": 1.081567620218517,
      "learning_rate": 6.584060199277897e-06,
      "loss": 0.4327713966369629,
      "num_input_tokens_seen": 193324568,
      "step": 11805,
      "train_runtime": 22698.3624,
      "train_tokens_per_second": 8517.115
    },
    {
      "epoch": 0.7951791004578508,
      "grad_norm": 1.145462654172766,
      "learning_rate": 6.581552096811428e-06,
      "loss": 0.3959984302520752,
      "num_input_tokens_seen": 193406488,
      "step": 11810,
      "train_runtime": 22707.9506,
      "train_tokens_per_second": 8517.127
    },
    {
      "epoch": 0.7955157554538109,
      "grad_norm": 1.1064775084982994,
      "learning_rate": 6.57904355206896e-06,
      "loss": 0.37884795665740967,
      "num_input_tokens_seen": 193488408,
      "step": 11815,
      "train_runtime": 22717.4686,
      "train_tokens_per_second": 8517.164
    },
    {
      "epoch": 0.7958524104497711,
      "grad_norm": 1.1680190978139913,
      "learning_rate": 6.576534565751997e-06,
      "loss": 0.3729418754577637,
      "num_input_tokens_seen": 193570328,
      "step": 11820,
      "train_runtime": 22727.0409,
      "train_tokens_per_second": 8517.181
    },
    {
      "epoch": 0.7961890654457312,
      "grad_norm": 0.9614557812120672,
      "learning_rate": 6.574025138562172e-06,
      "loss": 0.42534542083740234,
      "num_input_tokens_seen": 193652248,
      "step": 11825,
      "train_runtime": 22736.658,
      "train_tokens_per_second": 8517.182
    },
    {
      "epoch": 0.7965257204416913,
      "grad_norm": 0.804711588274061,
      "learning_rate": 6.571515271201234e-06,
      "loss": 0.35564727783203126,
      "num_input_tokens_seen": 193734168,
      "step": 11830,
      "train_runtime": 22746.2682,
      "train_tokens_per_second": 8517.185
    },
    {
      "epoch": 0.7968623754376515,
      "grad_norm": 1.08078788891279,
      "learning_rate": 6.569004964371062e-06,
      "loss": 0.3923297882080078,
      "num_input_tokens_seen": 193816088,
      "step": 11835,
      "train_runtime": 22755.7826,
      "train_tokens_per_second": 8517.224
    },
    {
      "epoch": 0.7971990304336116,
      "grad_norm": 1.234058232183272,
      "learning_rate": 6.566494218773656e-06,
      "loss": 0.4124558448791504,
      "num_input_tokens_seen": 193898008,
      "step": 11840,
      "train_runtime": 22765.4311,
      "train_tokens_per_second": 8517.212
    },
    {
      "epoch": 0.7975356854295718,
      "grad_norm": 1.3073530060892198,
      "learning_rate": 6.563983035111136e-06,
      "loss": 0.3947171688079834,
      "num_input_tokens_seen": 193979928,
      "step": 11845,
      "train_runtime": 22775.0499,
      "train_tokens_per_second": 8517.212
    },
    {
      "epoch": 0.7978723404255319,
      "grad_norm": 0.8984256959196911,
      "learning_rate": 6.561471414085748e-06,
      "loss": 0.3765054702758789,
      "num_input_tokens_seen": 194061848,
      "step": 11850,
      "train_runtime": 22784.5574,
      "train_tokens_per_second": 8517.253
    },
    {
      "epoch": 0.798208995421492,
      "grad_norm": 1.166108347814533,
      "learning_rate": 6.558959356399859e-06,
      "loss": 0.41008219718933103,
      "num_input_tokens_seen": 194143768,
      "step": 11855,
      "train_runtime": 22794.17,
      "train_tokens_per_second": 8517.255
    },
    {
      "epoch": 0.7985456504174522,
      "grad_norm": 0.8111938118721889,
      "learning_rate": 6.5564468627559544e-06,
      "loss": 0.3633738994598389,
      "num_input_tokens_seen": 194225688,
      "step": 11860,
      "train_runtime": 22803.7714,
      "train_tokens_per_second": 8517.262
    },
    {
      "epoch": 0.7988823054134123,
      "grad_norm": 1.1217341840878081,
      "learning_rate": 6.5539339338566495e-06,
      "loss": 0.44046831130981445,
      "num_input_tokens_seen": 194307608,
      "step": 11865,
      "train_runtime": 22813.283,
      "train_tokens_per_second": 8517.301
    },
    {
      "epoch": 0.7992189604093725,
      "grad_norm": 0.9908891153079643,
      "learning_rate": 6.5514205704046754e-06,
      "loss": 0.41256561279296877,
      "num_input_tokens_seen": 194389528,
      "step": 11870,
      "train_runtime": 22822.9203,
      "train_tokens_per_second": 8517.294
    },
    {
      "epoch": 0.7995556154053326,
      "grad_norm": 0.9333737814233134,
      "learning_rate": 6.548906773102886e-06,
      "loss": 0.34127624034881593,
      "num_input_tokens_seen": 194470920,
      "step": 11875,
      "train_runtime": 22832.8545,
      "train_tokens_per_second": 8517.153
    },
    {
      "epoch": 0.7998922704012927,
      "grad_norm": 1.1141277843244866,
      "learning_rate": 6.546392542654257e-06,
      "loss": 0.40090317726135255,
      "num_input_tokens_seen": 194552704,
      "step": 11880,
      "train_runtime": 22842.9181,
      "train_tokens_per_second": 8516.981
    },
    {
      "epoch": 0.8002289253972529,
      "grad_norm": 1.0873181287141174,
      "learning_rate": 6.543877879761882e-06,
      "loss": 0.41518774032592776,
      "num_input_tokens_seen": 194634624,
      "step": 11885,
      "train_runtime": 22852.481,
      "train_tokens_per_second": 8517.002
    },
    {
      "epoch": 0.800565580393213,
      "grad_norm": 1.103804162743698,
      "learning_rate": 6.5413627851289825e-06,
      "loss": 0.355963921546936,
      "num_input_tokens_seen": 194716544,
      "step": 11890,
      "train_runtime": 22862.0927,
      "train_tokens_per_second": 8517.004
    },
    {
      "epoch": 0.8009022353891732,
      "grad_norm": 0.8799919406388531,
      "learning_rate": 6.538847259458896e-06,
      "loss": 0.36477663516998293,
      "num_input_tokens_seen": 194798464,
      "step": 11895,
      "train_runtime": 22871.6541,
      "train_tokens_per_second": 8517.026
    },
    {
      "epoch": 0.8012388903851333,
      "grad_norm": 1.0407587596299794,
      "learning_rate": 6.5363313034550805e-06,
      "loss": 0.38185982704162597,
      "num_input_tokens_seen": 194880384,
      "step": 11900,
      "train_runtime": 22881.2316,
      "train_tokens_per_second": 8517.041
    },
    {
      "epoch": 0.8015755453810934,
      "grad_norm": 1.0704031472831044,
      "learning_rate": 6.533814917821114e-06,
      "loss": 0.3962876558303833,
      "num_input_tokens_seen": 194962304,
      "step": 11905,
      "train_runtime": 22890.8606,
      "train_tokens_per_second": 8517.037
    },
    {
      "epoch": 0.8019122003770536,
      "grad_norm": 1.3817647765863412,
      "learning_rate": 6.531298103260696e-06,
      "loss": 0.37332735061645506,
      "num_input_tokens_seen": 195044224,
      "step": 11910,
      "train_runtime": 22900.4799,
      "train_tokens_per_second": 8517.037
    },
    {
      "epoch": 0.8022488553730137,
      "grad_norm": 0.9817461877170581,
      "learning_rate": 6.528780860477648e-06,
      "loss": 0.38665645122528075,
      "num_input_tokens_seen": 195126144,
      "step": 11915,
      "train_runtime": 22909.985,
      "train_tokens_per_second": 8517.079
    },
    {
      "epoch": 0.8025855103689739,
      "grad_norm": 1.0177303557406574,
      "learning_rate": 6.526263190175907e-06,
      "loss": 0.3676005840301514,
      "num_input_tokens_seen": 195208064,
      "step": 11920,
      "train_runtime": 22919.5879,
      "train_tokens_per_second": 8517.084
    },
    {
      "epoch": 0.802922165364934,
      "grad_norm": 1.115059311619263,
      "learning_rate": 6.523745093059531e-06,
      "loss": 0.3937248229980469,
      "num_input_tokens_seen": 195289984,
      "step": 11925,
      "train_runtime": 22929.2177,
      "train_tokens_per_second": 8517.08
    },
    {
      "epoch": 0.8032588203608941,
      "grad_norm": 1.177316525718224,
      "learning_rate": 6.521226569832699e-06,
      "loss": 0.3917156934738159,
      "num_input_tokens_seen": 195371904,
      "step": 11930,
      "train_runtime": 22938.7283,
      "train_tokens_per_second": 8517.12
    },
    {
      "epoch": 0.8035954753568543,
      "grad_norm": 0.9274443576966519,
      "learning_rate": 6.518707621199707e-06,
      "loss": 0.3545417785644531,
      "num_input_tokens_seen": 195453824,
      "step": 11935,
      "train_runtime": 22948.3272,
      "train_tokens_per_second": 8517.127
    },
    {
      "epoch": 0.8039321303528144,
      "grad_norm": 1.2324040796036284,
      "learning_rate": 6.516188247864972e-06,
      "loss": 0.43592357635498047,
      "num_input_tokens_seen": 195535744,
      "step": 11940,
      "train_runtime": 22957.9379,
      "train_tokens_per_second": 8517.13
    },
    {
      "epoch": 0.8042687853487746,
      "grad_norm": 1.0955448888254689,
      "learning_rate": 6.513668450533028e-06,
      "loss": 0.3929403066635132,
      "num_input_tokens_seen": 195617384,
      "step": 11945,
      "train_runtime": 22967.915,
      "train_tokens_per_second": 8516.985
    },
    {
      "epoch": 0.8046054403447347,
      "grad_norm": 1.114880713862524,
      "learning_rate": 6.511148229908531e-06,
      "loss": 0.4233663558959961,
      "num_input_tokens_seen": 195699304,
      "step": 11950,
      "train_runtime": 22977.4908,
      "train_tokens_per_second": 8517.001
    },
    {
      "epoch": 0.8049420953406948,
      "grad_norm": 1.091712153914362,
      "learning_rate": 6.508627586696248e-06,
      "loss": 0.36252195835113527,
      "num_input_tokens_seen": 195781224,
      "step": 11955,
      "train_runtime": 22987.1032,
      "train_tokens_per_second": 8517.003
    },
    {
      "epoch": 0.805278750336655,
      "grad_norm": 1.010678772387954,
      "learning_rate": 6.5061065216010695e-06,
      "loss": 0.35509600639343264,
      "num_input_tokens_seen": 195863144,
      "step": 11960,
      "train_runtime": 22996.656,
      "train_tokens_per_second": 8517.027
    },
    {
      "epoch": 0.8056154053326151,
      "grad_norm": 1.0667163933373998,
      "learning_rate": 6.5035850353280075e-06,
      "loss": 0.33986654281616213,
      "num_input_tokens_seen": 195945064,
      "step": 11965,
      "train_runtime": 23006.2242,
      "train_tokens_per_second": 8517.046
    },
    {
      "epoch": 0.8059520603285752,
      "grad_norm": 0.9545448946011269,
      "learning_rate": 6.501063128582183e-06,
      "loss": 0.39634659290313723,
      "num_input_tokens_seen": 196026984,
      "step": 11970,
      "train_runtime": 23015.8435,
      "train_tokens_per_second": 8517.045
    },
    {
      "epoch": 0.8062887153245354,
      "grad_norm": 1.3484646832550313,
      "learning_rate": 6.498540802068843e-06,
      "loss": 0.4073770999908447,
      "num_input_tokens_seen": 196108904,
      "step": 11975,
      "train_runtime": 23025.4637,
      "train_tokens_per_second": 8517.045
    },
    {
      "epoch": 0.8066253703204955,
      "grad_norm": 1.1393589032733449,
      "learning_rate": 6.4960180564933454e-06,
      "loss": 0.36584551334381105,
      "num_input_tokens_seen": 196190824,
      "step": 11980,
      "train_runtime": 23034.9672,
      "train_tokens_per_second": 8517.087
    },
    {
      "epoch": 0.8069620253164557,
      "grad_norm": 1.2809561862026826,
      "learning_rate": 6.493494892561167e-06,
      "loss": 0.3895594596862793,
      "num_input_tokens_seen": 196272744,
      "step": 11985,
      "train_runtime": 23044.5312,
      "train_tokens_per_second": 8517.107
    },
    {
      "epoch": 0.8072986803124158,
      "grad_norm": 1.1641066841435856,
      "learning_rate": 6.490971310977907e-06,
      "loss": 0.4029975891113281,
      "num_input_tokens_seen": 196354664,
      "step": 11990,
      "train_runtime": 23054.1683,
      "train_tokens_per_second": 8517.1
    },
    {
      "epoch": 0.807635335308376,
      "grad_norm": 0.9873064689431406,
      "learning_rate": 6.488447312449273e-06,
      "loss": 0.3821958065032959,
      "num_input_tokens_seen": 196436584,
      "step": 11995,
      "train_runtime": 23063.6721,
      "train_tokens_per_second": 8517.143
    },
    {
      "epoch": 0.8079719903043361,
      "grad_norm": 1.2645371750379932,
      "learning_rate": 6.485922897681095e-06,
      "loss": 0.4058509349822998,
      "num_input_tokens_seen": 196518504,
      "step": 12000,
      "train_runtime": 23073.2712,
      "train_tokens_per_second": 8517.15
    },
    {
      "epoch": 0.8083086453002962,
      "grad_norm": 1.0421863628176669,
      "learning_rate": 6.483398067379316e-06,
      "loss": 0.4123128890991211,
      "num_input_tokens_seen": 196600424,
      "step": 12005,
      "train_runtime": 23082.8681,
      "train_tokens_per_second": 8517.158
    },
    {
      "epoch": 0.8086453002962564,
      "grad_norm": 1.1548426592284802,
      "learning_rate": 6.480872822249998e-06,
      "loss": 0.4001296520233154,
      "num_input_tokens_seen": 196682344,
      "step": 12010,
      "train_runtime": 23092.3683,
      "train_tokens_per_second": 8517.201
    },
    {
      "epoch": 0.8089819552922165,
      "grad_norm": 1.0969313421319082,
      "learning_rate": 6.478347162999318e-06,
      "loss": 0.3900286197662354,
      "num_input_tokens_seen": 196764264,
      "step": 12015,
      "train_runtime": 23101.9716,
      "train_tokens_per_second": 8517.207
    },
    {
      "epoch": 0.8093186102881766,
      "grad_norm": 0.884973057721478,
      "learning_rate": 6.475821090333568e-06,
      "loss": 0.38578505516052247,
      "num_input_tokens_seen": 196846184,
      "step": 12020,
      "train_runtime": 23111.5785,
      "train_tokens_per_second": 8517.211
    },
    {
      "epoch": 0.8096552652841368,
      "grad_norm": 1.0834437012404352,
      "learning_rate": 6.473294604959156e-06,
      "loss": 0.40588083267211916,
      "num_input_tokens_seen": 196928104,
      "step": 12025,
      "train_runtime": 23121.1308,
      "train_tokens_per_second": 8517.235
    },
    {
      "epoch": 0.8099919202800969,
      "grad_norm": 1.2251898928404032,
      "learning_rate": 6.470767707582605e-06,
      "loss": 0.37718245983123777,
      "num_input_tokens_seen": 197010024,
      "step": 12030,
      "train_runtime": 23130.6935,
      "train_tokens_per_second": 8517.255
    },
    {
      "epoch": 0.8103285752760571,
      "grad_norm": 1.1635966717564665,
      "learning_rate": 6.468240398910555e-06,
      "loss": 0.38895611763000487,
      "num_input_tokens_seen": 197091712,
      "step": 12035,
      "train_runtime": 23140.764,
      "train_tokens_per_second": 8517.079
    },
    {
      "epoch": 0.8106652302720172,
      "grad_norm": 1.1201623368568925,
      "learning_rate": 6.4657126796497595e-06,
      "loss": 0.3765002727508545,
      "num_input_tokens_seen": 197173632,
      "step": 12040,
      "train_runtime": 23150.3576,
      "train_tokens_per_second": 8517.088
    },
    {
      "epoch": 0.8110018852679773,
      "grad_norm": 1.0368198663484232,
      "learning_rate": 6.463184550507087e-06,
      "loss": 0.34524168968200686,
      "num_input_tokens_seen": 197255552,
      "step": 12045,
      "train_runtime": 23159.8579,
      "train_tokens_per_second": 8517.131
    },
    {
      "epoch": 0.8113385402639375,
      "grad_norm": 1.044379550931588,
      "learning_rate": 6.46065601218952e-06,
      "loss": 0.32676827907562256,
      "num_input_tokens_seen": 197337472,
      "step": 12050,
      "train_runtime": 23169.4793,
      "train_tokens_per_second": 8517.13
    },
    {
      "epoch": 0.8116751952598976,
      "grad_norm": 1.0015366537586678,
      "learning_rate": 6.458127065404156e-06,
      "loss": 0.36401000022888186,
      "num_input_tokens_seen": 197419392,
      "step": 12055,
      "train_runtime": 23179.0809,
      "train_tokens_per_second": 8517.136
    },
    {
      "epoch": 0.8120118502558578,
      "grad_norm": 0.9557904011757521,
      "learning_rate": 6.45559771085821e-06,
      "loss": 0.37982959747314454,
      "num_input_tokens_seen": 197501312,
      "step": 12060,
      "train_runtime": 23188.5899,
      "train_tokens_per_second": 8517.176
    },
    {
      "epoch": 0.8123485052518179,
      "grad_norm": 0.912188777739182,
      "learning_rate": 6.453067949259005e-06,
      "loss": 0.3773773193359375,
      "num_input_tokens_seen": 197583232,
      "step": 12065,
      "train_runtime": 23198.2029,
      "train_tokens_per_second": 8517.178
    },
    {
      "epoch": 0.812685160247778,
      "grad_norm": 0.9347259102749834,
      "learning_rate": 6.450537781313982e-06,
      "loss": 0.3539607048034668,
      "num_input_tokens_seen": 197665152,
      "step": 12070,
      "train_runtime": 23207.8067,
      "train_tokens_per_second": 8517.184
    },
    {
      "epoch": 0.8130218152437382,
      "grad_norm": 1.0662898365898357,
      "learning_rate": 6.448007207730693e-06,
      "loss": 0.3832110404968262,
      "num_input_tokens_seen": 197747072,
      "step": 12075,
      "train_runtime": 23217.3097,
      "train_tokens_per_second": 8517.226
    },
    {
      "epoch": 0.8133584702396983,
      "grad_norm": 1.0670933576134045,
      "learning_rate": 6.445476229216805e-06,
      "loss": 0.3876621723175049,
      "num_input_tokens_seen": 197828992,
      "step": 12080,
      "train_runtime": 23226.9377,
      "train_tokens_per_second": 8517.222
    },
    {
      "epoch": 0.8136951252356585,
      "grad_norm": 1.0719600679610373,
      "learning_rate": 6.442944846480099e-06,
      "loss": 0.42005095481872556,
      "num_input_tokens_seen": 197910912,
      "step": 12085,
      "train_runtime": 23236.5484,
      "train_tokens_per_second": 8517.225
    },
    {
      "epoch": 0.8140317802316186,
      "grad_norm": 1.2270402588357918,
      "learning_rate": 6.44041306022847e-06,
      "loss": 0.4167222023010254,
      "num_input_tokens_seen": 197992832,
      "step": 12090,
      "train_runtime": 23246.1083,
      "train_tokens_per_second": 8517.246
    },
    {
      "epoch": 0.8143684352275787,
      "grad_norm": 1.283581756454972,
      "learning_rate": 6.43788087116992e-06,
      "loss": 0.41501708030700685,
      "num_input_tokens_seen": 198074752,
      "step": 12095,
      "train_runtime": 23255.6733,
      "train_tokens_per_second": 8517.266
    },
    {
      "epoch": 0.8147050902235389,
      "grad_norm": 1.142093840844487,
      "learning_rate": 6.43534828001257e-06,
      "loss": 0.42397799491882326,
      "num_input_tokens_seen": 198156672,
      "step": 12100,
      "train_runtime": 23265.2275,
      "train_tokens_per_second": 8517.289
    },
    {
      "epoch": 0.815041745219499,
      "grad_norm": 1.0333466060593044,
      "learning_rate": 6.43281528746465e-06,
      "loss": 0.38325963020324705,
      "num_input_tokens_seen": 198238592,
      "step": 12105,
      "train_runtime": 23274.7855,
      "train_tokens_per_second": 8517.311
    },
    {
      "epoch": 0.8153784002154592,
      "grad_norm": 0.9264926469427985,
      "learning_rate": 6.430281894234504e-06,
      "loss": 0.3503131628036499,
      "num_input_tokens_seen": 198320512,
      "step": 12110,
      "train_runtime": 23284.3459,
      "train_tokens_per_second": 8517.332
    },
    {
      "epoch": 0.8157150552114193,
      "grad_norm": 0.9871264220121552,
      "learning_rate": 6.427748101030587e-06,
      "loss": 0.41138300895690916,
      "num_input_tokens_seen": 198402432,
      "step": 12115,
      "train_runtime": 23293.9481,
      "train_tokens_per_second": 8517.338
    },
    {
      "epoch": 0.8160517102073794,
      "grad_norm": 1.0395475902326745,
      "learning_rate": 6.425213908561468e-06,
      "loss": 0.3902572154998779,
      "num_input_tokens_seen": 198484352,
      "step": 12120,
      "train_runtime": 23303.5166,
      "train_tokens_per_second": 8517.356
    },
    {
      "epoch": 0.8163883652033396,
      "grad_norm": 1.058162527889334,
      "learning_rate": 6.422679317535823e-06,
      "loss": 0.36531405448913573,
      "num_input_tokens_seen": 198566272,
      "step": 12125,
      "train_runtime": 23313.0816,
      "train_tokens_per_second": 8517.376
    },
    {
      "epoch": 0.8167250201992997,
      "grad_norm": 1.2282041725669794,
      "learning_rate": 6.420144328662441e-06,
      "loss": 0.4516299724578857,
      "num_input_tokens_seen": 198648192,
      "step": 12130,
      "train_runtime": 23322.6791,
      "train_tokens_per_second": 8517.383
    },
    {
      "epoch": 0.8170616751952599,
      "grad_norm": 1.017715093116237,
      "learning_rate": 6.417608942650229e-06,
      "loss": 0.40655813217163084,
      "num_input_tokens_seen": 198730112,
      "step": 12135,
      "train_runtime": 23332.268,
      "train_tokens_per_second": 8517.394
    },
    {
      "epoch": 0.81739833019122,
      "grad_norm": 1.1530464273122851,
      "learning_rate": 6.415073160208196e-06,
      "loss": 0.40397191047668457,
      "num_input_tokens_seen": 198812032,
      "step": 12140,
      "train_runtime": 23341.8443,
      "train_tokens_per_second": 8517.409
    },
    {
      "epoch": 0.8177349851871801,
      "grad_norm": 1.0476623956880369,
      "learning_rate": 6.412536982045465e-06,
      "loss": 0.4034117698669434,
      "num_input_tokens_seen": 198893144,
      "step": 12145,
      "train_runtime": 23351.6396,
      "train_tokens_per_second": 8517.31
    },
    {
      "epoch": 0.8180716401831403,
      "grad_norm": 1.1289344964625445,
      "learning_rate": 6.410000408871273e-06,
      "loss": 0.4203841209411621,
      "num_input_tokens_seen": 198975064,
      "step": 12150,
      "train_runtime": 23361.2443,
      "train_tokens_per_second": 8517.314
    },
    {
      "epoch": 0.8184082951791004,
      "grad_norm": 1.1444389624116762,
      "learning_rate": 6.407463441394961e-06,
      "loss": 0.35439090728759765,
      "num_input_tokens_seen": 199056984,
      "step": 12155,
      "train_runtime": 23370.7562,
      "train_tokens_per_second": 8517.353
    },
    {
      "epoch": 0.8187449501750605,
      "grad_norm": 1.0107670285229327,
      "learning_rate": 6.4049260803259874e-06,
      "loss": 0.383499813079834,
      "num_input_tokens_seen": 199138904,
      "step": 12160,
      "train_runtime": 23380.3032,
      "train_tokens_per_second": 8517.379
    },
    {
      "epoch": 0.8190816051710207,
      "grad_norm": 1.1279699388523547,
      "learning_rate": 6.402388326373914e-06,
      "loss": 0.3826887607574463,
      "num_input_tokens_seen": 199220824,
      "step": 12165,
      "train_runtime": 23389.926,
      "train_tokens_per_second": 8517.377
    },
    {
      "epoch": 0.8194182601669808,
      "grad_norm": 1.0828546574002802,
      "learning_rate": 6.399850180248418e-06,
      "loss": 0.434041690826416,
      "num_input_tokens_seen": 199302744,
      "step": 12170,
      "train_runtime": 23399.4294,
      "train_tokens_per_second": 8517.419
    },
    {
      "epoch": 0.819754915162941,
      "grad_norm": 1.2134669252540535,
      "learning_rate": 6.397311642659283e-06,
      "loss": 0.4202021598815918,
      "num_input_tokens_seen": 199384664,
      "step": 12175,
      "train_runtime": 23409.0382,
      "train_tokens_per_second": 8517.422
    },
    {
      "epoch": 0.8200915701589011,
      "grad_norm": 1.2397047914079928,
      "learning_rate": 6.394772714316402e-06,
      "loss": 0.4205122470855713,
      "num_input_tokens_seen": 199466584,
      "step": 12180,
      "train_runtime": 23418.5978,
      "train_tokens_per_second": 8517.444
    },
    {
      "epoch": 0.8204282251548612,
      "grad_norm": 1.0426000371864343,
      "learning_rate": 6.392233395929779e-06,
      "loss": 0.3533951282501221,
      "num_input_tokens_seen": 199548504,
      "step": 12185,
      "train_runtime": 23428.103,
      "train_tokens_per_second": 8517.484
    },
    {
      "epoch": 0.8207648801508214,
      "grad_norm": 0.9481219296946103,
      "learning_rate": 6.389693688209526e-06,
      "loss": 0.3958866596221924,
      "num_input_tokens_seen": 199630424,
      "step": 12190,
      "train_runtime": 23437.7202,
      "train_tokens_per_second": 8517.485
    },
    {
      "epoch": 0.8211015351467815,
      "grad_norm": 1.067673653811966,
      "learning_rate": 6.387153591865867e-06,
      "loss": 0.40551605224609377,
      "num_input_tokens_seen": 199712344,
      "step": 12195,
      "train_runtime": 23447.3266,
      "train_tokens_per_second": 8517.489
    },
    {
      "epoch": 0.8214381901427417,
      "grad_norm": 1.1955613990294176,
      "learning_rate": 6.384613107609127e-06,
      "loss": 0.3806902408599854,
      "num_input_tokens_seen": 199794264,
      "step": 12200,
      "train_runtime": 23456.8784,
      "train_tokens_per_second": 8517.513
    },
    {
      "epoch": 0.8217748451387018,
      "grad_norm": 0.9592703838389383,
      "learning_rate": 6.382072236149747e-06,
      "loss": 0.40950784683227537,
      "num_input_tokens_seen": 199876184,
      "step": 12205,
      "train_runtime": 23466.4865,
      "train_tokens_per_second": 8517.516
    },
    {
      "epoch": 0.8221115001346619,
      "grad_norm": 0.9282673483064048,
      "learning_rate": 6.379530978198273e-06,
      "loss": 0.3469409465789795,
      "num_input_tokens_seen": 199958104,
      "step": 12210,
      "train_runtime": 23476.1126,
      "train_tokens_per_second": 8517.513
    },
    {
      "epoch": 0.8224481551306222,
      "grad_norm": 0.9661077023259856,
      "learning_rate": 6.376989334465361e-06,
      "loss": 0.36581876277923586,
      "num_input_tokens_seen": 200040024,
      "step": 12215,
      "train_runtime": 23485.6819,
      "train_tokens_per_second": 8517.531
    },
    {
      "epoch": 0.8227848101265823,
      "grad_norm": 0.9626948676041277,
      "learning_rate": 6.374447305661772e-06,
      "loss": 0.395578670501709,
      "num_input_tokens_seen": 200121944,
      "step": 12220,
      "train_runtime": 23495.2419,
      "train_tokens_per_second": 8517.552
    },
    {
      "epoch": 0.8231214651225425,
      "grad_norm": 0.9758171710974448,
      "learning_rate": 6.3719048924983776e-06,
      "loss": 0.35456998348236085,
      "num_input_tokens_seen": 200203840,
      "step": 12225,
      "train_runtime": 23505.4392,
      "train_tokens_per_second": 8517.341
    },
    {
      "epoch": 0.8234581201185026,
      "grad_norm": 1.1796135437693578,
      "learning_rate": 6.369362095686152e-06,
      "loss": 0.37596678733825684,
      "num_input_tokens_seen": 200285760,
      "step": 12230,
      "train_runtime": 23515.0558,
      "train_tokens_per_second": 8517.341
    },
    {
      "epoch": 0.8237947751144628,
      "grad_norm": 1.1850411908146414,
      "learning_rate": 6.366818915936185e-06,
      "loss": 0.3835036277770996,
      "num_input_tokens_seen": 200367680,
      "step": 12235,
      "train_runtime": 23524.5655,
      "train_tokens_per_second": 8517.381
    },
    {
      "epoch": 0.8241314301104229,
      "grad_norm": 1.213744141110959,
      "learning_rate": 6.364275353959667e-06,
      "loss": 0.38692240715026854,
      "num_input_tokens_seen": 200449600,
      "step": 12240,
      "train_runtime": 23534.2014,
      "train_tokens_per_second": 8517.374
    },
    {
      "epoch": 0.824468085106383,
      "grad_norm": 0.9825850891001194,
      "learning_rate": 6.3617314104678966e-06,
      "loss": 0.3559601306915283,
      "num_input_tokens_seen": 200531520,
      "step": 12245,
      "train_runtime": 23543.8026,
      "train_tokens_per_second": 8517.38
    },
    {
      "epoch": 0.8248047401023432,
      "grad_norm": 1.0318644834073387,
      "learning_rate": 6.359187086172278e-06,
      "loss": 0.3925612926483154,
      "num_input_tokens_seen": 200613104,
      "step": 12250,
      "train_runtime": 23553.7137,
      "train_tokens_per_second": 8517.26
    },
    {
      "epoch": 0.8251413950983033,
      "grad_norm": 0.9991102133227786,
      "learning_rate": 6.356642381784326e-06,
      "loss": 0.4034078598022461,
      "num_input_tokens_seen": 200695024,
      "step": 12255,
      "train_runtime": 23563.3218,
      "train_tokens_per_second": 8517.264
    },
    {
      "epoch": 0.8254780500942634,
      "grad_norm": 1.1748919237223259,
      "learning_rate": 6.354097298015658e-06,
      "loss": 0.360455846786499,
      "num_input_tokens_seen": 200776280,
      "step": 12260,
      "train_runtime": 23573.1618,
      "train_tokens_per_second": 8517.155
    },
    {
      "epoch": 0.8258147050902236,
      "grad_norm": 1.0961792645555513,
      "learning_rate": 6.3515518355779985e-06,
      "loss": 0.4229265689849854,
      "num_input_tokens_seen": 200858200,
      "step": 12265,
      "train_runtime": 23582.7168,
      "train_tokens_per_second": 8517.178
    },
    {
      "epoch": 0.8261513600861837,
      "grad_norm": 1.0443005941219694,
      "learning_rate": 6.349005995183177e-06,
      "loss": 0.37026448249816896,
      "num_input_tokens_seen": 200940120,
      "step": 12270,
      "train_runtime": 23592.2868,
      "train_tokens_per_second": 8517.196
    },
    {
      "epoch": 0.8264880150821439,
      "grad_norm": 0.9261142273846037,
      "learning_rate": 6.346459777543131e-06,
      "loss": 0.3520486354827881,
      "num_input_tokens_seen": 201022040,
      "step": 12275,
      "train_runtime": 23601.9172,
      "train_tokens_per_second": 8517.191
    },
    {
      "epoch": 0.826824670078104,
      "grad_norm": 1.1003597267446081,
      "learning_rate": 6.343913183369902e-06,
      "loss": 0.37940876483917235,
      "num_input_tokens_seen": 201103960,
      "step": 12280,
      "train_runtime": 23611.4814,
      "train_tokens_per_second": 8517.211
    },
    {
      "epoch": 0.8271613250740641,
      "grad_norm": 1.0889990808033208,
      "learning_rate": 6.341366213375637e-06,
      "loss": 0.36359481811523436,
      "num_input_tokens_seen": 201185880,
      "step": 12285,
      "train_runtime": 23621.0413,
      "train_tokens_per_second": 8517.232
    },
    {
      "epoch": 0.8274979800700243,
      "grad_norm": 1.0306930785479596,
      "learning_rate": 6.338818868272587e-06,
      "loss": 0.3915954828262329,
      "num_input_tokens_seen": 201267800,
      "step": 12290,
      "train_runtime": 23630.6048,
      "train_tokens_per_second": 8517.251
    },
    {
      "epoch": 0.8278346350659844,
      "grad_norm": 1.0486087693697996,
      "learning_rate": 6.336271148773111e-06,
      "loss": 0.41307506561279295,
      "num_input_tokens_seen": 201349720,
      "step": 12295,
      "train_runtime": 23640.1746,
      "train_tokens_per_second": 8517.269
    },
    {
      "epoch": 0.8281712900619446,
      "grad_norm": 1.1457172111626488,
      "learning_rate": 6.333723055589669e-06,
      "loss": 0.4439035415649414,
      "num_input_tokens_seen": 201431640,
      "step": 12300,
      "train_runtime": 23649.7513,
      "train_tokens_per_second": 8517.284
    },
    {
      "epoch": 0.8285079450579047,
      "grad_norm": 0.9653327008387976,
      "learning_rate": 6.331174589434826e-06,
      "loss": 0.37995400428771975,
      "num_input_tokens_seen": 201513560,
      "step": 12305,
      "train_runtime": 23659.3669,
      "train_tokens_per_second": 8517.285
    },
    {
      "epoch": 0.8288446000538648,
      "grad_norm": 1.0246719756582532,
      "learning_rate": 6.328625751021254e-06,
      "loss": 0.37443127632141116,
      "num_input_tokens_seen": 201595304,
      "step": 12310,
      "train_runtime": 23668.976,
      "train_tokens_per_second": 8517.28
    },
    {
      "epoch": 0.829181255049825,
      "grad_norm": 0.9220718628902451,
      "learning_rate": 6.326076541061729e-06,
      "loss": 0.3694486618041992,
      "num_input_tokens_seen": 201677224,
      "step": 12315,
      "train_runtime": 23678.4839,
      "train_tokens_per_second": 8517.32
    },
    {
      "epoch": 0.8295179100457851,
      "grad_norm": 1.2703203495116842,
      "learning_rate": 6.323526960269127e-06,
      "loss": 0.40633602142333985,
      "num_input_tokens_seen": 201759144,
      "step": 12320,
      "train_runtime": 23688.1142,
      "train_tokens_per_second": 8517.316
    },
    {
      "epoch": 0.8298545650417453,
      "grad_norm": 1.0597783532526814,
      "learning_rate": 6.3209770093564315e-06,
      "loss": 0.3764262437820435,
      "num_input_tokens_seen": 201841064,
      "step": 12325,
      "train_runtime": 23697.7259,
      "train_tokens_per_second": 8517.318
    },
    {
      "epoch": 0.8301912200377054,
      "grad_norm": 1.1025746951195035,
      "learning_rate": 6.318426689036727e-06,
      "loss": 0.3744697093963623,
      "num_input_tokens_seen": 201922984,
      "step": 12330,
      "train_runtime": 23707.2334,
      "train_tokens_per_second": 8517.358
    },
    {
      "epoch": 0.8305278750336655,
      "grad_norm": 1.1779714739854246,
      "learning_rate": 6.315876000023202e-06,
      "loss": 0.38825459480285646,
      "num_input_tokens_seen": 202004904,
      "step": 12335,
      "train_runtime": 23716.8466,
      "train_tokens_per_second": 8517.359
    },
    {
      "epoch": 0.8308645300296257,
      "grad_norm": 1.2182079707515698,
      "learning_rate": 6.313324943029151e-06,
      "loss": 0.4178600311279297,
      "num_input_tokens_seen": 202086824,
      "step": 12340,
      "train_runtime": 23726.4335,
      "train_tokens_per_second": 8517.37
    },
    {
      "epoch": 0.8312011850255858,
      "grad_norm": 0.9113503542966664,
      "learning_rate": 6.310773518767967e-06,
      "loss": 0.3702669620513916,
      "num_input_tokens_seen": 202168576,
      "step": 12345,
      "train_runtime": 23736.4521,
      "train_tokens_per_second": 8517.22
    },
    {
      "epoch": 0.831537840021546,
      "grad_norm": 1.414391680861735,
      "learning_rate": 6.30822172795315e-06,
      "loss": 0.402003812789917,
      "num_input_tokens_seen": 202250496,
      "step": 12350,
      "train_runtime": 23746.0768,
      "train_tokens_per_second": 8517.217
    },
    {
      "epoch": 0.8318744950175061,
      "grad_norm": 1.0982281375831697,
      "learning_rate": 6.3056695712982965e-06,
      "loss": 0.4147545337677002,
      "num_input_tokens_seen": 202331560,
      "step": 12355,
      "train_runtime": 23755.8856,
      "train_tokens_per_second": 8517.113
    },
    {
      "epoch": 0.8322111500134662,
      "grad_norm": 1.134024127482114,
      "learning_rate": 6.303117049517111e-06,
      "loss": 0.40347728729248045,
      "num_input_tokens_seen": 202413480,
      "step": 12360,
      "train_runtime": 23765.441,
      "train_tokens_per_second": 8517.135
    },
    {
      "epoch": 0.8325478050094264,
      "grad_norm": 1.0180745670919271,
      "learning_rate": 6.300564163323398e-06,
      "loss": 0.3790436267852783,
      "num_input_tokens_seen": 202495400,
      "step": 12365,
      "train_runtime": 23775.0158,
      "train_tokens_per_second": 8517.151
    },
    {
      "epoch": 0.8328844600053865,
      "grad_norm": 1.191901890298438,
      "learning_rate": 6.298010913431065e-06,
      "loss": 0.42172679901123045,
      "num_input_tokens_seen": 202577320,
      "step": 12370,
      "train_runtime": 23784.6477,
      "train_tokens_per_second": 8517.146
    },
    {
      "epoch": 0.8332211150013467,
      "grad_norm": 1.065011616283591,
      "learning_rate": 6.295457300554119e-06,
      "loss": 0.3750187158584595,
      "num_input_tokens_seen": 202659240,
      "step": 12375,
      "train_runtime": 23794.2036,
      "train_tokens_per_second": 8517.168
    },
    {
      "epoch": 0.8335577699973068,
      "grad_norm": 1.0375542052403128,
      "learning_rate": 6.292903325406668e-06,
      "loss": 0.4101543426513672,
      "num_input_tokens_seen": 202741160,
      "step": 12380,
      "train_runtime": 23803.7675,
      "train_tokens_per_second": 8517.188
    },
    {
      "epoch": 0.8338944249932669,
      "grad_norm": 1.1435818752526334,
      "learning_rate": 6.2903489887029255e-06,
      "loss": 0.3572885274887085,
      "num_input_tokens_seen": 202823080,
      "step": 12385,
      "train_runtime": 23813.3724,
      "train_tokens_per_second": 8517.193
    },
    {
      "epoch": 0.8342310799892271,
      "grad_norm": 1.1141194325196606,
      "learning_rate": 6.287794291157204e-06,
      "loss": 0.3707418441772461,
      "num_input_tokens_seen": 202905000,
      "step": 12390,
      "train_runtime": 23822.9956,
      "train_tokens_per_second": 8517.191
    },
    {
      "epoch": 0.8345677349851872,
      "grad_norm": 1.070498727039049,
      "learning_rate": 6.285239233483915e-06,
      "loss": 0.4076559066772461,
      "num_input_tokens_seen": 202986920,
      "step": 12395,
      "train_runtime": 23832.4978,
      "train_tokens_per_second": 8517.232
    },
    {
      "epoch": 0.8349043899811474,
      "grad_norm": 0.9621530161864223,
      "learning_rate": 6.2826838163975724e-06,
      "loss": 0.38537092208862306,
      "num_input_tokens_seen": 203068840,
      "step": 12400,
      "train_runtime": 23842.1364,
      "train_tokens_per_second": 8517.225
    },
    {
      "epoch": 0.8352410449771075,
      "grad_norm": 1.043866185263024,
      "learning_rate": 6.2801280406127905e-06,
      "loss": 0.38937807083129883,
      "num_input_tokens_seen": 203150760,
      "step": 12405,
      "train_runtime": 23851.7543,
      "train_tokens_per_second": 8517.225
    },
    {
      "epoch": 0.8355776999730676,
      "grad_norm": 1.1177451816672384,
      "learning_rate": 6.2775719068442845e-06,
      "loss": 0.3432033061981201,
      "num_input_tokens_seen": 203232680,
      "step": 12410,
      "train_runtime": 23861.2846,
      "train_tokens_per_second": 8517.256
    },
    {
      "epoch": 0.8359143549690278,
      "grad_norm": 1.1445857002135398,
      "learning_rate": 6.275015415806869e-06,
      "loss": 0.35495357513427733,
      "num_input_tokens_seen": 203314600,
      "step": 12415,
      "train_runtime": 23870.9217,
      "train_tokens_per_second": 8517.25
    },
    {
      "epoch": 0.8362510099649879,
      "grad_norm": 0.9424408299867487,
      "learning_rate": 6.272458568215458e-06,
      "loss": 0.42028193473815917,
      "num_input_tokens_seen": 203396520,
      "step": 12420,
      "train_runtime": 23880.552,
      "train_tokens_per_second": 8517.245
    },
    {
      "epoch": 0.836587664960948,
      "grad_norm": 1.1451186916830456,
      "learning_rate": 6.269901364785066e-06,
      "loss": 0.420613956451416,
      "num_input_tokens_seen": 203478440,
      "step": 12425,
      "train_runtime": 23890.1049,
      "train_tokens_per_second": 8517.269
    },
    {
      "epoch": 0.8369243199569082,
      "grad_norm": 1.0936821281721505,
      "learning_rate": 6.2673438062308055e-06,
      "loss": 0.4359039306640625,
      "num_input_tokens_seen": 203560360,
      "step": 12430,
      "train_runtime": 23899.6913,
      "train_tokens_per_second": 8517.28
    },
    {
      "epoch": 0.8372609749528683,
      "grad_norm": 1.1706207377160593,
      "learning_rate": 6.264785893267892e-06,
      "loss": 0.34352860450744627,
      "num_input_tokens_seen": 203640536,
      "step": 12435,
      "train_runtime": 23909.3536,
      "train_tokens_per_second": 8517.191
    },
    {
      "epoch": 0.8375976299488285,
      "grad_norm": 1.025560905119301,
      "learning_rate": 6.2622276266116366e-06,
      "loss": 0.38109307289123534,
      "num_input_tokens_seen": 203722456,
      "step": 12440,
      "train_runtime": 23918.9776,
      "train_tokens_per_second": 8517.189
    },
    {
      "epoch": 0.8379342849447886,
      "grad_norm": 0.8123264471735799,
      "learning_rate": 6.25966900697745e-06,
      "loss": 0.3804809331893921,
      "num_input_tokens_seen": 203804376,
      "step": 12445,
      "train_runtime": 23928.4885,
      "train_tokens_per_second": 8517.227
    },
    {
      "epoch": 0.8382709399407487,
      "grad_norm": 1.0827693547292387,
      "learning_rate": 6.257110035080843e-06,
      "loss": 0.3584766387939453,
      "num_input_tokens_seen": 203886296,
      "step": 12450,
      "train_runtime": 23938.1123,
      "train_tokens_per_second": 8517.225
    },
    {
      "epoch": 0.8386075949367089,
      "grad_norm": 1.0136704338218887,
      "learning_rate": 6.254550711637422e-06,
      "loss": 0.44614219665527344,
      "num_input_tokens_seen": 203968216,
      "step": 12455,
      "train_runtime": 23947.7287,
      "train_tokens_per_second": 8517.226
    },
    {
      "epoch": 0.838944249932669,
      "grad_norm": 1.3256910289348252,
      "learning_rate": 6.251991037362897e-06,
      "loss": 0.42407984733581544,
      "num_input_tokens_seen": 204050136,
      "step": 12460,
      "train_runtime": 23957.2423,
      "train_tokens_per_second": 8517.263
    },
    {
      "epoch": 0.8392809049286292,
      "grad_norm": 1.0786443944356836,
      "learning_rate": 6.249431012973068e-06,
      "loss": 0.39177899360656737,
      "num_input_tokens_seen": 204132056,
      "step": 12465,
      "train_runtime": 23966.8547,
      "train_tokens_per_second": 8517.265
    },
    {
      "epoch": 0.8396175599245893,
      "grad_norm": 0.9662952769368471,
      "learning_rate": 6.246870639183843e-06,
      "loss": 0.37220475673675535,
      "num_input_tokens_seen": 204213976,
      "step": 12470,
      "train_runtime": 23976.4711,
      "train_tokens_per_second": 8517.266
    },
    {
      "epoch": 0.8399542149205494,
      "grad_norm": 1.038233584644086,
      "learning_rate": 6.2443099167112185e-06,
      "loss": 0.4026665210723877,
      "num_input_tokens_seen": 204295896,
      "step": 12475,
      "train_runtime": 23985.98,
      "train_tokens_per_second": 8517.305
    },
    {
      "epoch": 0.8402908699165096,
      "grad_norm": 1.1359068977887432,
      "learning_rate": 6.241748846271293e-06,
      "loss": 0.4052535057067871,
      "num_input_tokens_seen": 204377816,
      "step": 12480,
      "train_runtime": 23995.5475,
      "train_tokens_per_second": 8517.322
    },
    {
      "epoch": 0.8406275249124697,
      "grad_norm": 1.208100063099847,
      "learning_rate": 6.239187428580263e-06,
      "loss": 0.41015052795410156,
      "num_input_tokens_seen": 204459520,
      "step": 12485,
      "train_runtime": 24005.6737,
      "train_tokens_per_second": 8517.133
    },
    {
      "epoch": 0.8409641799084299,
      "grad_norm": 1.249550232874411,
      "learning_rate": 6.236625664354421e-06,
      "loss": 0.4093649864196777,
      "num_input_tokens_seen": 204541440,
      "step": 12490,
      "train_runtime": 24015.235,
      "train_tokens_per_second": 8517.153
    },
    {
      "epoch": 0.84130083490439,
      "grad_norm": 1.0290738041037568,
      "learning_rate": 6.2340635543101545e-06,
      "loss": 0.35519957542419434,
      "num_input_tokens_seen": 204623360,
      "step": 12495,
      "train_runtime": 24024.8025,
      "train_tokens_per_second": 8517.171
    },
    {
      "epoch": 0.8416374899003501,
      "grad_norm": 1.2436223467300287,
      "learning_rate": 6.23150109916395e-06,
      "loss": 0.38255503177642824,
      "num_input_tokens_seen": 204705280,
      "step": 12500,
      "train_runtime": 24034.4114,
      "train_tokens_per_second": 8517.175
    },
    {
      "epoch": 0.8419741448963103,
      "grad_norm": 0.9851478265605231,
      "learning_rate": 6.22893829963239e-06,
      "loss": 0.3604919195175171,
      "num_input_tokens_seen": 204787200,
      "step": 12505,
      "train_runtime": 24043.9763,
      "train_tokens_per_second": 8517.194
    },
    {
      "epoch": 0.8423107998922704,
      "grad_norm": 1.2499128403376476,
      "learning_rate": 6.226375156432153e-06,
      "loss": 0.4043235778808594,
      "num_input_tokens_seen": 204869120,
      "step": 12510,
      "train_runtime": 24053.5568,
      "train_tokens_per_second": 8517.207
    },
    {
      "epoch": 0.8426474548882306,
      "grad_norm": 0.9186092873272774,
      "learning_rate": 6.2238116702800145e-06,
      "loss": 0.38563032150268556,
      "num_input_tokens_seen": 204951040,
      "step": 12515,
      "train_runtime": 24063.1825,
      "train_tokens_per_second": 8517.204
    },
    {
      "epoch": 0.8429841098841907,
      "grad_norm": 1.5965849107826127,
      "learning_rate": 6.221247841892846e-06,
      "loss": 0.38136603832244875,
      "num_input_tokens_seen": 205032960,
      "step": 12520,
      "train_runtime": 24072.8033,
      "train_tokens_per_second": 8517.203
    },
    {
      "epoch": 0.8433207648801508,
      "grad_norm": 1.2176550549077452,
      "learning_rate": 6.218683671987611e-06,
      "loss": 0.43294601440429686,
      "num_input_tokens_seen": 205114880,
      "step": 12525,
      "train_runtime": 24082.3128,
      "train_tokens_per_second": 8517.242
    },
    {
      "epoch": 0.843657419876111,
      "grad_norm": 1.1641971561712545,
      "learning_rate": 6.216119161281373e-06,
      "loss": 0.4296513557434082,
      "num_input_tokens_seen": 205196800,
      "step": 12530,
      "train_runtime": 24091.9239,
      "train_tokens_per_second": 8517.244
    },
    {
      "epoch": 0.8439940748720711,
      "grad_norm": 0.9758125771885114,
      "learning_rate": 6.213554310491291e-06,
      "loss": 0.3587090492248535,
      "num_input_tokens_seen": 205278720,
      "step": 12535,
      "train_runtime": 24101.5554,
      "train_tokens_per_second": 8517.24
    },
    {
      "epoch": 0.8443307298680313,
      "grad_norm": 1.1037187722397845,
      "learning_rate": 6.210989120334616e-06,
      "loss": 0.4120126724243164,
      "num_input_tokens_seen": 205360640,
      "step": 12540,
      "train_runtime": 24111.0623,
      "train_tokens_per_second": 8517.279
    },
    {
      "epoch": 0.8446673848639914,
      "grad_norm": 1.5141948026269907,
      "learning_rate": 6.208423591528694e-06,
      "loss": 0.38531808853149413,
      "num_input_tokens_seen": 205442560,
      "step": 12545,
      "train_runtime": 24120.6721,
      "train_tokens_per_second": 8517.282
    },
    {
      "epoch": 0.8450040398599515,
      "grad_norm": 1.0702784321314733,
      "learning_rate": 6.205857724790969e-06,
      "loss": 0.36019234657287597,
      "num_input_tokens_seen": 205524208,
      "step": 12550,
      "train_runtime": 24130.2903,
      "train_tokens_per_second": 8517.27
    },
    {
      "epoch": 0.8453406948559117,
      "grad_norm": 1.0721553828257369,
      "learning_rate": 6.203291520838974e-06,
      "loss": 0.41349115371704104,
      "num_input_tokens_seen": 205606128,
      "step": 12555,
      "train_runtime": 24139.7998,
      "train_tokens_per_second": 8517.309
    },
    {
      "epoch": 0.8456773498518718,
      "grad_norm": 1.0765644946650115,
      "learning_rate": 6.200724980390344e-06,
      "loss": 0.35385828018188475,
      "num_input_tokens_seen": 205687904,
      "step": 12560,
      "train_runtime": 24149.8512,
      "train_tokens_per_second": 8517.15
    },
    {
      "epoch": 0.846014004847832,
      "grad_norm": 1.014810661973727,
      "learning_rate": 6.198158104162804e-06,
      "loss": 0.35282297134399415,
      "num_input_tokens_seen": 205769824,
      "step": 12565,
      "train_runtime": 24159.3974,
      "train_tokens_per_second": 8517.175
    },
    {
      "epoch": 0.8463506598437921,
      "grad_norm": 1.0853902602638903,
      "learning_rate": 6.1955908928741685e-06,
      "loss": 0.4100193500518799,
      "num_input_tokens_seen": 205851744,
      "step": 12570,
      "train_runtime": 24168.9594,
      "train_tokens_per_second": 8517.195
    },
    {
      "epoch": 0.8466873148397522,
      "grad_norm": 1.1674494995403932,
      "learning_rate": 6.193023347242353e-06,
      "loss": 0.4032124042510986,
      "num_input_tokens_seen": 205933664,
      "step": 12575,
      "train_runtime": 24178.5263,
      "train_tokens_per_second": 8517.213
    },
    {
      "epoch": 0.8470239698357124,
      "grad_norm": 0.9972769552063951,
      "learning_rate": 6.190455467985361e-06,
      "loss": 0.4085115432739258,
      "num_input_tokens_seen": 206015584,
      "step": 12580,
      "train_runtime": 24188.15,
      "train_tokens_per_second": 8517.211
    },
    {
      "epoch": 0.8473606248316725,
      "grad_norm": 1.0254523054579296,
      "learning_rate": 6.187887255821295e-06,
      "loss": 0.39946975708007815,
      "num_input_tokens_seen": 206097504,
      "step": 12585,
      "train_runtime": 24197.767,
      "train_tokens_per_second": 8517.212
    },
    {
      "epoch": 0.8476972798276327,
      "grad_norm": 0.9559065219614445,
      "learning_rate": 6.185318711468345e-06,
      "loss": 0.37187628746032714,
      "num_input_tokens_seen": 206179424,
      "step": 12590,
      "train_runtime": 24207.2779,
      "train_tokens_per_second": 8517.249
    },
    {
      "epoch": 0.8480339348235928,
      "grad_norm": 0.9579947162889101,
      "learning_rate": 6.182749835644799e-06,
      "loss": 0.38562028408050536,
      "num_input_tokens_seen": 206261176,
      "step": 12595,
      "train_runtime": 24217.4412,
      "train_tokens_per_second": 8517.051
    },
    {
      "epoch": 0.8483705898195529,
      "grad_norm": 1.0532581599654325,
      "learning_rate": 6.180180629069031e-06,
      "loss": 0.36572628021240233,
      "num_input_tokens_seen": 206343096,
      "step": 12600,
      "train_runtime": 24227.084,
      "train_tokens_per_second": 8517.042
    },
    {
      "epoch": 0.8487072448155131,
      "grad_norm": 1.0818777241213493,
      "learning_rate": 6.177611092459514e-06,
      "loss": 0.38304476737976073,
      "num_input_tokens_seen": 206425016,
      "step": 12605,
      "train_runtime": 24236.5938,
      "train_tokens_per_second": 8517.08
    },
    {
      "epoch": 0.8490438998114732,
      "grad_norm": 0.9763276071978552,
      "learning_rate": 6.175041226534809e-06,
      "loss": 0.3938347339630127,
      "num_input_tokens_seen": 206506936,
      "step": 12610,
      "train_runtime": 24246.2199,
      "train_tokens_per_second": 8517.078
    },
    {
      "epoch": 0.8493805548074334,
      "grad_norm": 1.0607191598511,
      "learning_rate": 6.172471032013574e-06,
      "loss": 0.3613172769546509,
      "num_input_tokens_seen": 206588856,
      "step": 12615,
      "train_runtime": 24255.8355,
      "train_tokens_per_second": 8517.079
    },
    {
      "epoch": 0.8497172098033935,
      "grad_norm": 0.9927413022572201,
      "learning_rate": 6.169900509614553e-06,
      "loss": 0.38379502296447754,
      "num_input_tokens_seen": 206670776,
      "step": 12620,
      "train_runtime": 24265.3849,
      "train_tokens_per_second": 8517.103
    },
    {
      "epoch": 0.8500538647993536,
      "grad_norm": 0.9818221139858906,
      "learning_rate": 6.167329660056585e-06,
      "loss": 0.39607911109924315,
      "num_input_tokens_seen": 206752696,
      "step": 12625,
      "train_runtime": 24274.9534,
      "train_tokens_per_second": 8517.12
    },
    {
      "epoch": 0.8503905197953138,
      "grad_norm": 1.0418229939789005,
      "learning_rate": 6.1647584840586e-06,
      "loss": 0.3966355562210083,
      "num_input_tokens_seen": 206834616,
      "step": 12630,
      "train_runtime": 24284.5875,
      "train_tokens_per_second": 8517.115
    },
    {
      "epoch": 0.8507271747912739,
      "grad_norm": 1.008037737838642,
      "learning_rate": 6.1621869823396175e-06,
      "loss": 0.365296745300293,
      "num_input_tokens_seen": 206916536,
      "step": 12635,
      "train_runtime": 24294.1396,
      "train_tokens_per_second": 8517.138
    },
    {
      "epoch": 0.851063829787234,
      "grad_norm": 1.006827502432729,
      "learning_rate": 6.159615155618752e-06,
      "loss": 0.4125082015991211,
      "num_input_tokens_seen": 206998456,
      "step": 12640,
      "train_runtime": 24303.7092,
      "train_tokens_per_second": 8517.155
    },
    {
      "epoch": 0.8514004847831942,
      "grad_norm": 1.3866341134859304,
      "learning_rate": 6.157043004615207e-06,
      "loss": 0.429721736907959,
      "num_input_tokens_seen": 207080376,
      "step": 12645,
      "train_runtime": 24313.3176,
      "train_tokens_per_second": 8517.158
    },
    {
      "epoch": 0.8517371397791543,
      "grad_norm": 1.267722945888068,
      "learning_rate": 6.1544705300482755e-06,
      "loss": 0.397010350227356,
      "num_input_tokens_seen": 207162296,
      "step": 12650,
      "train_runtime": 24322.9196,
      "train_tokens_per_second": 8517.164
    },
    {
      "epoch": 0.8520737947751145,
      "grad_norm": 1.0504057989437596,
      "learning_rate": 6.151897732637341e-06,
      "loss": 0.3667948007583618,
      "num_input_tokens_seen": 207244216,
      "step": 12655,
      "train_runtime": 24332.7347,
      "train_tokens_per_second": 8517.095
    },
    {
      "epoch": 0.8524104497710746,
      "grad_norm": 0.9648318331241333,
      "learning_rate": 6.149324613101877e-06,
      "loss": 0.3911270618438721,
      "num_input_tokens_seen": 207326136,
      "step": 12660,
      "train_runtime": 24342.3325,
      "train_tokens_per_second": 8517.102
    },
    {
      "epoch": 0.8527471047670347,
      "grad_norm": 0.972699718474207,
      "learning_rate": 6.146751172161453e-06,
      "loss": 0.3483403205871582,
      "num_input_tokens_seen": 207408056,
      "step": 12665,
      "train_runtime": 24351.8834,
      "train_tokens_per_second": 8517.126
    },
    {
      "epoch": 0.8530837597629949,
      "grad_norm": 1.24529556610003,
      "learning_rate": 6.144177410535719e-06,
      "loss": 0.3742656946182251,
      "num_input_tokens_seen": 207489784,
      "step": 12670,
      "train_runtime": 24362.0808,
      "train_tokens_per_second": 8516.916
    },
    {
      "epoch": 0.853420414758955,
      "grad_norm": 0.8480831557194123,
      "learning_rate": 6.14160332894442e-06,
      "loss": 0.3897249698638916,
      "num_input_tokens_seen": 207571704,
      "step": 12675,
      "train_runtime": 24371.709,
      "train_tokens_per_second": 8516.912
    },
    {
      "epoch": 0.8537570697549152,
      "grad_norm": 1.084563739739534,
      "learning_rate": 6.139028928107391e-06,
      "loss": 0.3788467884063721,
      "num_input_tokens_seen": 207653624,
      "step": 12680,
      "train_runtime": 24381.3215,
      "train_tokens_per_second": 8516.914
    },
    {
      "epoch": 0.8540937247508753,
      "grad_norm": 1.3989562394650255,
      "learning_rate": 6.136454208744553e-06,
      "loss": 0.343404221534729,
      "num_input_tokens_seen": 207735544,
      "step": 12685,
      "train_runtime": 24390.8725,
      "train_tokens_per_second": 8516.938
    },
    {
      "epoch": 0.8544303797468354,
      "grad_norm": 1.3044555735321477,
      "learning_rate": 6.133879171575921e-06,
      "loss": 0.4109031677246094,
      "num_input_tokens_seen": 207817464,
      "step": 12690,
      "train_runtime": 24400.655,
      "train_tokens_per_second": 8516.881
    },
    {
      "epoch": 0.8547670347427956,
      "grad_norm": 0.9905490669439904,
      "learning_rate": 6.131303817321595e-06,
      "loss": 0.4308831214904785,
      "num_input_tokens_seen": 207899384,
      "step": 12695,
      "train_runtime": 24410.2815,
      "train_tokens_per_second": 8516.878
    },
    {
      "epoch": 0.8551036897387557,
      "grad_norm": 1.0253859121624858,
      "learning_rate": 6.128728146701763e-06,
      "loss": 0.41512699127197267,
      "num_input_tokens_seen": 207981304,
      "step": 12700,
      "train_runtime": 24419.8558,
      "train_tokens_per_second": 8516.893
    },
    {
      "epoch": 0.8554403447347159,
      "grad_norm": 0.9351374722490997,
      "learning_rate": 6.126152160436705e-06,
      "loss": 0.3627490043640137,
      "num_input_tokens_seen": 208063224,
      "step": 12705,
      "train_runtime": 24429.3733,
      "train_tokens_per_second": 8516.928
    },
    {
      "epoch": 0.855776999730676,
      "grad_norm": 1.172165086705657,
      "learning_rate": 6.123575859246784e-06,
      "loss": 0.38263764381408694,
      "num_input_tokens_seen": 208145144,
      "step": 12710,
      "train_runtime": 24438.9489,
      "train_tokens_per_second": 8516.943
    },
    {
      "epoch": 0.8561136547266361,
      "grad_norm": 1.1394504237605463,
      "learning_rate": 6.120999243852459e-06,
      "loss": 0.38510804176330565,
      "num_input_tokens_seen": 208227064,
      "step": 12715,
      "train_runtime": 24448.7614,
      "train_tokens_per_second": 8516.876
    },
    {
      "epoch": 0.8564503097225963,
      "grad_norm": 1.0794016480707622,
      "learning_rate": 6.118422314974269e-06,
      "loss": 0.4187830924987793,
      "num_input_tokens_seen": 208308984,
      "step": 12720,
      "train_runtime": 24458.269,
      "train_tokens_per_second": 8516.914
    },
    {
      "epoch": 0.8567869647185564,
      "grad_norm": 1.2167846896943582,
      "learning_rate": 6.115845073332845e-06,
      "loss": 0.38674187660217285,
      "num_input_tokens_seen": 208390904,
      "step": 12725,
      "train_runtime": 24467.8908,
      "train_tokens_per_second": 8516.913
    },
    {
      "epoch": 0.8571236197145166,
      "grad_norm": 1.1196298847893655,
      "learning_rate": 6.113267519648905e-06,
      "loss": 0.45670170783996583,
      "num_input_tokens_seen": 208472824,
      "step": 12730,
      "train_runtime": 24477.524,
      "train_tokens_per_second": 8516.908
    },
    {
      "epoch": 0.8574602747104767,
      "grad_norm": 0.8965949624755021,
      "learning_rate": 6.110689654643253e-06,
      "loss": 0.35608079433441164,
      "num_input_tokens_seen": 208554744,
      "step": 12735,
      "train_runtime": 24487.0368,
      "train_tokens_per_second": 8516.945
    },
    {
      "epoch": 0.8577969297064368,
      "grad_norm": 0.9365377244152477,
      "learning_rate": 6.1081114790367805e-06,
      "loss": 0.3701732635498047,
      "num_input_tokens_seen": 208636664,
      "step": 12740,
      "train_runtime": 24496.6479,
      "train_tokens_per_second": 8516.947
    },
    {
      "epoch": 0.858133584702397,
      "grad_norm": 0.9192347604127943,
      "learning_rate": 6.105532993550467e-06,
      "loss": 0.3988960742950439,
      "num_input_tokens_seen": 208718584,
      "step": 12745,
      "train_runtime": 24506.2576,
      "train_tokens_per_second": 8516.951
    },
    {
      "epoch": 0.8584702396983571,
      "grad_norm": 0.988887319907229,
      "learning_rate": 6.1029541989053765e-06,
      "loss": 0.37209904193878174,
      "num_input_tokens_seen": 208800504,
      "step": 12750,
      "train_runtime": 24515.8119,
      "train_tokens_per_second": 8516.973
    },
    {
      "epoch": 0.8588068946943173,
      "grad_norm": 0.9878974417585824,
      "learning_rate": 6.100375095822661e-06,
      "loss": 0.37455215454101565,
      "num_input_tokens_seen": 208882424,
      "step": 12755,
      "train_runtime": 24525.378,
      "train_tokens_per_second": 8516.991
    },
    {
      "epoch": 0.8591435496902774,
      "grad_norm": 0.9467337740782716,
      "learning_rate": 6.097795685023558e-06,
      "loss": 0.3663607358932495,
      "num_input_tokens_seen": 208964344,
      "step": 12760,
      "train_runtime": 24535.0064,
      "train_tokens_per_second": 8516.988
    },
    {
      "epoch": 0.8594802046862375,
      "grad_norm": 0.986395230506417,
      "learning_rate": 6.095215967229396e-06,
      "loss": 0.37414090633392333,
      "num_input_tokens_seen": 209046264,
      "step": 12765,
      "train_runtime": 24544.5699,
      "train_tokens_per_second": 8517.007
    },
    {
      "epoch": 0.8598168596821977,
      "grad_norm": 1.0751763213355119,
      "learning_rate": 6.092635943161578e-06,
      "loss": 0.3935889720916748,
      "num_input_tokens_seen": 209128184,
      "step": 12770,
      "train_runtime": 24554.1298,
      "train_tokens_per_second": 8517.027
    },
    {
      "epoch": 0.8601535146781578,
      "grad_norm": 0.9988729226247074,
      "learning_rate": 6.090055613541603e-06,
      "loss": 0.3850706100463867,
      "num_input_tokens_seen": 209210104,
      "step": 12775,
      "train_runtime": 24563.7414,
      "train_tokens_per_second": 8517.029
    },
    {
      "epoch": 0.860490169674118,
      "grad_norm": 1.154986965494469,
      "learning_rate": 6.087474979091052e-06,
      "loss": 0.36183667182922363,
      "num_input_tokens_seen": 209291456,
      "step": 12780,
      "train_runtime": 24573.5877,
      "train_tokens_per_second": 8516.927
    },
    {
      "epoch": 0.8608268246700781,
      "grad_norm": 1.2184391478875918,
      "learning_rate": 6.084894040531591e-06,
      "loss": 0.3522505283355713,
      "num_input_tokens_seen": 209373376,
      "step": 12785,
      "train_runtime": 24583.1537,
      "train_tokens_per_second": 8516.945
    },
    {
      "epoch": 0.8611634796660382,
      "grad_norm": 1.063481360820538,
      "learning_rate": 6.0823127985849705e-06,
      "loss": 0.3700291156768799,
      "num_input_tokens_seen": 209454296,
      "step": 12790,
      "train_runtime": 24592.7721,
      "train_tokens_per_second": 8516.905
    },
    {
      "epoch": 0.8615001346619984,
      "grad_norm": 1.1151457857634808,
      "learning_rate": 6.079731253973028e-06,
      "loss": 0.36481671333312987,
      "num_input_tokens_seen": 209536216,
      "step": 12795,
      "train_runtime": 24602.3873,
      "train_tokens_per_second": 8516.906
    },
    {
      "epoch": 0.8618367896579585,
      "grad_norm": 1.1930661400296838,
      "learning_rate": 6.077149407417683e-06,
      "loss": 0.38528549671173096,
      "num_input_tokens_seen": 209618136,
      "step": 12800,
      "train_runtime": 24611.8925,
      "train_tokens_per_second": 8516.945
    },
    {
      "epoch": 0.8621734446539187,
      "grad_norm": 1.1372809583324768,
      "learning_rate": 6.07456725964094e-06,
      "loss": 0.3991963863372803,
      "num_input_tokens_seen": 209700056,
      "step": 12805,
      "train_runtime": 24621.5222,
      "train_tokens_per_second": 8516.941
    },
    {
      "epoch": 0.8625100996498788,
      "grad_norm": 0.9355957294894782,
      "learning_rate": 6.07198481136489e-06,
      "loss": 0.3565434455871582,
      "num_input_tokens_seen": 209781976,
      "step": 12810,
      "train_runtime": 24631.1407,
      "train_tokens_per_second": 8516.941
    },
    {
      "epoch": 0.8628467546458389,
      "grad_norm": 0.9477886166156955,
      "learning_rate": 6.069402063311706e-06,
      "loss": 0.341782808303833,
      "num_input_tokens_seen": 209863896,
      "step": 12815,
      "train_runtime": 24640.6412,
      "train_tokens_per_second": 8516.982
    },
    {
      "epoch": 0.8631834096417991,
      "grad_norm": 6.960119575989202,
      "learning_rate": 6.066819016203646e-06,
      "loss": 0.39874954223632814,
      "num_input_tokens_seen": 209945816,
      "step": 12820,
      "train_runtime": 24650.2702,
      "train_tokens_per_second": 8516.978
    },
    {
      "epoch": 0.8635200646377592,
      "grad_norm": 1.0102137211536524,
      "learning_rate": 6.064235670763048e-06,
      "loss": 0.3995970726013184,
      "num_input_tokens_seen": 210027736,
      "step": 12825,
      "train_runtime": 24659.8913,
      "train_tokens_per_second": 8516.977
    },
    {
      "epoch": 0.8638567196337193,
      "grad_norm": 1.1122391733692416,
      "learning_rate": 6.06165202771234e-06,
      "loss": 0.41887388229370115,
      "num_input_tokens_seen": 210109656,
      "step": 12830,
      "train_runtime": 24669.4013,
      "train_tokens_per_second": 8517.015
    },
    {
      "epoch": 0.8641933746296795,
      "grad_norm": 0.8734765193034626,
      "learning_rate": 6.059068087774026e-06,
      "loss": 0.3529968738555908,
      "num_input_tokens_seen": 210191576,
      "step": 12835,
      "train_runtime": 24679.0398,
      "train_tokens_per_second": 8517.008
    },
    {
      "epoch": 0.8645300296256396,
      "grad_norm": 0.9399958755989317,
      "learning_rate": 6.0564838516707005e-06,
      "loss": 0.40218229293823243,
      "num_input_tokens_seen": 210273496,
      "step": 12840,
      "train_runtime": 24688.6582,
      "train_tokens_per_second": 8517.008
    },
    {
      "epoch": 0.8648666846215998,
      "grad_norm": 0.8431016481325407,
      "learning_rate": 6.053899320125033e-06,
      "loss": 0.3454217672348022,
      "num_input_tokens_seen": 210355416,
      "step": 12845,
      "train_runtime": 24698.2224,
      "train_tokens_per_second": 8517.027
    },
    {
      "epoch": 0.8652033396175599,
      "grad_norm": 1.2638535680229515,
      "learning_rate": 6.051314493859782e-06,
      "loss": 0.4248453140258789,
      "num_input_tokens_seen": 210437336,
      "step": 12850,
      "train_runtime": 24707.792,
      "train_tokens_per_second": 8517.043
    },
    {
      "epoch": 0.86553999461352,
      "grad_norm": 0.9733718218594603,
      "learning_rate": 6.048729373597786e-06,
      "loss": 0.3699317455291748,
      "num_input_tokens_seen": 210519256,
      "step": 12855,
      "train_runtime": 24717.4314,
      "train_tokens_per_second": 8517.036
    },
    {
      "epoch": 0.8658766496094802,
      "grad_norm": 1.042288304928766,
      "learning_rate": 6.046143960061963e-06,
      "loss": 0.3595315933227539,
      "num_input_tokens_seen": 210600712,
      "step": 12860,
      "train_runtime": 24727.3223,
      "train_tokens_per_second": 8516.923
    },
    {
      "epoch": 0.8662133046054403,
      "grad_norm": 1.1488589507402658,
      "learning_rate": 6.043558253975319e-06,
      "loss": 0.40372447967529296,
      "num_input_tokens_seen": 210682632,
      "step": 12865,
      "train_runtime": 24736.8614,
      "train_tokens_per_second": 8516.951
    },
    {
      "epoch": 0.8665499596014005,
      "grad_norm": 1.0381735574641524,
      "learning_rate": 6.040972256060936e-06,
      "loss": 0.3791553020477295,
      "num_input_tokens_seen": 210764552,
      "step": 12870,
      "train_runtime": 24746.4785,
      "train_tokens_per_second": 8516.951
    },
    {
      "epoch": 0.8668866145973606,
      "grad_norm": 1.024151963439361,
      "learning_rate": 6.038385967041982e-06,
      "loss": 0.3995301961898804,
      "num_input_tokens_seen": 210846472,
      "step": 12875,
      "train_runtime": 24756.0995,
      "train_tokens_per_second": 8516.95
    },
    {
      "epoch": 0.8672232695933207,
      "grad_norm": 0.9964088962982749,
      "learning_rate": 6.035799387641703e-06,
      "loss": 0.37049198150634766,
      "num_input_tokens_seen": 210928392,
      "step": 12880,
      "train_runtime": 24765.6072,
      "train_tokens_per_second": 8516.989
    },
    {
      "epoch": 0.8675599245892809,
      "grad_norm": 1.0699245025859971,
      "learning_rate": 6.033212518583427e-06,
      "loss": 0.39513068199157714,
      "num_input_tokens_seen": 211010312,
      "step": 12885,
      "train_runtime": 24775.2279,
      "train_tokens_per_second": 8516.988
    },
    {
      "epoch": 0.867896579585241,
      "grad_norm": 1.0454545726509514,
      "learning_rate": 6.030625360590567e-06,
      "loss": 0.3762871265411377,
      "num_input_tokens_seen": 211092232,
      "step": 12890,
      "train_runtime": 24784.8002,
      "train_tokens_per_second": 8517.004
    },
    {
      "epoch": 0.8682332345812012,
      "grad_norm": 0.9511235169318901,
      "learning_rate": 6.028037914386609e-06,
      "loss": 0.341230583190918,
      "num_input_tokens_seen": 211174152,
      "step": 12895,
      "train_runtime": 24794.3101,
      "train_tokens_per_second": 8517.041
    },
    {
      "epoch": 0.8685698895771613,
      "grad_norm": 0.9832662098471635,
      "learning_rate": 6.025450180695128e-06,
      "loss": 0.4097433567047119,
      "num_input_tokens_seen": 211256072,
      "step": 12900,
      "train_runtime": 24803.9252,
      "train_tokens_per_second": 8517.042
    },
    {
      "epoch": 0.8689065445731214,
      "grad_norm": 1.1306897704071193,
      "learning_rate": 6.022862160239774e-06,
      "loss": 0.3573839902877808,
      "num_input_tokens_seen": 211337992,
      "step": 12905,
      "train_runtime": 24813.5623,
      "train_tokens_per_second": 8517.036
    },
    {
      "epoch": 0.8692431995690816,
      "grad_norm": 0.9765382332255762,
      "learning_rate": 6.020273853744277e-06,
      "loss": 0.36747925281524657,
      "num_input_tokens_seen": 211419912,
      "step": 12910,
      "train_runtime": 24823.0686,
      "train_tokens_per_second": 8517.074
    },
    {
      "epoch": 0.8695798545650417,
      "grad_norm": 1.0734053310824607,
      "learning_rate": 6.017685261932452e-06,
      "loss": 0.3605794906616211,
      "num_input_tokens_seen": 211501592,
      "step": 12915,
      "train_runtime": 24833.199,
      "train_tokens_per_second": 8516.889
    },
    {
      "epoch": 0.8699165095610019,
      "grad_norm": 1.0125207663286042,
      "learning_rate": 6.015096385528189e-06,
      "loss": 0.4061676025390625,
      "num_input_tokens_seen": 211583512,
      "step": 12920,
      "train_runtime": 24842.8264,
      "train_tokens_per_second": 8516.886
    },
    {
      "epoch": 0.870253164556962,
      "grad_norm": 1.2219924570321596,
      "learning_rate": 6.012507225255457e-06,
      "loss": 0.4059577941894531,
      "num_input_tokens_seen": 211665432,
      "step": 12925,
      "train_runtime": 24852.3773,
      "train_tokens_per_second": 8516.909
    },
    {
      "epoch": 0.8705898195529221,
      "grad_norm": 1.174835305890871,
      "learning_rate": 6.009917781838309e-06,
      "loss": 0.3827080249786377,
      "num_input_tokens_seen": 211747352,
      "step": 12930,
      "train_runtime": 24861.9517,
      "train_tokens_per_second": 8516.924
    },
    {
      "epoch": 0.8709264745488823,
      "grad_norm": 1.0731750693658717,
      "learning_rate": 6.007328056000873e-06,
      "loss": 0.3908916711807251,
      "num_input_tokens_seen": 211829272,
      "step": 12935,
      "train_runtime": 24871.578,
      "train_tokens_per_second": 8516.921
    },
    {
      "epoch": 0.8712631295448424,
      "grad_norm": 1.1025548365821445,
      "learning_rate": 6.004738048467359e-06,
      "loss": 0.39035496711730955,
      "num_input_tokens_seen": 211911192,
      "step": 12940,
      "train_runtime": 24881.1472,
      "train_tokens_per_second": 8516.938
    },
    {
      "epoch": 0.8715997845408026,
      "grad_norm": 1.264546682941175,
      "learning_rate": 6.002147759962054e-06,
      "loss": 0.3933008432388306,
      "num_input_tokens_seen": 211993112,
      "step": 12945,
      "train_runtime": 24890.7315,
      "train_tokens_per_second": 8516.95
    },
    {
      "epoch": 0.8719364395367627,
      "grad_norm": 1.090838809397159,
      "learning_rate": 5.999557191209323e-06,
      "loss": 0.4247895240783691,
      "num_input_tokens_seen": 212075032,
      "step": 12950,
      "train_runtime": 24900.3464,
      "train_tokens_per_second": 8516.951
    },
    {
      "epoch": 0.8722730945327228,
      "grad_norm": 1.0742233106335413,
      "learning_rate": 5.996966342933611e-06,
      "loss": 0.34872260093688967,
      "num_input_tokens_seen": 212156952,
      "step": 12955,
      "train_runtime": 24909.9809,
      "train_tokens_per_second": 8516.946
    },
    {
      "epoch": 0.872609749528683,
      "grad_norm": 1.0170990376826845,
      "learning_rate": 5.99437521585944e-06,
      "loss": 0.3815558671951294,
      "num_input_tokens_seen": 212238872,
      "step": 12960,
      "train_runtime": 24919.4924,
      "train_tokens_per_second": 8516.982
    },
    {
      "epoch": 0.8729464045246431,
      "grad_norm": 1.2368623515223593,
      "learning_rate": 5.99178381071141e-06,
      "loss": 0.4271837711334229,
      "num_input_tokens_seen": 212320792,
      "step": 12965,
      "train_runtime": 24929.1183,
      "train_tokens_per_second": 8516.98
    },
    {
      "epoch": 0.8732830595206033,
      "grad_norm": 1.046146789905438,
      "learning_rate": 5.989192128214201e-06,
      "loss": 0.3787795066833496,
      "num_input_tokens_seen": 212402712,
      "step": 12970,
      "train_runtime": 24938.7529,
      "train_tokens_per_second": 8516.974
    },
    {
      "epoch": 0.8736197145165634,
      "grad_norm": 1.0255461706212052,
      "learning_rate": 5.986600169092568e-06,
      "loss": 0.34870476722717286,
      "num_input_tokens_seen": 212484632,
      "step": 12975,
      "train_runtime": 24948.2795,
      "train_tokens_per_second": 8517.005
    },
    {
      "epoch": 0.8739563695125235,
      "grad_norm": 1.0564813119928724,
      "learning_rate": 5.984007934071343e-06,
      "loss": 0.4178572177886963,
      "num_input_tokens_seen": 212566552,
      "step": 12980,
      "train_runtime": 24957.9243,
      "train_tokens_per_second": 8516.996
    },
    {
      "epoch": 0.8742930245084837,
      "grad_norm": 1.031269243944212,
      "learning_rate": 5.981415423875436e-06,
      "loss": 0.3572450876235962,
      "num_input_tokens_seen": 212648472,
      "step": 12985,
      "train_runtime": 24967.5461,
      "train_tokens_per_second": 8516.995
    },
    {
      "epoch": 0.8746296795044438,
      "grad_norm": 1.1898805489849587,
      "learning_rate": 5.978822639229836e-06,
      "loss": 0.38425345420837403,
      "num_input_tokens_seen": 212730392,
      "step": 12990,
      "train_runtime": 24977.0983,
      "train_tokens_per_second": 8517.018
    },
    {
      "epoch": 0.874966334500404,
      "grad_norm": 0.9885859305079346,
      "learning_rate": 5.976229580859607e-06,
      "loss": 0.3741785526275635,
      "num_input_tokens_seen": 212812312,
      "step": 12995,
      "train_runtime": 24986.6698,
      "train_tokens_per_second": 8517.034
    },
    {
      "epoch": 0.8753029894963641,
      "grad_norm": 1.1434293776074398,
      "learning_rate": 5.9736362494898895e-06,
      "loss": 0.33244478702545166,
      "num_input_tokens_seen": 212894232,
      "step": 13000,
      "train_runtime": 24996.2395,
      "train_tokens_per_second": 8517.05
    },
    {
      "epoch": 0.8756396444923242,
      "grad_norm": 1.0283179509144218,
      "learning_rate": 5.971042645845897e-06,
      "loss": 0.3788741588592529,
      "num_input_tokens_seen": 212975984,
      "step": 13005,
      "train_runtime": 25006.2425,
      "train_tokens_per_second": 8516.913
    },
    {
      "epoch": 0.8759762994882844,
      "grad_norm": 0.8554155282287854,
      "learning_rate": 5.968448770652926e-06,
      "loss": 0.3932424306869507,
      "num_input_tokens_seen": 213057904,
      "step": 13010,
      "train_runtime": 25015.8039,
      "train_tokens_per_second": 8516.932
    },
    {
      "epoch": 0.8763129544842445,
      "grad_norm": 1.204389743873943,
      "learning_rate": 5.965854624636345e-06,
      "loss": 0.4185153007507324,
      "num_input_tokens_seen": 213138984,
      "step": 13015,
      "train_runtime": 25025.6139,
      "train_tokens_per_second": 8516.833
    },
    {
      "epoch": 0.8766496094802047,
      "grad_norm": 1.0717441301051351,
      "learning_rate": 5.963260208521598e-06,
      "loss": 0.39586901664733887,
      "num_input_tokens_seen": 213220904,
      "step": 13020,
      "train_runtime": 25035.2069,
      "train_tokens_per_second": 8516.842
    },
    {
      "epoch": 0.8769862644761648,
      "grad_norm": 1.1252917394580462,
      "learning_rate": 5.960665523034205e-06,
      "loss": 0.4190385818481445,
      "num_input_tokens_seen": 213302824,
      "step": 13025,
      "train_runtime": 25044.7224,
      "train_tokens_per_second": 8516.877
    },
    {
      "epoch": 0.8773229194721249,
      "grad_norm": 1.1322021368813449,
      "learning_rate": 5.958070568899762e-06,
      "loss": 0.3579248905181885,
      "num_input_tokens_seen": 213384744,
      "step": 13030,
      "train_runtime": 25054.345,
      "train_tokens_per_second": 8516.876
    },
    {
      "epoch": 0.8776595744680851,
      "grad_norm": 0.9959088584994243,
      "learning_rate": 5.955475346843939e-06,
      "loss": 0.35210480690002444,
      "num_input_tokens_seen": 213466664,
      "step": 13035,
      "train_runtime": 25063.9647,
      "train_tokens_per_second": 8516.875
    },
    {
      "epoch": 0.8779962294640452,
      "grad_norm": 1.106544907160144,
      "learning_rate": 5.952879857592482e-06,
      "loss": 0.3361725091934204,
      "num_input_tokens_seen": 213548584,
      "step": 13040,
      "train_runtime": 25073.4771,
      "train_tokens_per_second": 8516.911
    },
    {
      "epoch": 0.8783328844600053,
      "grad_norm": 1.149784602059803,
      "learning_rate": 5.950284101871212e-06,
      "loss": 0.43806753158569334,
      "num_input_tokens_seen": 213630504,
      "step": 13045,
      "train_runtime": 25083.1175,
      "train_tokens_per_second": 8516.904
    },
    {
      "epoch": 0.8786695394559655,
      "grad_norm": 1.0479634284705353,
      "learning_rate": 5.947688080406023e-06,
      "loss": 0.3688490629196167,
      "num_input_tokens_seen": 213712424,
      "step": 13050,
      "train_runtime": 25092.6775,
      "train_tokens_per_second": 8516.924
    },
    {
      "epoch": 0.8790061944519256,
      "grad_norm": 1.1468571540771395,
      "learning_rate": 5.945091793922885e-06,
      "loss": 0.33827009201049807,
      "num_input_tokens_seen": 213794344,
      "step": 13055,
      "train_runtime": 25102.1878,
      "train_tokens_per_second": 8516.961
    },
    {
      "epoch": 0.8793428494478858,
      "grad_norm": 1.0864011251634422,
      "learning_rate": 5.942495243147839e-06,
      "loss": 0.3847355365753174,
      "num_input_tokens_seen": 213875920,
      "step": 13060,
      "train_runtime": 25112.212,
      "train_tokens_per_second": 8516.809
    },
    {
      "epoch": 0.8796795044438459,
      "grad_norm": 1.0497126710050548,
      "learning_rate": 5.939898428807004e-06,
      "loss": 0.36087913513183595,
      "num_input_tokens_seen": 213957840,
      "step": 13065,
      "train_runtime": 25121.8235,
      "train_tokens_per_second": 8516.812
    },
    {
      "epoch": 0.880016159439806,
      "grad_norm": 1.173259589268004,
      "learning_rate": 5.937301351626571e-06,
      "loss": 0.3780477523803711,
      "num_input_tokens_seen": 214039760,
      "step": 13070,
      "train_runtime": 25131.3807,
      "train_tokens_per_second": 8516.833
    },
    {
      "epoch": 0.8803528144357662,
      "grad_norm": 1.1507948134267674,
      "learning_rate": 5.934704012332803e-06,
      "loss": 0.3914018630981445,
      "num_input_tokens_seen": 214121680,
      "step": 13075,
      "train_runtime": 25140.943,
      "train_tokens_per_second": 8516.852
    },
    {
      "epoch": 0.8806894694317263,
      "grad_norm": 1.201721380550295,
      "learning_rate": 5.93210641165204e-06,
      "loss": 0.4252922058105469,
      "num_input_tokens_seen": 214203600,
      "step": 13080,
      "train_runtime": 25150.563,
      "train_tokens_per_second": 8516.851
    },
    {
      "epoch": 0.8810261244276865,
      "grad_norm": 1.13409369778441,
      "learning_rate": 5.92950855031069e-06,
      "loss": 0.4121957778930664,
      "num_input_tokens_seen": 214285520,
      "step": 13085,
      "train_runtime": 25160.1286,
      "train_tokens_per_second": 8516.869
    },
    {
      "epoch": 0.8813627794236466,
      "grad_norm": 1.0651323347467618,
      "learning_rate": 5.926910429035239e-06,
      "loss": 0.387033486366272,
      "num_input_tokens_seen": 214367440,
      "step": 13090,
      "train_runtime": 25169.7184,
      "train_tokens_per_second": 8516.879
    },
    {
      "epoch": 0.8816994344196067,
      "grad_norm": 1.065392036278659,
      "learning_rate": 5.924312048552241e-06,
      "loss": 0.37127485275268557,
      "num_input_tokens_seen": 214449360,
      "step": 13095,
      "train_runtime": 25179.3403,
      "train_tokens_per_second": 8516.878
    },
    {
      "epoch": 0.8820360894155669,
      "grad_norm": 0.9392504984074519,
      "learning_rate": 5.9217134095883265e-06,
      "loss": 0.3579894542694092,
      "num_input_tokens_seen": 214531280,
      "step": 13100,
      "train_runtime": 25188.9821,
      "train_tokens_per_second": 8516.87
    },
    {
      "epoch": 0.882372744411527,
      "grad_norm": 1.1504446997095106,
      "learning_rate": 5.919114512870197e-06,
      "loss": 0.39468092918395997,
      "num_input_tokens_seen": 214613200,
      "step": 13105,
      "train_runtime": 25198.4883,
      "train_tokens_per_second": 8516.908
    },
    {
      "epoch": 0.8827093994074872,
      "grad_norm": 1.0347504247280896,
      "learning_rate": 5.916515359124623e-06,
      "loss": 0.4036856174468994,
      "num_input_tokens_seen": 214695120,
      "step": 13110,
      "train_runtime": 25208.0485,
      "train_tokens_per_second": 8516.927
    },
    {
      "epoch": 0.8830460544034473,
      "grad_norm": 1.1333222401994638,
      "learning_rate": 5.913915949078453e-06,
      "loss": 0.3907550573348999,
      "num_input_tokens_seen": 214777040,
      "step": 13115,
      "train_runtime": 25217.6894,
      "train_tokens_per_second": 8516.92
    },
    {
      "epoch": 0.8833827093994074,
      "grad_norm": 1.0653072208685688,
      "learning_rate": 5.911316283458601e-06,
      "loss": 0.3955994129180908,
      "num_input_tokens_seen": 214858960,
      "step": 13120,
      "train_runtime": 25227.2087,
      "train_tokens_per_second": 8516.953
    },
    {
      "epoch": 0.8837193643953676,
      "grad_norm": 0.9887445531188537,
      "learning_rate": 5.908716362992057e-06,
      "loss": 0.39969863891601565,
      "num_input_tokens_seen": 214940880,
      "step": 13125,
      "train_runtime": 25236.8357,
      "train_tokens_per_second": 8516.95
    },
    {
      "epoch": 0.8840560193913277,
      "grad_norm": 1.1458670687169092,
      "learning_rate": 5.90611618840588e-06,
      "loss": 0.4121832847595215,
      "num_input_tokens_seen": 215022800,
      "step": 13130,
      "train_runtime": 25246.4635,
      "train_tokens_per_second": 8516.947
    },
    {
      "epoch": 0.8843926743872879,
      "grad_norm": 1.1355974068643566,
      "learning_rate": 5.903515760427198e-06,
      "loss": 0.4260763645172119,
      "num_input_tokens_seen": 215104720,
      "step": 13135,
      "train_runtime": 25255.9794,
      "train_tokens_per_second": 8516.982
    },
    {
      "epoch": 0.884729329383248,
      "grad_norm": 1.238434862587458,
      "learning_rate": 5.900915079783218e-06,
      "loss": 0.423964786529541,
      "num_input_tokens_seen": 215186640,
      "step": 13140,
      "train_runtime": 25265.5605,
      "train_tokens_per_second": 8516.994
    },
    {
      "epoch": 0.8850659843792081,
      "grad_norm": 1.0079517596823742,
      "learning_rate": 5.898314147201207e-06,
      "loss": 0.3572171688079834,
      "num_input_tokens_seen": 215268560,
      "step": 13145,
      "train_runtime": 25275.1799,
      "train_tokens_per_second": 8516.994
    },
    {
      "epoch": 0.8854026393751683,
      "grad_norm": 1.2049485803056033,
      "learning_rate": 5.895712963408511e-06,
      "loss": 0.40389337539672854,
      "num_input_tokens_seen": 215350480,
      "step": 13150,
      "train_runtime": 25284.7286,
      "train_tokens_per_second": 8517.018
    },
    {
      "epoch": 0.8857392943711284,
      "grad_norm": 1.1441833941277397,
      "learning_rate": 5.893111529132539e-06,
      "loss": 0.4169490337371826,
      "num_input_tokens_seen": 215432400,
      "step": 13155,
      "train_runtime": 25294.2809,
      "train_tokens_per_second": 8517.04
    },
    {
      "epoch": 0.8860759493670886,
      "grad_norm": 1.0291030685817386,
      "learning_rate": 5.8905098451007755e-06,
      "loss": 0.40878639221191404,
      "num_input_tokens_seen": 215514320,
      "step": 13160,
      "train_runtime": 25303.9061,
      "train_tokens_per_second": 8517.038
    },
    {
      "epoch": 0.8864126043630487,
      "grad_norm": 1.1419610951197028,
      "learning_rate": 5.887907912040774e-06,
      "loss": 0.40551371574401857,
      "num_input_tokens_seen": 215596240,
      "step": 13165,
      "train_runtime": 25313.4671,
      "train_tokens_per_second": 8517.057
    },
    {
      "epoch": 0.8867492593590088,
      "grad_norm": 1.1983053000768888,
      "learning_rate": 5.885305730680158e-06,
      "loss": 0.3918898582458496,
      "num_input_tokens_seen": 215678160,
      "step": 13170,
      "train_runtime": 25323.0237,
      "train_tokens_per_second": 8517.078
    },
    {
      "epoch": 0.887085914354969,
      "grad_norm": 1.030152083768244,
      "learning_rate": 5.882703301746616e-06,
      "loss": 0.40273585319519045,
      "num_input_tokens_seen": 215760080,
      "step": 13175,
      "train_runtime": 25332.631,
      "train_tokens_per_second": 8517.081
    },
    {
      "epoch": 0.8874225693509291,
      "grad_norm": 0.9557838185977388,
      "learning_rate": 5.88010062596791e-06,
      "loss": 0.41527252197265624,
      "num_input_tokens_seen": 215842000,
      "step": 13180,
      "train_runtime": 25342.2394,
      "train_tokens_per_second": 8517.085
    },
    {
      "epoch": 0.8877592243468893,
      "grad_norm": 0.809614824175685,
      "learning_rate": 5.877497704071871e-06,
      "loss": 0.40056381225585935,
      "num_input_tokens_seen": 215923920,
      "step": 13185,
      "train_runtime": 25351.7441,
      "train_tokens_per_second": 8517.123
    },
    {
      "epoch": 0.8880958793428494,
      "grad_norm": 1.085215282735591,
      "learning_rate": 5.874894536786396e-06,
      "loss": 0.37878296375274656,
      "num_input_tokens_seen": 216005840,
      "step": 13190,
      "train_runtime": 25361.2856,
      "train_tokens_per_second": 8517.149
    },
    {
      "epoch": 0.8884325343388096,
      "grad_norm": 1.0138475012560217,
      "learning_rate": 5.8722911248394555e-06,
      "loss": 0.37706592082977297,
      "num_input_tokens_seen": 216087760,
      "step": 13195,
      "train_runtime": 25370.8891,
      "train_tokens_per_second": 8517.154
    },
    {
      "epoch": 0.8887691893347698,
      "grad_norm": 1.2714534182124566,
      "learning_rate": 5.8696874689590824e-06,
      "loss": 0.3964494228363037,
      "num_input_tokens_seen": 216169128,
      "step": 13200,
      "train_runtime": 25380.6851,
      "train_tokens_per_second": 8517.072
    },
    {
      "epoch": 0.8891058443307299,
      "grad_norm": 1.1254085394984394,
      "learning_rate": 5.867083569873383e-06,
      "loss": 0.41181058883666993,
      "num_input_tokens_seen": 216251048,
      "step": 13205,
      "train_runtime": 25390.29,
      "train_tokens_per_second": 8517.077
    },
    {
      "epoch": 0.8894424993266901,
      "grad_norm": 1.1193654885876345,
      "learning_rate": 5.8644794283105255e-06,
      "loss": 0.3789928436279297,
      "num_input_tokens_seen": 216332968,
      "step": 13210,
      "train_runtime": 25399.9107,
      "train_tokens_per_second": 8517.076
    },
    {
      "epoch": 0.8897791543226502,
      "grad_norm": 1.006977279555679,
      "learning_rate": 5.861875044998755e-06,
      "loss": 0.42157158851623533,
      "num_input_tokens_seen": 216414888,
      "step": 13215,
      "train_runtime": 25409.4152,
      "train_tokens_per_second": 8517.114
    },
    {
      "epoch": 0.8901158093186103,
      "grad_norm": 1.0946273550662278,
      "learning_rate": 5.859270420666377e-06,
      "loss": 0.3826438426971436,
      "num_input_tokens_seen": 216496808,
      "step": 13220,
      "train_runtime": 25419.0416,
      "train_tokens_per_second": 8517.111
    },
    {
      "epoch": 0.8904524643145705,
      "grad_norm": 1.19277960180467,
      "learning_rate": 5.856665556041764e-06,
      "loss": 0.3548551082611084,
      "num_input_tokens_seen": 216578728,
      "step": 13225,
      "train_runtime": 25428.6477,
      "train_tokens_per_second": 8517.115
    },
    {
      "epoch": 0.8907891193105306,
      "grad_norm": 1.0740753015773656,
      "learning_rate": 5.8540604518533605e-06,
      "loss": 0.41751847267150877,
      "num_input_tokens_seen": 216660648,
      "step": 13230,
      "train_runtime": 25438.2013,
      "train_tokens_per_second": 8517.137
    },
    {
      "epoch": 0.8911257743064908,
      "grad_norm": 1.3130506900961512,
      "learning_rate": 5.851455108829675e-06,
      "loss": 0.3883676052093506,
      "num_input_tokens_seen": 216742568,
      "step": 13235,
      "train_runtime": 25447.7668,
      "train_tokens_per_second": 8517.155
    },
    {
      "epoch": 0.8914624293024509,
      "grad_norm": 1.090685227643034,
      "learning_rate": 5.848849527699283e-06,
      "loss": 0.4008179187774658,
      "num_input_tokens_seen": 216824488,
      "step": 13240,
      "train_runtime": 25457.3339,
      "train_tokens_per_second": 8517.172
    },
    {
      "epoch": 0.891799084298411,
      "grad_norm": 1.1001790454689377,
      "learning_rate": 5.846243709190827e-06,
      "loss": 0.3757540464401245,
      "num_input_tokens_seen": 216906408,
      "step": 13245,
      "train_runtime": 25466.9003,
      "train_tokens_per_second": 8517.189
    },
    {
      "epoch": 0.8921357392943712,
      "grad_norm": 1.0156741824693976,
      "learning_rate": 5.843637654033019e-06,
      "loss": 0.3955357551574707,
      "num_input_tokens_seen": 216988328,
      "step": 13250,
      "train_runtime": 25476.4698,
      "train_tokens_per_second": 8517.205
    },
    {
      "epoch": 0.8924723942903313,
      "grad_norm": 0.9589003649621386,
      "learning_rate": 5.841031362954629e-06,
      "loss": 0.4298526763916016,
      "num_input_tokens_seen": 217070248,
      "step": 13255,
      "train_runtime": 25486.0843,
      "train_tokens_per_second": 8517.207
    },
    {
      "epoch": 0.8928090492862915,
      "grad_norm": 1.307173882621213,
      "learning_rate": 5.8384248366845e-06,
      "loss": 0.39066765308380125,
      "num_input_tokens_seen": 217152168,
      "step": 13260,
      "train_runtime": 25495.653,
      "train_tokens_per_second": 8517.223
    },
    {
      "epoch": 0.8931457042822516,
      "grad_norm": 1.1935525903200412,
      "learning_rate": 5.8358180759515396e-06,
      "loss": 0.3975389003753662,
      "num_input_tokens_seen": 217234088,
      "step": 13265,
      "train_runtime": 25505.2341,
      "train_tokens_per_second": 8517.236
    },
    {
      "epoch": 0.8934823592782117,
      "grad_norm": 0.9305269362279894,
      "learning_rate": 5.833211081484719e-06,
      "loss": 0.36778080463409424,
      "num_input_tokens_seen": 217316008,
      "step": 13270,
      "train_runtime": 25514.8547,
      "train_tokens_per_second": 8517.235
    },
    {
      "epoch": 0.8938190142741719,
      "grad_norm": 0.9537997740022591,
      "learning_rate": 5.830603854013077e-06,
      "loss": 0.3790589809417725,
      "num_input_tokens_seen": 217397928,
      "step": 13275,
      "train_runtime": 25524.3902,
      "train_tokens_per_second": 8517.262
    },
    {
      "epoch": 0.894155669270132,
      "grad_norm": 1.01019903823514,
      "learning_rate": 5.827996394265716e-06,
      "loss": 0.36084914207458496,
      "num_input_tokens_seen": 217479848,
      "step": 13280,
      "train_runtime": 25533.9051,
      "train_tokens_per_second": 8517.297
    },
    {
      "epoch": 0.8944923242660922,
      "grad_norm": 1.0429592922033057,
      "learning_rate": 5.825388702971802e-06,
      "loss": 0.3859630346298218,
      "num_input_tokens_seen": 217561768,
      "step": 13285,
      "train_runtime": 25543.5185,
      "train_tokens_per_second": 8517.298
    },
    {
      "epoch": 0.8948289792620523,
      "grad_norm": 1.1798509530007661,
      "learning_rate": 5.822780780860571e-06,
      "loss": 0.40363273620605467,
      "num_input_tokens_seen": 217643688,
      "step": 13290,
      "train_runtime": 25553.111,
      "train_tokens_per_second": 8517.307
    },
    {
      "epoch": 0.8951656342580124,
      "grad_norm": 0.9764097193825932,
      "learning_rate": 5.820172628661317e-06,
      "loss": 0.4012331008911133,
      "num_input_tokens_seen": 217725608,
      "step": 13295,
      "train_runtime": 25562.6288,
      "train_tokens_per_second": 8517.34
    },
    {
      "epoch": 0.8955022892539726,
      "grad_norm": 0.9604999162627207,
      "learning_rate": 5.8175642471034045e-06,
      "loss": 0.3614973068237305,
      "num_input_tokens_seen": 217807528,
      "step": 13300,
      "train_runtime": 25572.2065,
      "train_tokens_per_second": 8517.354
    },
    {
      "epoch": 0.8958389442499327,
      "grad_norm": 1.085805032928612,
      "learning_rate": 5.8149556369162565e-06,
      "loss": 0.3471692562103271,
      "num_input_tokens_seen": 217888648,
      "step": 13305,
      "train_runtime": 25582.0077,
      "train_tokens_per_second": 8517.261
    },
    {
      "epoch": 0.8961755992458929,
      "grad_norm": 1.1175629075309486,
      "learning_rate": 5.812346798829361e-06,
      "loss": 0.43292818069458006,
      "num_input_tokens_seen": 217970568,
      "step": 13310,
      "train_runtime": 25591.5105,
      "train_tokens_per_second": 8517.3
    },
    {
      "epoch": 0.896512254241853,
      "grad_norm": 0.8591033380482113,
      "learning_rate": 5.809737733572276e-06,
      "loss": 0.36238095760345457,
      "num_input_tokens_seen": 218052488,
      "step": 13315,
      "train_runtime": 25601.113,
      "train_tokens_per_second": 8517.305
    },
    {
      "epoch": 0.8968489092378131,
      "grad_norm": 1.1841422990715753,
      "learning_rate": 5.8071284418746166e-06,
      "loss": 0.38277482986450195,
      "num_input_tokens_seen": 218134408,
      "step": 13320,
      "train_runtime": 25610.7276,
      "train_tokens_per_second": 8517.306
    },
    {
      "epoch": 0.8971855642337733,
      "grad_norm": 0.9013341151236144,
      "learning_rate": 5.80451892446606e-06,
      "loss": 0.3993748903274536,
      "num_input_tokens_seen": 218216328,
      "step": 13325,
      "train_runtime": 25620.2358,
      "train_tokens_per_second": 8517.343
    },
    {
      "epoch": 0.8975222192297334,
      "grad_norm": 1.069034186124228,
      "learning_rate": 5.801909182076354e-06,
      "loss": 0.37286944389343263,
      "num_input_tokens_seen": 218298248,
      "step": 13330,
      "train_runtime": 25629.8598,
      "train_tokens_per_second": 8517.341
    },
    {
      "epoch": 0.8978588742256935,
      "grad_norm": 0.9057764324767907,
      "learning_rate": 5.7992992154353e-06,
      "loss": 0.40991816520690916,
      "num_input_tokens_seen": 218380168,
      "step": 13335,
      "train_runtime": 25639.4654,
      "train_tokens_per_second": 8517.345
    },
    {
      "epoch": 0.8981955292216537,
      "grad_norm": 1.242908734854803,
      "learning_rate": 5.79668902527277e-06,
      "loss": 0.37459659576416016,
      "num_input_tokens_seen": 218462088,
      "step": 13340,
      "train_runtime": 25649.0155,
      "train_tokens_per_second": 8517.367
    },
    {
      "epoch": 0.8985321842176138,
      "grad_norm": 1.0300164477181037,
      "learning_rate": 5.794078612318695e-06,
      "loss": 0.3669215202331543,
      "num_input_tokens_seen": 218544008,
      "step": 13345,
      "train_runtime": 25658.589,
      "train_tokens_per_second": 8517.382
    },
    {
      "epoch": 0.898868839213574,
      "grad_norm": 1.069081382112595,
      "learning_rate": 5.791467977303068e-06,
      "loss": 0.4005726337432861,
      "num_input_tokens_seen": 218625736,
      "step": 13350,
      "train_runtime": 25668.6276,
      "train_tokens_per_second": 8517.235
    },
    {
      "epoch": 0.8992054942095341,
      "grad_norm": 1.2783459795537697,
      "learning_rate": 5.788857120955945e-06,
      "loss": 0.4104879379272461,
      "num_input_tokens_seen": 218707656,
      "step": 13355,
      "train_runtime": 25678.1951,
      "train_tokens_per_second": 8517.252
    },
    {
      "epoch": 0.8995421492054942,
      "grad_norm": 0.9821893877709679,
      "learning_rate": 5.786246044007443e-06,
      "loss": 0.36040165424346926,
      "num_input_tokens_seen": 218789576,
      "step": 13360,
      "train_runtime": 25687.7568,
      "train_tokens_per_second": 8517.271
    },
    {
      "epoch": 0.8998788042014544,
      "grad_norm": 1.0648027665159028,
      "learning_rate": 5.783634747187743e-06,
      "loss": 0.3941212177276611,
      "num_input_tokens_seen": 218871496,
      "step": 13365,
      "train_runtime": 25697.3726,
      "train_tokens_per_second": 8517.271
    },
    {
      "epoch": 0.9002154591974145,
      "grad_norm": 1.3702816634670723,
      "learning_rate": 5.781023231227084e-06,
      "loss": 0.34082622528076173,
      "num_input_tokens_seen": 218953416,
      "step": 13370,
      "train_runtime": 25707.0005,
      "train_tokens_per_second": 8517.268
    },
    {
      "epoch": 0.9005521141933747,
      "grad_norm": 1.1515911587382406,
      "learning_rate": 5.778411496855769e-06,
      "loss": 0.39595139026641846,
      "num_input_tokens_seen": 219035336,
      "step": 13375,
      "train_runtime": 25716.5149,
      "train_tokens_per_second": 8517.302
    },
    {
      "epoch": 0.9008887691893348,
      "grad_norm": 1.200752997668739,
      "learning_rate": 5.775799544804161e-06,
      "loss": 0.4081613063812256,
      "num_input_tokens_seen": 219117256,
      "step": 13380,
      "train_runtime": 25726.1789,
      "train_tokens_per_second": 8517.287
    },
    {
      "epoch": 0.9012254241852949,
      "grad_norm": 1.076502861430893,
      "learning_rate": 5.7731873758026845e-06,
      "loss": 0.36323587894439696,
      "num_input_tokens_seen": 219199176,
      "step": 13385,
      "train_runtime": 25735.7923,
      "train_tokens_per_second": 8517.289
    },
    {
      "epoch": 0.9015620791812551,
      "grad_norm": 1.157376796409587,
      "learning_rate": 5.770574990581823e-06,
      "loss": 0.3689774990081787,
      "num_input_tokens_seen": 219280440,
      "step": 13390,
      "train_runtime": 25745.5342,
      "train_tokens_per_second": 8517.222
    },
    {
      "epoch": 0.9018987341772152,
      "grad_norm": 1.1606065528066092,
      "learning_rate": 5.767962389872124e-06,
      "loss": 0.351528263092041,
      "num_input_tokens_seen": 219362360,
      "step": 13395,
      "train_runtime": 25755.1561,
      "train_tokens_per_second": 8517.221
    },
    {
      "epoch": 0.9022353891731754,
      "grad_norm": 1.1119132403304697,
      "learning_rate": 5.765349574404191e-06,
      "loss": 0.3788151264190674,
      "num_input_tokens_seen": 219444280,
      "step": 13400,
      "train_runtime": 25764.723,
      "train_tokens_per_second": 8517.238
    },
    {
      "epoch": 0.9025720441691355,
      "grad_norm": 1.0293754951740146,
      "learning_rate": 5.7627365449086905e-06,
      "loss": 0.41572017669677735,
      "num_input_tokens_seen": 219526200,
      "step": 13405,
      "train_runtime": 25774.2339,
      "train_tokens_per_second": 8517.274
    },
    {
      "epoch": 0.9029086991650956,
      "grad_norm": 1.011645803718891,
      "learning_rate": 5.760123302116345e-06,
      "loss": 0.4466106414794922,
      "num_input_tokens_seen": 219608120,
      "step": 13410,
      "train_runtime": 25783.848,
      "train_tokens_per_second": 8517.275
    },
    {
      "epoch": 0.9032453541610558,
      "grad_norm": 1.2526061821881167,
      "learning_rate": 5.757509846757945e-06,
      "loss": 0.4064896583557129,
      "num_input_tokens_seen": 219689536,
      "step": 13415,
      "train_runtime": 25793.4572,
      "train_tokens_per_second": 8517.258
    },
    {
      "epoch": 0.9035820091570159,
      "grad_norm": 1.2701896005060207,
      "learning_rate": 5.75489617956433e-06,
      "loss": 0.40890793800354003,
      "num_input_tokens_seen": 219771456,
      "step": 13420,
      "train_runtime": 25802.9637,
      "train_tokens_per_second": 8517.295
    },
    {
      "epoch": 0.9039186641529761,
      "grad_norm": 1.0380720968218131,
      "learning_rate": 5.752282301266406e-06,
      "loss": 0.41172060966491697,
      "num_input_tokens_seen": 219853376,
      "step": 13425,
      "train_runtime": 25812.5895,
      "train_tokens_per_second": 8517.293
    },
    {
      "epoch": 0.9042553191489362,
      "grad_norm": 0.9896982953998387,
      "learning_rate": 5.749668212595136e-06,
      "loss": 0.39811458587646487,
      "num_input_tokens_seen": 219935296,
      "step": 13430,
      "train_runtime": 25822.211,
      "train_tokens_per_second": 8517.291
    },
    {
      "epoch": 0.9045919741448963,
      "grad_norm": 1.055665396357445,
      "learning_rate": 5.747053914281539e-06,
      "loss": 0.3725238800048828,
      "num_input_tokens_seen": 220017216,
      "step": 13435,
      "train_runtime": 25831.7705,
      "train_tokens_per_second": 8517.311
    },
    {
      "epoch": 0.9049286291408565,
      "grad_norm": 1.2507228618832429,
      "learning_rate": 5.744439407056699e-06,
      "loss": 0.3659604787826538,
      "num_input_tokens_seen": 220099136,
      "step": 13440,
      "train_runtime": 25841.3327,
      "train_tokens_per_second": 8517.329
    },
    {
      "epoch": 0.9052652841368166,
      "grad_norm": 1.0100892402299313,
      "learning_rate": 5.741824691651752e-06,
      "loss": 0.34780867099761964,
      "num_input_tokens_seen": 220181056,
      "step": 13445,
      "train_runtime": 25850.9429,
      "train_tokens_per_second": 8517.332
    },
    {
      "epoch": 0.9056019391327768,
      "grad_norm": 1.2458099444462045,
      "learning_rate": 5.739209768797896e-06,
      "loss": 0.4094665050506592,
      "num_input_tokens_seen": 220262656,
      "step": 13450,
      "train_runtime": 25861.602,
      "train_tokens_per_second": 8516.976
    },
    {
      "epoch": 0.9059385941287369,
      "grad_norm": 1.0973085845244253,
      "learning_rate": 5.736594639226385e-06,
      "loss": 0.36223413944244387,
      "num_input_tokens_seen": 220343816,
      "step": 13455,
      "train_runtime": 25871.3232,
      "train_tokens_per_second": 8516.913
    },
    {
      "epoch": 0.906275249124697,
      "grad_norm": 1.1899655551758022,
      "learning_rate": 5.7339793036685306e-06,
      "loss": 0.441160774230957,
      "num_input_tokens_seen": 220425736,
      "step": 13460,
      "train_runtime": 25880.9438,
      "train_tokens_per_second": 8516.913
    },
    {
      "epoch": 0.9066119041206572,
      "grad_norm": 1.0308681081062279,
      "learning_rate": 5.731363762855706e-06,
      "loss": 0.34939844608306886,
      "num_input_tokens_seen": 220507304,
      "step": 13465,
      "train_runtime": 25890.979,
      "train_tokens_per_second": 8516.762
    },
    {
      "epoch": 0.9069485591166173,
      "grad_norm": 1.0391298181614699,
      "learning_rate": 5.728748017519337e-06,
      "loss": 0.45427370071411133,
      "num_input_tokens_seen": 220588072,
      "step": 13470,
      "train_runtime": 25900.578,
      "train_tokens_per_second": 8516.724
    },
    {
      "epoch": 0.9072852141125775,
      "grad_norm": 0.9482689246093134,
      "learning_rate": 5.726132068390908e-06,
      "loss": 0.36444659233093263,
      "num_input_tokens_seen": 220669992,
      "step": 13475,
      "train_runtime": 25910.2354,
      "train_tokens_per_second": 8516.711
    },
    {
      "epoch": 0.9076218691085376,
      "grad_norm": 1.3113885019533114,
      "learning_rate": 5.723515916201962e-06,
      "loss": 0.43723430633544924,
      "num_input_tokens_seen": 220751912,
      "step": 13480,
      "train_runtime": 25919.8524,
      "train_tokens_per_second": 8516.712
    },
    {
      "epoch": 0.9079585241044977,
      "grad_norm": 0.8701587865320481,
      "learning_rate": 5.720899561684098e-06,
      "loss": 0.35997614860534666,
      "num_input_tokens_seen": 220833672,
      "step": 13485,
      "train_runtime": 25929.9253,
      "train_tokens_per_second": 8516.556
    },
    {
      "epoch": 0.9082951791004579,
      "grad_norm": 1.088509453210904,
      "learning_rate": 5.71828300556897e-06,
      "loss": 0.4157142639160156,
      "num_input_tokens_seen": 220915592,
      "step": 13490,
      "train_runtime": 25939.5082,
      "train_tokens_per_second": 8516.568
    },
    {
      "epoch": 0.908631834096418,
      "grad_norm": 0.9254853161397372,
      "learning_rate": 5.715666248588289e-06,
      "loss": 0.3815330982208252,
      "num_input_tokens_seen": 220997512,
      "step": 13495,
      "train_runtime": 25949.1509,
      "train_tokens_per_second": 8516.56
    },
    {
      "epoch": 0.9089684890923782,
      "grad_norm": 1.1009932260308872,
      "learning_rate": 5.713049291473825e-06,
      "loss": 0.43400139808654786,
      "num_input_tokens_seen": 221079432,
      "step": 13500,
      "train_runtime": 25958.7253,
      "train_tokens_per_second": 8516.575
    },
    {
      "epoch": 0.9093051440883383,
      "grad_norm": 1.0279424480462809,
      "learning_rate": 5.710432134957399e-06,
      "loss": 0.3411779165267944,
      "num_input_tokens_seen": 221160904,
      "step": 13505,
      "train_runtime": 25968.6318,
      "train_tokens_per_second": 8516.463
    },
    {
      "epoch": 0.9096417990842984,
      "grad_norm": 1.0510254394803393,
      "learning_rate": 5.707814779770892e-06,
      "loss": 0.349976634979248,
      "num_input_tokens_seen": 221242824,
      "step": 13510,
      "train_runtime": 25978.2624,
      "train_tokens_per_second": 8516.46
    },
    {
      "epoch": 0.9099784540802586,
      "grad_norm": 1.0932035635898176,
      "learning_rate": 5.7051972266462395e-06,
      "loss": 0.37895641326904295,
      "num_input_tokens_seen": 221324744,
      "step": 13515,
      "train_runtime": 25987.8876,
      "train_tokens_per_second": 8516.458
    },
    {
      "epoch": 0.9103151090762187,
      "grad_norm": 1.0753967128952648,
      "learning_rate": 5.702579476315431e-06,
      "loss": 0.3765723466873169,
      "num_input_tokens_seen": 221406664,
      "step": 13520,
      "train_runtime": 25997.3884,
      "train_tokens_per_second": 8516.496
    },
    {
      "epoch": 0.9106517640721788,
      "grad_norm": 1.0462949751063277,
      "learning_rate": 5.699961529510513e-06,
      "loss": 0.364529013633728,
      "num_input_tokens_seen": 221488584,
      "step": 13525,
      "train_runtime": 26006.9946,
      "train_tokens_per_second": 8516.501
    },
    {
      "epoch": 0.910988419068139,
      "grad_norm": 1.0765927128300101,
      "learning_rate": 5.697343386963586e-06,
      "loss": 0.3986340045928955,
      "num_input_tokens_seen": 221570504,
      "step": 13530,
      "train_runtime": 26016.6272,
      "train_tokens_per_second": 8516.496
    },
    {
      "epoch": 0.9113250740640991,
      "grad_norm": 0.9996274240019065,
      "learning_rate": 5.694725049406801e-06,
      "loss": 0.3741201877593994,
      "num_input_tokens_seen": 221652424,
      "step": 13535,
      "train_runtime": 26026.1295,
      "train_tokens_per_second": 8516.534
    },
    {
      "epoch": 0.9116617290600593,
      "grad_norm": 0.8939911877435521,
      "learning_rate": 5.692106517572373e-06,
      "loss": 0.35800273418426515,
      "num_input_tokens_seen": 221734344,
      "step": 13540,
      "train_runtime": 26035.7515,
      "train_tokens_per_second": 8516.533
    },
    {
      "epoch": 0.9119983840560194,
      "grad_norm": 1.227194137168269,
      "learning_rate": 5.689487792192566e-06,
      "loss": 0.40900325775146484,
      "num_input_tokens_seen": 221816264,
      "step": 13545,
      "train_runtime": 26045.3663,
      "train_tokens_per_second": 8516.535
    },
    {
      "epoch": 0.9123350390519795,
      "grad_norm": 1.1069450032477717,
      "learning_rate": 5.686868873999696e-06,
      "loss": 0.39687137603759765,
      "num_input_tokens_seen": 221898184,
      "step": 13550,
      "train_runtime": 26054.907,
      "train_tokens_per_second": 8516.56
    },
    {
      "epoch": 0.9126716940479397,
      "grad_norm": 1.0148166922772401,
      "learning_rate": 5.684249763726135e-06,
      "loss": 0.41794528961181643,
      "num_input_tokens_seen": 221980104,
      "step": 13555,
      "train_runtime": 26064.4752,
      "train_tokens_per_second": 8516.577
    },
    {
      "epoch": 0.9130083490438998,
      "grad_norm": 0.8814720143773621,
      "learning_rate": 5.681630462104308e-06,
      "loss": 0.3668773651123047,
      "num_input_tokens_seen": 222062024,
      "step": 13560,
      "train_runtime": 26074.1014,
      "train_tokens_per_second": 8516.574
    },
    {
      "epoch": 0.91334500403986,
      "grad_norm": 1.0208667216544416,
      "learning_rate": 5.679010969866697e-06,
      "loss": 0.40363645553588867,
      "num_input_tokens_seen": 222143944,
      "step": 13565,
      "train_runtime": 26083.6562,
      "train_tokens_per_second": 8516.595
    },
    {
      "epoch": 0.9136816590358201,
      "grad_norm": 1.3066633261857525,
      "learning_rate": 5.676391287745835e-06,
      "loss": 0.3697827339172363,
      "num_input_tokens_seen": 222225864,
      "step": 13570,
      "train_runtime": 26093.2259,
      "train_tokens_per_second": 8516.611
    },
    {
      "epoch": 0.9140183140317802,
      "grad_norm": 0.8926025744127621,
      "learning_rate": 5.673771416474305e-06,
      "loss": 0.36525013446807864,
      "num_input_tokens_seen": 222307784,
      "step": 13575,
      "train_runtime": 26102.8541,
      "train_tokens_per_second": 8516.608
    },
    {
      "epoch": 0.9143549690277404,
      "grad_norm": 1.1600597734115676,
      "learning_rate": 5.671151356784747e-06,
      "loss": 0.36576104164123535,
      "num_input_tokens_seen": 222389704,
      "step": 13580,
      "train_runtime": 26112.4885,
      "train_tokens_per_second": 8516.603
    },
    {
      "epoch": 0.9146916240237005,
      "grad_norm": 1.1879247077484647,
      "learning_rate": 5.668531109409851e-06,
      "loss": 0.4101104736328125,
      "num_input_tokens_seen": 222471624,
      "step": 13585,
      "train_runtime": 26122.0004,
      "train_tokens_per_second": 8516.638
    },
    {
      "epoch": 0.9150282790196607,
      "grad_norm": 0.944108861074926,
      "learning_rate": 5.665910675082362e-06,
      "loss": 0.3337234973907471,
      "num_input_tokens_seen": 222553544,
      "step": 13590,
      "train_runtime": 26131.6114,
      "train_tokens_per_second": 8516.641
    },
    {
      "epoch": 0.9153649340156208,
      "grad_norm": 0.9923415275641072,
      "learning_rate": 5.663290054535078e-06,
      "loss": 0.34328718185424806,
      "num_input_tokens_seen": 222635464,
      "step": 13595,
      "train_runtime": 26141.2397,
      "train_tokens_per_second": 8516.638
    },
    {
      "epoch": 0.9157015890115809,
      "grad_norm": 1.1690575828230054,
      "learning_rate": 5.660669248500846e-06,
      "loss": 0.44451122283935546,
      "num_input_tokens_seen": 222717384,
      "step": 13600,
      "train_runtime": 26150.7492,
      "train_tokens_per_second": 8516.673
    },
    {
      "epoch": 0.9160382440075411,
      "grad_norm": 1.088482836093137,
      "learning_rate": 5.658048257712563e-06,
      "loss": 0.35999507904052735,
      "num_input_tokens_seen": 222799304,
      "step": 13605,
      "train_runtime": 26160.3422,
      "train_tokens_per_second": 8516.682
    },
    {
      "epoch": 0.9163748990035012,
      "grad_norm": 0.9403031581114163,
      "learning_rate": 5.655427082903184e-06,
      "loss": 0.39266014099121094,
      "num_input_tokens_seen": 222881224,
      "step": 13610,
      "train_runtime": 26169.9703,
      "train_tokens_per_second": 8516.679
    },
    {
      "epoch": 0.9167115539994614,
      "grad_norm": 1.3407429061679124,
      "learning_rate": 5.652805724805711e-06,
      "loss": 0.3707202196121216,
      "num_input_tokens_seen": 222963144,
      "step": 13615,
      "train_runtime": 26179.4785,
      "train_tokens_per_second": 8516.714
    },
    {
      "epoch": 0.9170482089954215,
      "grad_norm": 1.15707304383087,
      "learning_rate": 5.650184184153199e-06,
      "loss": 0.40958614349365235,
      "num_input_tokens_seen": 223045064,
      "step": 13620,
      "train_runtime": 26189.1,
      "train_tokens_per_second": 8516.714
    },
    {
      "epoch": 0.9173848639913816,
      "grad_norm": 1.124504565375372,
      "learning_rate": 5.647562461678754e-06,
      "loss": 0.42523970603942873,
      "num_input_tokens_seen": 223126984,
      "step": 13625,
      "train_runtime": 26198.7159,
      "train_tokens_per_second": 8516.715
    },
    {
      "epoch": 0.9177215189873418,
      "grad_norm": 1.1503445822295972,
      "learning_rate": 5.64494055811553e-06,
      "loss": 0.3404831409454346,
      "num_input_tokens_seen": 223208904,
      "step": 13630,
      "train_runtime": 26208.2225,
      "train_tokens_per_second": 8516.751
    },
    {
      "epoch": 0.9180581739833019,
      "grad_norm": 1.1547571006065211,
      "learning_rate": 5.642318474196737e-06,
      "loss": 0.3564609050750732,
      "num_input_tokens_seen": 223290824,
      "step": 13635,
      "train_runtime": 26217.8576,
      "train_tokens_per_second": 8516.746
    },
    {
      "epoch": 0.9183948289792621,
      "grad_norm": 1.0356586717190501,
      "learning_rate": 5.63969621065563e-06,
      "loss": 0.36686675548553466,
      "num_input_tokens_seen": 223372744,
      "step": 13640,
      "train_runtime": 26227.4564,
      "train_tokens_per_second": 8516.752
    },
    {
      "epoch": 0.9187314839752222,
      "grad_norm": 1.0278677148321254,
      "learning_rate": 5.637073768225517e-06,
      "loss": 0.36862285137176515,
      "num_input_tokens_seen": 223454664,
      "step": 13645,
      "train_runtime": 26237.0043,
      "train_tokens_per_second": 8516.775
    },
    {
      "epoch": 0.9190681389711823,
      "grad_norm": 1.096754303512598,
      "learning_rate": 5.634451147639758e-06,
      "loss": 0.3575877666473389,
      "num_input_tokens_seen": 223536584,
      "step": 13650,
      "train_runtime": 26246.5647,
      "train_tokens_per_second": 8516.794
    },
    {
      "epoch": 0.9194047939671425,
      "grad_norm": 1.031970268689306,
      "learning_rate": 5.631828349631757e-06,
      "loss": 0.3892276525497437,
      "num_input_tokens_seen": 223618504,
      "step": 13655,
      "train_runtime": 26256.126,
      "train_tokens_per_second": 8516.813
    },
    {
      "epoch": 0.9197414489631026,
      "grad_norm": 1.0106500039678337,
      "learning_rate": 5.629205374934974e-06,
      "loss": 0.3825801134109497,
      "num_input_tokens_seen": 223700424,
      "step": 13660,
      "train_runtime": 26265.7067,
      "train_tokens_per_second": 8516.825
    },
    {
      "epoch": 0.9200781039590628,
      "grad_norm": 1.0701113788869956,
      "learning_rate": 5.626582224282913e-06,
      "loss": 0.3707162380218506,
      "num_input_tokens_seen": 223782344,
      "step": 13665,
      "train_runtime": 26275.2767,
      "train_tokens_per_second": 8516.841
    },
    {
      "epoch": 0.9204147589550229,
      "grad_norm": 0.9671474639526825,
      "learning_rate": 5.623958898409132e-06,
      "loss": 0.40628442764282224,
      "num_input_tokens_seen": 223864264,
      "step": 13670,
      "train_runtime": 26284.8903,
      "train_tokens_per_second": 8516.842
    },
    {
      "epoch": 0.920751413950983,
      "grad_norm": 0.9969321338486504,
      "learning_rate": 5.621335398047234e-06,
      "loss": 0.4171802043914795,
      "num_input_tokens_seen": 223946184,
      "step": 13675,
      "train_runtime": 26294.4492,
      "train_tokens_per_second": 8516.862
    },
    {
      "epoch": 0.9210880689469432,
      "grad_norm": 1.00553732868925,
      "learning_rate": 5.618711723930874e-06,
      "loss": 0.3713544368743896,
      "num_input_tokens_seen": 224027192,
      "step": 13680,
      "train_runtime": 26304.1804,
      "train_tokens_per_second": 8516.79
    },
    {
      "epoch": 0.9214247239429033,
      "grad_norm": 1.1097608361195124,
      "learning_rate": 5.616087876793753e-06,
      "loss": 0.39417462348937987,
      "num_input_tokens_seen": 224109112,
      "step": 13685,
      "train_runtime": 26313.7895,
      "train_tokens_per_second": 8516.794
    },
    {
      "epoch": 0.9217613789388635,
      "grad_norm": 1.0020508401360553,
      "learning_rate": 5.613463857369621e-06,
      "loss": 0.3967545747756958,
      "num_input_tokens_seen": 224191032,
      "step": 13690,
      "train_runtime": 26323.41,
      "train_tokens_per_second": 8516.793
    },
    {
      "epoch": 0.9220980339348236,
      "grad_norm": 0.9108581952482553,
      "learning_rate": 5.610839666392278e-06,
      "loss": 0.38825368881225586,
      "num_input_tokens_seen": 224272952,
      "step": 13695,
      "train_runtime": 26332.9229,
      "train_tokens_per_second": 8516.827
    },
    {
      "epoch": 0.9224346889307837,
      "grad_norm": 1.1386171059996026,
      "learning_rate": 5.608215304595571e-06,
      "loss": 0.38642661571502684,
      "num_input_tokens_seen": 224354872,
      "step": 13700,
      "train_runtime": 26342.5515,
      "train_tokens_per_second": 8516.824
    },
    {
      "epoch": 0.9227713439267439,
      "grad_norm": 0.920940064180783,
      "learning_rate": 5.605590772713393e-06,
      "loss": 0.3470944881439209,
      "num_input_tokens_seen": 224436792,
      "step": 13705,
      "train_runtime": 26352.1657,
      "train_tokens_per_second": 8516.825
    },
    {
      "epoch": 0.923107998922704,
      "grad_norm": 1.0421361236793638,
      "learning_rate": 5.602966071479687e-06,
      "loss": 0.39814512729644774,
      "num_input_tokens_seen": 224518712,
      "step": 13710,
      "train_runtime": 26361.669,
      "train_tokens_per_second": 8516.863
    },
    {
      "epoch": 0.9234446539186641,
      "grad_norm": 0.922861780496137,
      "learning_rate": 5.600341201628439e-06,
      "loss": 0.3873440265655518,
      "num_input_tokens_seen": 224600632,
      "step": 13715,
      "train_runtime": 26371.2741,
      "train_tokens_per_second": 8516.867
    },
    {
      "epoch": 0.9237813089146243,
      "grad_norm": 0.9337230260410979,
      "learning_rate": 5.597716163893692e-06,
      "loss": 0.3192357778549194,
      "num_input_tokens_seen": 224682552,
      "step": 13720,
      "train_runtime": 26380.8101,
      "train_tokens_per_second": 8516.894
    },
    {
      "epoch": 0.9241179639105844,
      "grad_norm": 1.1109921310452704,
      "learning_rate": 5.595090959009525e-06,
      "loss": 0.37081160545349123,
      "num_input_tokens_seen": 224764472,
      "step": 13725,
      "train_runtime": 26390.3174,
      "train_tokens_per_second": 8516.929
    },
    {
      "epoch": 0.9244546189065446,
      "grad_norm": 0.8969933817207177,
      "learning_rate": 5.592465587710068e-06,
      "loss": 0.32669639587402344,
      "num_input_tokens_seen": 224846392,
      "step": 13730,
      "train_runtime": 26399.9279,
      "train_tokens_per_second": 8516.932
    },
    {
      "epoch": 0.9247912739025047,
      "grad_norm": 1.1263534042689716,
      "learning_rate": 5.589840050729498e-06,
      "loss": 0.34850809574127195,
      "num_input_tokens_seen": 224928312,
      "step": 13735,
      "train_runtime": 26409.5287,
      "train_tokens_per_second": 8516.938
    },
    {
      "epoch": 0.9251279288984648,
      "grad_norm": 1.1699536716124816,
      "learning_rate": 5.587214348802039e-06,
      "loss": 0.4284060478210449,
      "num_input_tokens_seen": 225010232,
      "step": 13740,
      "train_runtime": 26419.0433,
      "train_tokens_per_second": 8516.971
    },
    {
      "epoch": 0.925464583894425,
      "grad_norm": 1.023169719883121,
      "learning_rate": 5.5845884826619615e-06,
      "loss": 0.36827936172485354,
      "num_input_tokens_seen": 225092152,
      "step": 13745,
      "train_runtime": 26428.663,
      "train_tokens_per_second": 8516.971
    },
    {
      "epoch": 0.9258012388903851,
      "grad_norm": 0.8803714087378165,
      "learning_rate": 5.5819624530435775e-06,
      "loss": 0.3848397731781006,
      "num_input_tokens_seen": 225174072,
      "step": 13750,
      "train_runtime": 26438.2914,
      "train_tokens_per_second": 8516.968
    },
    {
      "epoch": 0.9261378938863453,
      "grad_norm": 1.1165166894442782,
      "learning_rate": 5.57933626068125e-06,
      "loss": 0.42786407470703125,
      "num_input_tokens_seen": 225255992,
      "step": 13755,
      "train_runtime": 26447.7981,
      "train_tokens_per_second": 8517.004
    },
    {
      "epoch": 0.9264745488823054,
      "grad_norm": 1.081947254548166,
      "learning_rate": 5.5767099063093835e-06,
      "loss": 0.4044837474822998,
      "num_input_tokens_seen": 225337912,
      "step": 13760,
      "train_runtime": 26457.4272,
      "train_tokens_per_second": 8517.0
    },
    {
      "epoch": 0.9268112038782655,
      "grad_norm": 1.1834872902146478,
      "learning_rate": 5.574083390662431e-06,
      "loss": 0.418363094329834,
      "num_input_tokens_seen": 225419832,
      "step": 13765,
      "train_runtime": 26467.039,
      "train_tokens_per_second": 8517.002
    },
    {
      "epoch": 0.9271478588742257,
      "grad_norm": 1.1356692510536333,
      "learning_rate": 5.57145671447489e-06,
      "loss": 0.40317397117614745,
      "num_input_tokens_seen": 225501000,
      "step": 13770,
      "train_runtime": 26476.7971,
      "train_tokens_per_second": 8516.929
    },
    {
      "epoch": 0.9274845138701858,
      "grad_norm": 1.2564637024558913,
      "learning_rate": 5.5688298784813e-06,
      "loss": 0.40511884689331057,
      "num_input_tokens_seen": 225582920,
      "step": 13775,
      "train_runtime": 26486.3713,
      "train_tokens_per_second": 8516.943
    },
    {
      "epoch": 0.927821168866146,
      "grad_norm": 1.0978387396597515,
      "learning_rate": 5.566202883416249e-06,
      "loss": 0.4454645156860352,
      "num_input_tokens_seen": 225664840,
      "step": 13780,
      "train_runtime": 26496.0053,
      "train_tokens_per_second": 8516.938
    },
    {
      "epoch": 0.9281578238621061,
      "grad_norm": 1.1279263142257852,
      "learning_rate": 5.563575730014366e-06,
      "loss": 0.409511661529541,
      "num_input_tokens_seen": 225746760,
      "step": 13785,
      "train_runtime": 26505.5661,
      "train_tokens_per_second": 8516.957
    },
    {
      "epoch": 0.9284944788580662,
      "grad_norm": 1.1498478703076664,
      "learning_rate": 5.560948419010328e-06,
      "loss": 0.41030092239379884,
      "num_input_tokens_seen": 225828680,
      "step": 13790,
      "train_runtime": 26515.1451,
      "train_tokens_per_second": 8516.969
    },
    {
      "epoch": 0.9288311338540264,
      "grad_norm": 0.9436706169684567,
      "learning_rate": 5.558320951138853e-06,
      "loss": 0.38550262451171874,
      "num_input_tokens_seen": 225910600,
      "step": 13795,
      "train_runtime": 26524.7557,
      "train_tokens_per_second": 8516.972
    },
    {
      "epoch": 0.9291677888499865,
      "grad_norm": 0.9993331889300611,
      "learning_rate": 5.5556933271347046e-06,
      "loss": 0.3741105556488037,
      "num_input_tokens_seen": 225992520,
      "step": 13800,
      "train_runtime": 26534.3629,
      "train_tokens_per_second": 8516.976
    },
    {
      "epoch": 0.9295044438459467,
      "grad_norm": 1.0745636608449978,
      "learning_rate": 5.55306554773269e-06,
      "loss": 0.4287284851074219,
      "num_input_tokens_seen": 226074440,
      "step": 13805,
      "train_runtime": 26543.8668,
      "train_tokens_per_second": 8517.012
    },
    {
      "epoch": 0.9298410988419068,
      "grad_norm": 1.0139470827259973,
      "learning_rate": 5.550437613667658e-06,
      "loss": 0.38964333534240725,
      "num_input_tokens_seen": 226156360,
      "step": 13810,
      "train_runtime": 26553.5066,
      "train_tokens_per_second": 8517.005
    },
    {
      "epoch": 0.9301777538378669,
      "grad_norm": 1.0263007338584766,
      "learning_rate": 5.547809525674501e-06,
      "loss": 0.3971379041671753,
      "num_input_tokens_seen": 226238280,
      "step": 13815,
      "train_runtime": 26563.1225,
      "train_tokens_per_second": 8517.006
    },
    {
      "epoch": 0.9305144088338271,
      "grad_norm": 0.9642647156215189,
      "learning_rate": 5.545181284488159e-06,
      "loss": 0.4019437313079834,
      "num_input_tokens_seen": 226320200,
      "step": 13820,
      "train_runtime": 26572.6262,
      "train_tokens_per_second": 8517.043
    },
    {
      "epoch": 0.9308510638297872,
      "grad_norm": 1.1018213951697053,
      "learning_rate": 5.5425528908436085e-06,
      "loss": 0.38912985324859617,
      "num_input_tokens_seen": 226402120,
      "step": 13825,
      "train_runtime": 26582.2375,
      "train_tokens_per_second": 8517.045
    },
    {
      "epoch": 0.9311877188257474,
      "grad_norm": 1.0308581927543137,
      "learning_rate": 5.539924345475873e-06,
      "loss": 0.36619091033935547,
      "num_input_tokens_seen": 226484040,
      "step": 13830,
      "train_runtime": 26591.8552,
      "train_tokens_per_second": 8517.045
    },
    {
      "epoch": 0.9315243738217075,
      "grad_norm": 1.1364378972433298,
      "learning_rate": 5.5372956491200145e-06,
      "loss": 0.4014134407043457,
      "num_input_tokens_seen": 226565960,
      "step": 13835,
      "train_runtime": 26601.3605,
      "train_tokens_per_second": 8517.082
    },
    {
      "epoch": 0.9318610288176676,
      "grad_norm": 1.0071838083192726,
      "learning_rate": 5.534666802511143e-06,
      "loss": 0.3678145408630371,
      "num_input_tokens_seen": 226647880,
      "step": 13840,
      "train_runtime": 26610.9986,
      "train_tokens_per_second": 8517.075
    },
    {
      "epoch": 0.9321976838136278,
      "grad_norm": 1.2778122435556947,
      "learning_rate": 5.532037806384404e-06,
      "loss": 0.4043914318084717,
      "num_input_tokens_seen": 226729800,
      "step": 13845,
      "train_runtime": 26620.5571,
      "train_tokens_per_second": 8517.094
    },
    {
      "epoch": 0.9325343388095879,
      "grad_norm": 0.8778076085522655,
      "learning_rate": 5.529408661474989e-06,
      "loss": 0.3445703983306885,
      "num_input_tokens_seen": 226811720,
      "step": 13850,
      "train_runtime": 26630.1331,
      "train_tokens_per_second": 8517.108
    },
    {
      "epoch": 0.932870993805548,
      "grad_norm": 1.1090698031904593,
      "learning_rate": 5.526779368518129e-06,
      "loss": 0.39854631423950193,
      "num_input_tokens_seen": 226893640,
      "step": 13855,
      "train_runtime": 26639.6999,
      "train_tokens_per_second": 8517.124
    },
    {
      "epoch": 0.9332076488015082,
      "grad_norm": 1.0730481735975221,
      "learning_rate": 5.5241499282491e-06,
      "loss": 0.37203102111816405,
      "num_input_tokens_seen": 226975560,
      "step": 13860,
      "train_runtime": 26649.3044,
      "train_tokens_per_second": 8517.129
    },
    {
      "epoch": 0.9335443037974683,
      "grad_norm": 1.0536165587662556,
      "learning_rate": 5.5215203414032135e-06,
      "loss": 0.34066028594970704,
      "num_input_tokens_seen": 227057480,
      "step": 13865,
      "train_runtime": 26658.8592,
      "train_tokens_per_second": 8517.149
    },
    {
      "epoch": 0.9338809587934285,
      "grad_norm": 1.0994233972364824,
      "learning_rate": 5.518890608715828e-06,
      "loss": 0.3735603094100952,
      "num_input_tokens_seen": 227139400,
      "step": 13870,
      "train_runtime": 26668.4291,
      "train_tokens_per_second": 8517.165
    },
    {
      "epoch": 0.9342176137893886,
      "grad_norm": 1.2862123318487646,
      "learning_rate": 5.516260730922337e-06,
      "loss": 0.37006850242614747,
      "num_input_tokens_seen": 227221064,
      "step": 13875,
      "train_runtime": 26678.4335,
      "train_tokens_per_second": 8517.032
    },
    {
      "epoch": 0.9345542687853488,
      "grad_norm": 1.2790689857057629,
      "learning_rate": 5.51363070875818e-06,
      "loss": 0.4154044151306152,
      "num_input_tokens_seen": 227302984,
      "step": 13880,
      "train_runtime": 26688.0463,
      "train_tokens_per_second": 8517.033
    },
    {
      "epoch": 0.9348909237813089,
      "grad_norm": 0.987378400149711,
      "learning_rate": 5.511000542958832e-06,
      "loss": 0.41206817626953124,
      "num_input_tokens_seen": 227384904,
      "step": 13885,
      "train_runtime": 26697.5499,
      "train_tokens_per_second": 8517.07
    },
    {
      "epoch": 0.935227578777269,
      "grad_norm": 1.0611487687562342,
      "learning_rate": 5.508370234259812e-06,
      "loss": 0.362284255027771,
      "num_input_tokens_seen": 227466824,
      "step": 13890,
      "train_runtime": 26707.1582,
      "train_tokens_per_second": 8517.073
    },
    {
      "epoch": 0.9355642337732292,
      "grad_norm": 0.9014072404665111,
      "learning_rate": 5.505739783396678e-06,
      "loss": 0.3697743654251099,
      "num_input_tokens_seen": 227548744,
      "step": 13895,
      "train_runtime": 26716.7215,
      "train_tokens_per_second": 8517.091
    },
    {
      "epoch": 0.9359008887691893,
      "grad_norm": 1.0020433974817078,
      "learning_rate": 5.503109191105026e-06,
      "loss": 0.373736572265625,
      "num_input_tokens_seen": 227630664,
      "step": 13900,
      "train_runtime": 26726.2278,
      "train_tokens_per_second": 8517.127
    },
    {
      "epoch": 0.9362375437651494,
      "grad_norm": 1.0466900560218362,
      "learning_rate": 5.500478458120493e-06,
      "loss": 0.4124135494232178,
      "num_input_tokens_seen": 227712584,
      "step": 13905,
      "train_runtime": 26735.8411,
      "train_tokens_per_second": 8517.128
    },
    {
      "epoch": 0.9365741987611096,
      "grad_norm": 1.0420191463654818,
      "learning_rate": 5.497847585178754e-06,
      "loss": 0.4012413024902344,
      "num_input_tokens_seen": 227794472,
      "step": 13910,
      "train_runtime": 26745.419,
      "train_tokens_per_second": 8517.14
    },
    {
      "epoch": 0.9369108537570697,
      "grad_norm": 1.1446311056454694,
      "learning_rate": 5.495216573015526e-06,
      "loss": 0.37330875396728513,
      "num_input_tokens_seen": 227876392,
      "step": 13915,
      "train_runtime": 26754.9253,
      "train_tokens_per_second": 8517.175
    },
    {
      "epoch": 0.9372475087530299,
      "grad_norm": 1.2564811322788025,
      "learning_rate": 5.492585422366562e-06,
      "loss": 0.3951666831970215,
      "num_input_tokens_seen": 227958312,
      "step": 13920,
      "train_runtime": 26764.5262,
      "train_tokens_per_second": 8517.181
    },
    {
      "epoch": 0.93758416374899,
      "grad_norm": 1.009575075917734,
      "learning_rate": 5.489954133967657e-06,
      "loss": 0.3487447738647461,
      "num_input_tokens_seen": 228040096,
      "step": 13925,
      "train_runtime": 26774.6219,
      "train_tokens_per_second": 8517.024
    },
    {
      "epoch": 0.9379208187449501,
      "grad_norm": 1.120957153768455,
      "learning_rate": 5.48732270855464e-06,
      "loss": 0.39089312553405764,
      "num_input_tokens_seen": 228122016,
      "step": 13930,
      "train_runtime": 26784.1323,
      "train_tokens_per_second": 8517.058
    },
    {
      "epoch": 0.9382574737409103,
      "grad_norm": 1.070210779861485,
      "learning_rate": 5.48469114686338e-06,
      "loss": 0.3781121253967285,
      "num_input_tokens_seen": 228203936,
      "step": 13935,
      "train_runtime": 26793.703,
      "train_tokens_per_second": 8517.073
    },
    {
      "epoch": 0.9385941287368704,
      "grad_norm": 0.9616384502250399,
      "learning_rate": 5.482059449629788e-06,
      "loss": 0.36685502529144287,
      "num_input_tokens_seen": 228285856,
      "step": 13940,
      "train_runtime": 26803.3195,
      "train_tokens_per_second": 8517.074
    },
    {
      "epoch": 0.9389307837328306,
      "grad_norm": 1.1122311419887438,
      "learning_rate": 5.479427617589808e-06,
      "loss": 0.41621990203857423,
      "num_input_tokens_seen": 228367776,
      "step": 13945,
      "train_runtime": 26812.8673,
      "train_tokens_per_second": 8517.096
    },
    {
      "epoch": 0.9392674387287907,
      "grad_norm": 3.1537783988565424,
      "learning_rate": 5.4767956514794226e-06,
      "loss": 0.36086239814758303,
      "num_input_tokens_seen": 228449696,
      "step": 13950,
      "train_runtime": 26822.4307,
      "train_tokens_per_second": 8517.114
    },
    {
      "epoch": 0.9396040937247508,
      "grad_norm": 1.1395853791640627,
      "learning_rate": 5.474163552034655e-06,
      "loss": 0.3824437141418457,
      "num_input_tokens_seen": 228531616,
      "step": 13955,
      "train_runtime": 26832.063,
      "train_tokens_per_second": 8517.109
    },
    {
      "epoch": 0.939940748720711,
      "grad_norm": 0.8961455894996707,
      "learning_rate": 5.4715313199915595e-06,
      "loss": 0.3850611925125122,
      "num_input_tokens_seen": 228613536,
      "step": 13960,
      "train_runtime": 26841.6162,
      "train_tokens_per_second": 8517.13
    },
    {
      "epoch": 0.9402774037166711,
      "grad_norm": 1.1688568400870651,
      "learning_rate": 5.468898956086236e-06,
      "loss": 0.38628010749816893,
      "num_input_tokens_seen": 228695456,
      "step": 13965,
      "train_runtime": 26851.1882,
      "train_tokens_per_second": 8517.145
    },
    {
      "epoch": 0.9406140587126313,
      "grad_norm": 1.0988732809649142,
      "learning_rate": 5.466266461054815e-06,
      "loss": 0.3749761819839478,
      "num_input_tokens_seen": 228777376,
      "step": 13970,
      "train_runtime": 26860.7938,
      "train_tokens_per_second": 8517.149
    },
    {
      "epoch": 0.9409507137085914,
      "grad_norm": 1.029370725476351,
      "learning_rate": 5.463633835633464e-06,
      "loss": 0.408290958404541,
      "num_input_tokens_seen": 228859296,
      "step": 13975,
      "train_runtime": 26870.3657,
      "train_tokens_per_second": 8517.163
    },
    {
      "epoch": 0.9412873687045515,
      "grad_norm": 0.9975296304663407,
      "learning_rate": 5.46100108055839e-06,
      "loss": 0.4061539649963379,
      "num_input_tokens_seen": 228941216,
      "step": 13980,
      "train_runtime": 26879.9429,
      "train_tokens_per_second": 8517.176
    },
    {
      "epoch": 0.9416240237005117,
      "grad_norm": 0.9855459393319435,
      "learning_rate": 5.458368196565833e-06,
      "loss": 0.3168904781341553,
      "num_input_tokens_seen": 229023136,
      "step": 13985,
      "train_runtime": 26889.553,
      "train_tokens_per_second": 8517.179
    },
    {
      "epoch": 0.9419606786964718,
      "grad_norm": 1.1937262191920743,
      "learning_rate": 5.455735184392073e-06,
      "loss": 0.42143821716308594,
      "num_input_tokens_seen": 229105056,
      "step": 13990,
      "train_runtime": 26899.181,
      "train_tokens_per_second": 8517.176
    },
    {
      "epoch": 0.942297333692432,
      "grad_norm": 1.1372534440500173,
      "learning_rate": 5.4531020447734204e-06,
      "loss": 0.3975001096725464,
      "num_input_tokens_seen": 229186976,
      "step": 13995,
      "train_runtime": 26908.6938,
      "train_tokens_per_second": 8517.209
    },
    {
      "epoch": 0.9426339886883921,
      "grad_norm": 0.9075119101813912,
      "learning_rate": 5.450468778446226e-06,
      "loss": 0.3940162420272827,
      "num_input_tokens_seen": 229268896,
      "step": 14000,
      "train_runtime": 26918.3387,
      "train_tokens_per_second": 8517.201
    },
    {
      "epoch": 0.9429706436843522,
      "grad_norm": 1.149162910955268,
      "learning_rate": 5.447835386146875e-06,
      "loss": 0.39708359241485597,
      "num_input_tokens_seen": 229350816,
      "step": 14005,
      "train_runtime": 26927.9504,
      "train_tokens_per_second": 8517.203
    },
    {
      "epoch": 0.9433072986803124,
      "grad_norm": 1.0554839255398847,
      "learning_rate": 5.445201868611784e-06,
      "loss": 0.42700948715209963,
      "num_input_tokens_seen": 229432736,
      "step": 14010,
      "train_runtime": 26937.4606,
      "train_tokens_per_second": 8517.237
    },
    {
      "epoch": 0.9436439536762725,
      "grad_norm": 0.875144813160166,
      "learning_rate": 5.442568226577413e-06,
      "loss": 0.38336780071258547,
      "num_input_tokens_seen": 229514656,
      "step": 14015,
      "train_runtime": 26947.0804,
      "train_tokens_per_second": 8517.236
    },
    {
      "epoch": 0.9439806086722327,
      "grad_norm": 0.9826336472889399,
      "learning_rate": 5.439934460780247e-06,
      "loss": 0.3388866424560547,
      "num_input_tokens_seen": 229596576,
      "step": 14020,
      "train_runtime": 26956.6717,
      "train_tokens_per_second": 8517.245
    },
    {
      "epoch": 0.9443172636681928,
      "grad_norm": 0.8801534847769833,
      "learning_rate": 5.437300571956813e-06,
      "loss": 0.3640773773193359,
      "num_input_tokens_seen": 229678256,
      "step": 14025,
      "train_runtime": 26966.6473,
      "train_tokens_per_second": 8517.123
    },
    {
      "epoch": 0.9446539186641529,
      "grad_norm": 1.0044527819454245,
      "learning_rate": 5.434666560843667e-06,
      "loss": 0.3668461084365845,
      "num_input_tokens_seen": 229760176,
      "step": 14030,
      "train_runtime": 26976.2896,
      "train_tokens_per_second": 8517.116
    },
    {
      "epoch": 0.9449905736601131,
      "grad_norm": 1.0401258410318013,
      "learning_rate": 5.432032428177405e-06,
      "loss": 0.37430486679077146,
      "num_input_tokens_seen": 229842096,
      "step": 14035,
      "train_runtime": 26985.9137,
      "train_tokens_per_second": 8517.114
    },
    {
      "epoch": 0.9453272286560732,
      "grad_norm": 1.1783844687131306,
      "learning_rate": 5.429398174694651e-06,
      "loss": 0.35134210586547854,
      "num_input_tokens_seen": 229924016,
      "step": 14040,
      "train_runtime": 26995.468,
      "train_tokens_per_second": 8517.134
    },
    {
      "epoch": 0.9456638836520334,
      "grad_norm": 0.9155325338850514,
      "learning_rate": 5.426763801132067e-06,
      "loss": 0.3786123275756836,
      "num_input_tokens_seen": 230005128,
      "step": 14045,
      "train_runtime": 27005.2275,
      "train_tokens_per_second": 8517.059
    },
    {
      "epoch": 0.9460005386479935,
      "grad_norm": 1.1571949057634812,
      "learning_rate": 5.424129308226347e-06,
      "loss": 0.4304826259613037,
      "num_input_tokens_seen": 230087048,
      "step": 14050,
      "train_runtime": 27014.8452,
      "train_tokens_per_second": 8517.06
    },
    {
      "epoch": 0.9463371936439536,
      "grad_norm": 1.1473892725557235,
      "learning_rate": 5.421494696714219e-06,
      "loss": 0.45882453918457033,
      "num_input_tokens_seen": 230168968,
      "step": 14055,
      "train_runtime": 27024.3965,
      "train_tokens_per_second": 8517.081
    },
    {
      "epoch": 0.9466738486399138,
      "grad_norm": 1.1103178196986059,
      "learning_rate": 5.418859967332442e-06,
      "loss": 0.4218428611755371,
      "num_input_tokens_seen": 230250888,
      "step": 14060,
      "train_runtime": 27033.9064,
      "train_tokens_per_second": 8517.115
    },
    {
      "epoch": 0.9470105036358739,
      "grad_norm": 0.9938972558970175,
      "learning_rate": 5.416225120817811e-06,
      "loss": 0.3830850124359131,
      "num_input_tokens_seen": 230332808,
      "step": 14065,
      "train_runtime": 27043.5092,
      "train_tokens_per_second": 8517.12
    },
    {
      "epoch": 0.947347158631834,
      "grad_norm": 1.0430077191788674,
      "learning_rate": 5.413590157907153e-06,
      "loss": 0.36276907920837403,
      "num_input_tokens_seen": 230414728,
      "step": 14070,
      "train_runtime": 27053.0682,
      "train_tokens_per_second": 8517.138
    },
    {
      "epoch": 0.9476838136277942,
      "grad_norm": 0.9233634707022121,
      "learning_rate": 5.4109550793373255e-06,
      "loss": 0.3823229312896729,
      "num_input_tokens_seen": 230496648,
      "step": 14075,
      "train_runtime": 27062.6384,
      "train_tokens_per_second": 8517.154
    },
    {
      "epoch": 0.9480204686237543,
      "grad_norm": 1.310306126892947,
      "learning_rate": 5.40831988584522e-06,
      "loss": 0.4666844367980957,
      "num_input_tokens_seen": 230578568,
      "step": 14080,
      "train_runtime": 27072.2457,
      "train_tokens_per_second": 8517.157
    },
    {
      "epoch": 0.9483571236197145,
      "grad_norm": 1.143234127506132,
      "learning_rate": 5.40568457816776e-06,
      "loss": 0.4039937973022461,
      "num_input_tokens_seen": 230660488,
      "step": 14085,
      "train_runtime": 27081.8604,
      "train_tokens_per_second": 8517.158
    },
    {
      "epoch": 0.9486937786156746,
      "grad_norm": 1.1824861851847337,
      "learning_rate": 5.403049157041903e-06,
      "loss": 0.36440553665161135,
      "num_input_tokens_seen": 230742408,
      "step": 14090,
      "train_runtime": 27091.368,
      "train_tokens_per_second": 8517.193
    },
    {
      "epoch": 0.9490304336116347,
      "grad_norm": 1.1522580934142381,
      "learning_rate": 5.400413623204635e-06,
      "loss": 0.39713852405548095,
      "num_input_tokens_seen": 230824328,
      "step": 14095,
      "train_runtime": 27100.9929,
      "train_tokens_per_second": 8517.191
    },
    {
      "epoch": 0.9493670886075949,
      "grad_norm": 1.0293191167262399,
      "learning_rate": 5.397777977392973e-06,
      "loss": 0.37778582572937014,
      "num_input_tokens_seen": 230906248,
      "step": 14100,
      "train_runtime": 27110.5702,
      "train_tokens_per_second": 8517.204
    },
    {
      "epoch": 0.949703743603555,
      "grad_norm": 1.0931902748074898,
      "learning_rate": 5.395142220343967e-06,
      "loss": 0.41011343002319334,
      "num_input_tokens_seen": 230988168,
      "step": 14105,
      "train_runtime": 27120.0743,
      "train_tokens_per_second": 8517.24
    },
    {
      "epoch": 0.9500403985995152,
      "grad_norm": 0.8833690140750213,
      "learning_rate": 5.3925063527947e-06,
      "loss": 0.37215099334716795,
      "num_input_tokens_seen": 231070088,
      "step": 14110,
      "train_runtime": 27129.7018,
      "train_tokens_per_second": 8517.237
    },
    {
      "epoch": 0.9503770535954753,
      "grad_norm": 1.0399152478077114,
      "learning_rate": 5.389870375482283e-06,
      "loss": 0.38038837909698486,
      "num_input_tokens_seen": 231152008,
      "step": 14115,
      "train_runtime": 27139.3191,
      "train_tokens_per_second": 8517.237
    },
    {
      "epoch": 0.9507137085914354,
      "grad_norm": 1.0411527616147318,
      "learning_rate": 5.387234289143859e-06,
      "loss": 0.3381585836410522,
      "num_input_tokens_seen": 231233928,
      "step": 14120,
      "train_runtime": 27148.8295,
      "train_tokens_per_second": 8517.271
    },
    {
      "epoch": 0.9510503635873956,
      "grad_norm": 1.0828620479632793,
      "learning_rate": 5.384598094516601e-06,
      "loss": 0.3506008625030518,
      "num_input_tokens_seen": 231315848,
      "step": 14125,
      "train_runtime": 27158.465,
      "train_tokens_per_second": 8517.265
    },
    {
      "epoch": 0.9513870185833557,
      "grad_norm": 1.1225518735128026,
      "learning_rate": 5.381961792337712e-06,
      "loss": 0.407561731338501,
      "num_input_tokens_seen": 231397768,
      "step": 14130,
      "train_runtime": 27168.1042,
      "train_tokens_per_second": 8517.259
    },
    {
      "epoch": 0.9517236735793159,
      "grad_norm": 1.0477425679023333,
      "learning_rate": 5.3793253833444245e-06,
      "loss": 0.3701212406158447,
      "num_input_tokens_seen": 231479688,
      "step": 14135,
      "train_runtime": 27177.6567,
      "train_tokens_per_second": 8517.279
    },
    {
      "epoch": 0.952060328575276,
      "grad_norm": 0.9060114297923368,
      "learning_rate": 5.3766888682740045e-06,
      "loss": 0.37421298027038574,
      "num_input_tokens_seen": 231561608,
      "step": 14140,
      "train_runtime": 27187.1647,
      "train_tokens_per_second": 8517.314
    },
    {
      "epoch": 0.9523969835712361,
      "grad_norm": 1.1080688268476537,
      "learning_rate": 5.374052247863745e-06,
      "loss": 0.3997375965118408,
      "num_input_tokens_seen": 231643528,
      "step": 14145,
      "train_runtime": 27196.7389,
      "train_tokens_per_second": 8517.327
    },
    {
      "epoch": 0.9527336385671963,
      "grad_norm": 1.0315188522325203,
      "learning_rate": 5.371415522850966e-06,
      "loss": 0.37203350067138674,
      "num_input_tokens_seen": 231725448,
      "step": 14150,
      "train_runtime": 27206.288,
      "train_tokens_per_second": 8517.349
    },
    {
      "epoch": 0.9530702935631564,
      "grad_norm": 0.9631198288076108,
      "learning_rate": 5.368778693973019e-06,
      "loss": 0.3513765811920166,
      "num_input_tokens_seen": 231807368,
      "step": 14155,
      "train_runtime": 27215.8585,
      "train_tokens_per_second": 8517.364
    },
    {
      "epoch": 0.9534069485591166,
      "grad_norm": 0.9441967472268544,
      "learning_rate": 5.366141761967287e-06,
      "loss": 0.35953350067138673,
      "num_input_tokens_seen": 231889288,
      "step": 14160,
      "train_runtime": 27225.4792,
      "train_tokens_per_second": 8517.363
    },
    {
      "epoch": 0.9537436035550767,
      "grad_norm": 0.9964464162693835,
      "learning_rate": 5.363504727571179e-06,
      "loss": 0.3969372987747192,
      "num_input_tokens_seen": 231971208,
      "step": 14165,
      "train_runtime": 27235.0392,
      "train_tokens_per_second": 8517.381
    },
    {
      "epoch": 0.954080258551037,
      "grad_norm": 1.2394213312253732,
      "learning_rate": 5.360867591522133e-06,
      "loss": 0.4247234344482422,
      "num_input_tokens_seen": 232053128,
      "step": 14170,
      "train_runtime": 27244.6146,
      "train_tokens_per_second": 8517.394
    },
    {
      "epoch": 0.9544169135469971,
      "grad_norm": 1.0835824109281809,
      "learning_rate": 5.358230354557617e-06,
      "loss": 0.35178170204162595,
      "num_input_tokens_seen": 232135048,
      "step": 14175,
      "train_runtime": 27254.2408,
      "train_tokens_per_second": 8517.392
    },
    {
      "epoch": 0.9547535685429572,
      "grad_norm": 1.0898185451412634,
      "learning_rate": 5.355593017415123e-06,
      "loss": 0.4194737434387207,
      "num_input_tokens_seen": 232216968,
      "step": 14180,
      "train_runtime": 27263.8675,
      "train_tokens_per_second": 8517.389
    },
    {
      "epoch": 0.9550902235389174,
      "grad_norm": 1.0918640976246656,
      "learning_rate": 5.352955580832176e-06,
      "loss": 0.3953291654586792,
      "num_input_tokens_seen": 232298888,
      "step": 14185,
      "train_runtime": 27273.3777,
      "train_tokens_per_second": 8517.423
    },
    {
      "epoch": 0.9554268785348775,
      "grad_norm": 1.058057410461804,
      "learning_rate": 5.350318045546326e-06,
      "loss": 0.3809826135635376,
      "num_input_tokens_seen": 232380808,
      "step": 14190,
      "train_runtime": 27283.0061,
      "train_tokens_per_second": 8517.419
    },
    {
      "epoch": 0.9557635335308377,
      "grad_norm": 1.1975275748502414,
      "learning_rate": 5.347680412295152e-06,
      "loss": 0.41631031036376953,
      "num_input_tokens_seen": 232462728,
      "step": 14195,
      "train_runtime": 27292.6067,
      "train_tokens_per_second": 8517.425
    },
    {
      "epoch": 0.9561001885267978,
      "grad_norm": 1.1838332558228304,
      "learning_rate": 5.34504268181626e-06,
      "loss": 0.4016422271728516,
      "num_input_tokens_seen": 232544648,
      "step": 14200,
      "train_runtime": 27302.1138,
      "train_tokens_per_second": 8517.459
    },
    {
      "epoch": 0.9564368435227579,
      "grad_norm": 1.1886800149786396,
      "learning_rate": 5.342404854847282e-06,
      "loss": 0.39684252738952636,
      "num_input_tokens_seen": 232626568,
      "step": 14205,
      "train_runtime": 27311.7076,
      "train_tokens_per_second": 8517.467
    },
    {
      "epoch": 0.9567734985187181,
      "grad_norm": 1.0847934862536097,
      "learning_rate": 5.3397669321258776e-06,
      "loss": 0.36311337947845457,
      "num_input_tokens_seen": 232708488,
      "step": 14210,
      "train_runtime": 27321.3009,
      "train_tokens_per_second": 8517.475
    },
    {
      "epoch": 0.9571101535146782,
      "grad_norm": 1.2484093976439108,
      "learning_rate": 5.337128914389734e-06,
      "loss": 0.4206860542297363,
      "num_input_tokens_seen": 232790408,
      "step": 14215,
      "train_runtime": 27330.7984,
      "train_tokens_per_second": 8517.512
    },
    {
      "epoch": 0.9574468085106383,
      "grad_norm": 1.0596049210103202,
      "learning_rate": 5.334490802376565e-06,
      "loss": 0.40004701614379884,
      "num_input_tokens_seen": 232872328,
      "step": 14220,
      "train_runtime": 27340.4069,
      "train_tokens_per_second": 8517.515
    },
    {
      "epoch": 0.9577834635065985,
      "grad_norm": 1.2699498272379635,
      "learning_rate": 5.33185259682411e-06,
      "loss": 0.3817143440246582,
      "num_input_tokens_seen": 232954248,
      "step": 14225,
      "train_runtime": 27350.0252,
      "train_tokens_per_second": 8517.515
    },
    {
      "epoch": 0.9581201185025586,
      "grad_norm": 1.0883441159132248,
      "learning_rate": 5.329214298470134e-06,
      "loss": 0.3982752799987793,
      "num_input_tokens_seen": 233036168,
      "step": 14230,
      "train_runtime": 27359.5328,
      "train_tokens_per_second": 8517.549
    },
    {
      "epoch": 0.9584567734985188,
      "grad_norm": 1.1331186318684012,
      "learning_rate": 5.326575908052428e-06,
      "loss": 0.3835842847824097,
      "num_input_tokens_seen": 233118088,
      "step": 14235,
      "train_runtime": 27369.1357,
      "train_tokens_per_second": 8517.554
    },
    {
      "epoch": 0.9587934284944789,
      "grad_norm": 1.2055891002520784,
      "learning_rate": 5.323937426308813e-06,
      "loss": 0.44475631713867186,
      "num_input_tokens_seen": 233200008,
      "step": 14240,
      "train_runtime": 27378.7529,
      "train_tokens_per_second": 8517.554
    },
    {
      "epoch": 0.959130083490439,
      "grad_norm": 1.0215899232254462,
      "learning_rate": 5.321298853977128e-06,
      "loss": 0.33727293014526366,
      "num_input_tokens_seen": 233281928,
      "step": 14245,
      "train_runtime": 27388.3039,
      "train_tokens_per_second": 8517.575
    },
    {
      "epoch": 0.9594667384863992,
      "grad_norm": 1.1185581419723243,
      "learning_rate": 5.318660191795244e-06,
      "loss": 0.3662278652191162,
      "num_input_tokens_seen": 233363848,
      "step": 14250,
      "train_runtime": 27397.8819,
      "train_tokens_per_second": 8517.587
    },
    {
      "epoch": 0.9598033934823593,
      "grad_norm": 1.0680302003266315,
      "learning_rate": 5.316021440501053e-06,
      "loss": 0.3388226509094238,
      "num_input_tokens_seen": 233445768,
      "step": 14255,
      "train_runtime": 27407.507,
      "train_tokens_per_second": 8517.585
    },
    {
      "epoch": 0.9601400484783195,
      "grad_norm": 1.0818828609565578,
      "learning_rate": 5.313382600832474e-06,
      "loss": 0.37340354919433594,
      "num_input_tokens_seen": 233527688,
      "step": 14260,
      "train_runtime": 27417.0598,
      "train_tokens_per_second": 8517.605
    },
    {
      "epoch": 0.9604767034742796,
      "grad_norm": 1.1780183005270992,
      "learning_rate": 5.310743673527451e-06,
      "loss": 0.41126222610473634,
      "num_input_tokens_seen": 233609608,
      "step": 14265,
      "train_runtime": 27426.6289,
      "train_tokens_per_second": 8517.62
    },
    {
      "epoch": 0.9608133584702397,
      "grad_norm": 1.0045694526270215,
      "learning_rate": 5.3081046593239496e-06,
      "loss": 0.39419260025024416,
      "num_input_tokens_seen": 233691528,
      "step": 14270,
      "train_runtime": 27436.2201,
      "train_tokens_per_second": 8517.628
    },
    {
      "epoch": 0.9611500134661999,
      "grad_norm": 1.0042418927944714,
      "learning_rate": 5.305465558959964e-06,
      "loss": 0.3930588483810425,
      "num_input_tokens_seen": 233773448,
      "step": 14275,
      "train_runtime": 27445.7584,
      "train_tokens_per_second": 8517.653
    },
    {
      "epoch": 0.96148666846216,
      "grad_norm": 1.0953243056820736,
      "learning_rate": 5.302826373173506e-06,
      "loss": 0.3870546340942383,
      "num_input_tokens_seen": 233855368,
      "step": 14280,
      "train_runtime": 27455.2739,
      "train_tokens_per_second": 8517.685
    },
    {
      "epoch": 0.9618233234581202,
      "grad_norm": 1.0694840921667736,
      "learning_rate": 5.3001871027026194e-06,
      "loss": 0.3594978332519531,
      "num_input_tokens_seen": 233937288,
      "step": 14285,
      "train_runtime": 27464.8739,
      "train_tokens_per_second": 8517.69
    },
    {
      "epoch": 0.9621599784540803,
      "grad_norm": 1.142911050405202,
      "learning_rate": 5.2975477482853685e-06,
      "loss": 0.37065303325653076,
      "num_input_tokens_seen": 234019208,
      "step": 14290,
      "train_runtime": 27474.4922,
      "train_tokens_per_second": 8517.69
    },
    {
      "epoch": 0.9624966334500404,
      "grad_norm": 1.0157054252321864,
      "learning_rate": 5.294908310659838e-06,
      "loss": 0.340985107421875,
      "num_input_tokens_seen": 234101128,
      "step": 14295,
      "train_runtime": 27483.9896,
      "train_tokens_per_second": 8517.727
    },
    {
      "epoch": 0.9628332884460006,
      "grad_norm": 1.1340440094943676,
      "learning_rate": 5.292268790564138e-06,
      "loss": 0.4177098751068115,
      "num_input_tokens_seen": 234183048,
      "step": 14300,
      "train_runtime": 27493.5846,
      "train_tokens_per_second": 8517.734
    },
    {
      "epoch": 0.9631699434419607,
      "grad_norm": 1.0799793628307337,
      "learning_rate": 5.289629188736403e-06,
      "loss": 0.4064041614532471,
      "num_input_tokens_seen": 234264968,
      "step": 14305,
      "train_runtime": 27503.131,
      "train_tokens_per_second": 8517.756
    },
    {
      "epoch": 0.9635065984379209,
      "grad_norm": 1.146966782901632,
      "learning_rate": 5.286989505914788e-06,
      "loss": 0.3898423671722412,
      "num_input_tokens_seen": 234346888,
      "step": 14310,
      "train_runtime": 27512.6307,
      "train_tokens_per_second": 8517.793
    },
    {
      "epoch": 0.963843253433881,
      "grad_norm": 1.0757750307675,
      "learning_rate": 5.284349742837475e-06,
      "loss": 0.371588134765625,
      "num_input_tokens_seen": 234428808,
      "step": 14315,
      "train_runtime": 27522.2273,
      "train_tokens_per_second": 8517.799
    },
    {
      "epoch": 0.9641799084298411,
      "grad_norm": 1.2092459953849741,
      "learning_rate": 5.281709900242662e-06,
      "loss": 0.3671190023422241,
      "num_input_tokens_seen": 234510728,
      "step": 14320,
      "train_runtime": 27531.7815,
      "train_tokens_per_second": 8517.819
    },
    {
      "epoch": 0.9645165634258013,
      "grad_norm": 1.0531356402488679,
      "learning_rate": 5.279069978868574e-06,
      "loss": 0.3849951267242432,
      "num_input_tokens_seen": 234592648,
      "step": 14325,
      "train_runtime": 27541.2901,
      "train_tokens_per_second": 8517.853
    },
    {
      "epoch": 0.9648532184217614,
      "grad_norm": 0.9760840320846786,
      "learning_rate": 5.276429979453455e-06,
      "loss": 0.39730358123779297,
      "num_input_tokens_seen": 234674568,
      "step": 14330,
      "train_runtime": 27550.8689,
      "train_tokens_per_second": 8517.864
    },
    {
      "epoch": 0.9651898734177216,
      "grad_norm": 1.0569846476954652,
      "learning_rate": 5.273789902735575e-06,
      "loss": 0.40018014907836913,
      "num_input_tokens_seen": 234756488,
      "step": 14335,
      "train_runtime": 27560.4671,
      "train_tokens_per_second": 8517.87
    },
    {
      "epoch": 0.9655265284136817,
      "grad_norm": 1.0911959679162109,
      "learning_rate": 5.27114974945322e-06,
      "loss": 0.37135236263275145,
      "num_input_tokens_seen": 234838408,
      "step": 14340,
      "train_runtime": 27569.9755,
      "train_tokens_per_second": 8517.904
    },
    {
      "epoch": 0.9658631834096418,
      "grad_norm": 1.3594853409066383,
      "learning_rate": 5.2685095203447046e-06,
      "loss": 0.38461596965789796,
      "num_input_tokens_seen": 234920328,
      "step": 14345,
      "train_runtime": 27579.6109,
      "train_tokens_per_second": 8517.899
    },
    {
      "epoch": 0.966199838405602,
      "grad_norm": 1.016354653189513,
      "learning_rate": 5.265869216148357e-06,
      "loss": 0.3638118028640747,
      "num_input_tokens_seen": 235002248,
      "step": 14350,
      "train_runtime": 27589.2201,
      "train_tokens_per_second": 8517.901
    },
    {
      "epoch": 0.9665364934015621,
      "grad_norm": 1.0990760924039489,
      "learning_rate": 5.263228837602532e-06,
      "loss": 0.4342654228210449,
      "num_input_tokens_seen": 235084168,
      "step": 14355,
      "train_runtime": 27598.7803,
      "train_tokens_per_second": 8517.919
    },
    {
      "epoch": 0.9668731483975223,
      "grad_norm": 1.0069855511662638,
      "learning_rate": 5.2605883854456e-06,
      "loss": 0.4060643672943115,
      "num_input_tokens_seen": 235166088,
      "step": 14360,
      "train_runtime": 27608.3484,
      "train_tokens_per_second": 8517.934
    },
    {
      "epoch": 0.9672098033934824,
      "grad_norm": 0.9328926999924906,
      "learning_rate": 5.257947860415957e-06,
      "loss": 0.3785711765289307,
      "num_input_tokens_seen": 235248008,
      "step": 14365,
      "train_runtime": 27617.9473,
      "train_tokens_per_second": 8517.94
    },
    {
      "epoch": 0.9675464583894425,
      "grad_norm": 1.1408616264870555,
      "learning_rate": 5.255307263252019e-06,
      "loss": 0.38107995986938475,
      "num_input_tokens_seen": 235329928,
      "step": 14370,
      "train_runtime": 27627.5018,
      "train_tokens_per_second": 8517.959
    },
    {
      "epoch": 0.9678831133854027,
      "grad_norm": 1.0305190186659239,
      "learning_rate": 5.2526665946922185e-06,
      "loss": 0.41966900825500486,
      "num_input_tokens_seen": 235411848,
      "step": 14375,
      "train_runtime": 27637.0646,
      "train_tokens_per_second": 8517.976
    },
    {
      "epoch": 0.9682197683813628,
      "grad_norm": 0.9587051702731253,
      "learning_rate": 5.25002585547501e-06,
      "loss": 0.3553262233734131,
      "num_input_tokens_seen": 235493768,
      "step": 14380,
      "train_runtime": 27646.6735,
      "train_tokens_per_second": 8517.978
    },
    {
      "epoch": 0.968556423377323,
      "grad_norm": 1.091412823278883,
      "learning_rate": 5.2473850463388675e-06,
      "loss": 0.4188982486724854,
      "num_input_tokens_seen": 235575688,
      "step": 14385,
      "train_runtime": 27656.2885,
      "train_tokens_per_second": 8517.979
    },
    {
      "epoch": 0.9688930783732831,
      "grad_norm": 1.1739650157089556,
      "learning_rate": 5.244744168022288e-06,
      "loss": 0.3735010862350464,
      "num_input_tokens_seen": 235657608,
      "step": 14390,
      "train_runtime": 27665.7886,
      "train_tokens_per_second": 8518.015
    },
    {
      "epoch": 0.9692297333692432,
      "grad_norm": 1.0411634609232383,
      "learning_rate": 5.24210322126378e-06,
      "loss": 0.3947759628295898,
      "num_input_tokens_seen": 235739528,
      "step": 14395,
      "train_runtime": 27675.3545,
      "train_tokens_per_second": 8518.031
    },
    {
      "epoch": 0.9695663883652034,
      "grad_norm": 1.1876704361535173,
      "learning_rate": 5.23946220680188e-06,
      "loss": 0.3778035640716553,
      "num_input_tokens_seen": 235821448,
      "step": 14400,
      "train_runtime": 27684.955,
      "train_tokens_per_second": 8518.036
    },
    {
      "epoch": 0.9699030433611635,
      "grad_norm": 1.092517304386831,
      "learning_rate": 5.236821125375136e-06,
      "loss": 0.36174955368041994,
      "num_input_tokens_seen": 235903368,
      "step": 14405,
      "train_runtime": 27694.4598,
      "train_tokens_per_second": 8518.071
    },
    {
      "epoch": 0.9702396983571236,
      "grad_norm": 1.0710240207992408,
      "learning_rate": 5.234179977722119e-06,
      "loss": 0.3614306926727295,
      "num_input_tokens_seen": 235985288,
      "step": 14410,
      "train_runtime": 27704.0858,
      "train_tokens_per_second": 8518.068
    },
    {
      "epoch": 0.9705763533530838,
      "grad_norm": 1.0995041669310517,
      "learning_rate": 5.23153876458142e-06,
      "loss": 0.3931384563446045,
      "num_input_tokens_seen": 236067208,
      "step": 14415,
      "train_runtime": 27713.7026,
      "train_tokens_per_second": 8518.068
    },
    {
      "epoch": 0.9709130083490439,
      "grad_norm": 0.9743378255380298,
      "learning_rate": 5.2288974866916435e-06,
      "loss": 0.38429386615753175,
      "num_input_tokens_seen": 236149128,
      "step": 14420,
      "train_runtime": 27723.211,
      "train_tokens_per_second": 8518.102
    },
    {
      "epoch": 0.9712496633450041,
      "grad_norm": 1.0692646901011715,
      "learning_rate": 5.226256144791414e-06,
      "loss": 0.387548828125,
      "num_input_tokens_seen": 236231048,
      "step": 14425,
      "train_runtime": 27732.8208,
      "train_tokens_per_second": 8518.104
    },
    {
      "epoch": 0.9715863183409642,
      "grad_norm": 1.0420161827776007,
      "learning_rate": 5.223614739619375e-06,
      "loss": 0.37428627014160154,
      "num_input_tokens_seen": 236312968,
      "step": 14430,
      "train_runtime": 27742.4313,
      "train_tokens_per_second": 8518.106
    },
    {
      "epoch": 0.9719229733369243,
      "grad_norm": 0.9707412866582746,
      "learning_rate": 5.2209732719141855e-06,
      "loss": 0.36239075660705566,
      "num_input_tokens_seen": 236394888,
      "step": 14435,
      "train_runtime": 27751.9371,
      "train_tokens_per_second": 8518.14
    },
    {
      "epoch": 0.9722596283328845,
      "grad_norm": 1.0392640040682295,
      "learning_rate": 5.218331742414528e-06,
      "loss": 0.39075753688812254,
      "num_input_tokens_seen": 236476808,
      "step": 14440,
      "train_runtime": 27761.5741,
      "train_tokens_per_second": 8518.134
    },
    {
      "epoch": 0.9725962833288446,
      "grad_norm": 1.050559537352533,
      "learning_rate": 5.215690151859094e-06,
      "loss": 0.3597720146179199,
      "num_input_tokens_seen": 236558728,
      "step": 14445,
      "train_runtime": 27771.175,
      "train_tokens_per_second": 8518.139
    },
    {
      "epoch": 0.9729329383248048,
      "grad_norm": 1.0882758976709024,
      "learning_rate": 5.213048500986597e-06,
      "loss": 0.398499870300293,
      "num_input_tokens_seen": 236640648,
      "step": 14450,
      "train_runtime": 27780.7248,
      "train_tokens_per_second": 8518.16
    },
    {
      "epoch": 0.9732695933207649,
      "grad_norm": 1.1559772521389855,
      "learning_rate": 5.210406790535766e-06,
      "loss": 0.4068995475769043,
      "num_input_tokens_seen": 236722568,
      "step": 14455,
      "train_runtime": 27790.3008,
      "train_tokens_per_second": 8518.172
    },
    {
      "epoch": 0.973606248316725,
      "grad_norm": 1.0795976589818672,
      "learning_rate": 5.207765021245346e-06,
      "loss": 0.36990258693695066,
      "num_input_tokens_seen": 236804488,
      "step": 14460,
      "train_runtime": 27799.8946,
      "train_tokens_per_second": 8518.179
    },
    {
      "epoch": 0.9739429033126852,
      "grad_norm": 1.0223180043562103,
      "learning_rate": 5.2051231938541015e-06,
      "loss": 0.36394453048706055,
      "num_input_tokens_seen": 236886408,
      "step": 14465,
      "train_runtime": 27809.4383,
      "train_tokens_per_second": 8518.202
    },
    {
      "epoch": 0.9742795583086453,
      "grad_norm": 1.106975653422436,
      "learning_rate": 5.202481309100808e-06,
      "loss": 0.3847886323928833,
      "num_input_tokens_seen": 236968328,
      "step": 14470,
      "train_runtime": 27819.003,
      "train_tokens_per_second": 8518.218
    },
    {
      "epoch": 0.9746162133046055,
      "grad_norm": 1.1780555686345577,
      "learning_rate": 5.199839367724266e-06,
      "loss": 0.38728573322296145,
      "num_input_tokens_seen": 237050248,
      "step": 14475,
      "train_runtime": 27828.6354,
      "train_tokens_per_second": 8518.213
    },
    {
      "epoch": 0.9749528683005656,
      "grad_norm": 0.9808170297153732,
      "learning_rate": 5.197197370463278e-06,
      "loss": 0.3559496641159058,
      "num_input_tokens_seen": 237132168,
      "step": 14480,
      "train_runtime": 27838.1929,
      "train_tokens_per_second": 8518.231
    },
    {
      "epoch": 0.9752895232965257,
      "grad_norm": 1.1757324288765476,
      "learning_rate": 5.194555318056675e-06,
      "loss": 0.41885828971862793,
      "num_input_tokens_seen": 237214088,
      "step": 14485,
      "train_runtime": 27847.7493,
      "train_tokens_per_second": 8518.25
    },
    {
      "epoch": 0.9756261782924859,
      "grad_norm": 1.2981832844956966,
      "learning_rate": 5.191913211243299e-06,
      "loss": 0.3948268175125122,
      "num_input_tokens_seen": 237296008,
      "step": 14490,
      "train_runtime": 27857.3437,
      "train_tokens_per_second": 8518.257
    },
    {
      "epoch": 0.975962833288446,
      "grad_norm": 1.0410327559634627,
      "learning_rate": 5.1892710507620055e-06,
      "loss": 0.3662563323974609,
      "num_input_tokens_seen": 237377928,
      "step": 14495,
      "train_runtime": 27866.9008,
      "train_tokens_per_second": 8518.275
    },
    {
      "epoch": 0.9762994882844062,
      "grad_norm": 0.9934590588126008,
      "learning_rate": 5.186628837351666e-06,
      "loss": 0.3543086528778076,
      "num_input_tokens_seen": 237459848,
      "step": 14500,
      "train_runtime": 27876.4697,
      "train_tokens_per_second": 8518.29
    },
    {
      "epoch": 0.9766361432803663,
      "grad_norm": 0.9973082310765855,
      "learning_rate": 5.183986571751167e-06,
      "loss": 0.3894719123840332,
      "num_input_tokens_seen": 237541768,
      "step": 14505,
      "train_runtime": 27886.0847,
      "train_tokens_per_second": 8518.29
    },
    {
      "epoch": 0.9769727982763264,
      "grad_norm": 1.0958544680413247,
      "learning_rate": 5.181344254699408e-06,
      "loss": 0.37988078594207764,
      "num_input_tokens_seen": 237623688,
      "step": 14510,
      "train_runtime": 27895.6871,
      "train_tokens_per_second": 8518.295
    },
    {
      "epoch": 0.9773094532722866,
      "grad_norm": 0.9207487115356888,
      "learning_rate": 5.178701886935309e-06,
      "loss": 0.35365638732910154,
      "num_input_tokens_seen": 237705608,
      "step": 14515,
      "train_runtime": 27905.187,
      "train_tokens_per_second": 8518.331
    },
    {
      "epoch": 0.9776461082682467,
      "grad_norm": 1.3127397721677534,
      "learning_rate": 5.176059469197796e-06,
      "loss": 0.3813375949859619,
      "num_input_tokens_seen": 237787528,
      "step": 14520,
      "train_runtime": 27914.8193,
      "train_tokens_per_second": 8518.326
    },
    {
      "epoch": 0.9779827632642069,
      "grad_norm": 0.9898377153566703,
      "learning_rate": 5.173417002225815e-06,
      "loss": 0.3416694164276123,
      "num_input_tokens_seen": 237869448,
      "step": 14525,
      "train_runtime": 27924.4393,
      "train_tokens_per_second": 8518.325
    },
    {
      "epoch": 0.978319418260167,
      "grad_norm": 0.9855704653793925,
      "learning_rate": 5.170774486758322e-06,
      "loss": 0.4066044807434082,
      "num_input_tokens_seen": 237951368,
      "step": 14530,
      "train_runtime": 27933.9408,
      "train_tokens_per_second": 8518.36
    },
    {
      "epoch": 0.9786560732561271,
      "grad_norm": 1.0269841404808944,
      "learning_rate": 5.168131923534287e-06,
      "loss": 0.42747836112976073,
      "num_input_tokens_seen": 238033288,
      "step": 14535,
      "train_runtime": 27943.5571,
      "train_tokens_per_second": 8518.36
    },
    {
      "epoch": 0.9789927282520873,
      "grad_norm": 0.898180181003563,
      "learning_rate": 5.165489313292696e-06,
      "loss": 0.34574360847473146,
      "num_input_tokens_seen": 238115208,
      "step": 14540,
      "train_runtime": 27953.1645,
      "train_tokens_per_second": 8518.363
    },
    {
      "epoch": 0.9793293832480474,
      "grad_norm": 1.027767657074714,
      "learning_rate": 5.162846656772548e-06,
      "loss": 0.38202714920043945,
      "num_input_tokens_seen": 238197128,
      "step": 14545,
      "train_runtime": 27962.6651,
      "train_tokens_per_second": 8518.399
    },
    {
      "epoch": 0.9796660382440076,
      "grad_norm": 1.164626716102323,
      "learning_rate": 5.16020395471285e-06,
      "loss": 0.3559075117111206,
      "num_input_tokens_seen": 238279048,
      "step": 14550,
      "train_runtime": 27972.2811,
      "train_tokens_per_second": 8518.399
    },
    {
      "epoch": 0.9800026932399677,
      "grad_norm": 1.067164024215344,
      "learning_rate": 5.157561207852628e-06,
      "loss": 0.44065370559692385,
      "num_input_tokens_seen": 238360192,
      "step": 14555,
      "train_runtime": 27981.892,
      "train_tokens_per_second": 8518.373
    },
    {
      "epoch": 0.9803393482359278,
      "grad_norm": 1.0036912500492723,
      "learning_rate": 5.154918416930915e-06,
      "loss": 0.3384669542312622,
      "num_input_tokens_seen": 238442112,
      "step": 14560,
      "train_runtime": 27991.4425,
      "train_tokens_per_second": 8518.393
    },
    {
      "epoch": 0.980676003231888,
      "grad_norm": 1.1010995273885242,
      "learning_rate": 5.152275582686761e-06,
      "loss": 0.41267170906066897,
      "num_input_tokens_seen": 238524032,
      "step": 14565,
      "train_runtime": 28001.0139,
      "train_tokens_per_second": 8518.407
    },
    {
      "epoch": 0.9810126582278481,
      "grad_norm": 0.9864173133597162,
      "learning_rate": 5.149632705859227e-06,
      "loss": 0.38809547424316404,
      "num_input_tokens_seen": 238605952,
      "step": 14570,
      "train_runtime": 28010.5716,
      "train_tokens_per_second": 8518.425
    },
    {
      "epoch": 0.9813493132238083,
      "grad_norm": 1.2634870247964538,
      "learning_rate": 5.146989787187383e-06,
      "loss": 0.41362428665161133,
      "num_input_tokens_seen": 238687872,
      "step": 14575,
      "train_runtime": 28020.1181,
      "train_tokens_per_second": 8518.446
    },
    {
      "epoch": 0.9816859682197684,
      "grad_norm": 0.8723938451395796,
      "learning_rate": 5.144346827410314e-06,
      "loss": 0.3704625129699707,
      "num_input_tokens_seen": 238769792,
      "step": 14580,
      "train_runtime": 28029.687,
      "train_tokens_per_second": 8518.461
    },
    {
      "epoch": 0.9820226232157285,
      "grad_norm": 1.1410659347271626,
      "learning_rate": 5.141703827267117e-06,
      "loss": 0.43834972381591797,
      "num_input_tokens_seen": 238851712,
      "step": 14585,
      "train_runtime": 28039.318,
      "train_tokens_per_second": 8518.457
    },
    {
      "epoch": 0.9823592782116887,
      "grad_norm": 1.2500077680520665,
      "learning_rate": 5.139060787496893e-06,
      "loss": 0.42435832023620607,
      "num_input_tokens_seen": 238933632,
      "step": 14590,
      "train_runtime": 28048.8756,
      "train_tokens_per_second": 8518.475
    },
    {
      "epoch": 0.9826959332076488,
      "grad_norm": 1.0372048726479555,
      "learning_rate": 5.136417708838766e-06,
      "loss": 0.3808101177215576,
      "num_input_tokens_seen": 239015552,
      "step": 14595,
      "train_runtime": 28058.4477,
      "train_tokens_per_second": 8518.488
    },
    {
      "epoch": 0.983032588203609,
      "grad_norm": 1.0449891332029901,
      "learning_rate": 5.13377459203186e-06,
      "loss": 0.3923546075820923,
      "num_input_tokens_seen": 239097472,
      "step": 14600,
      "train_runtime": 28068.0901,
      "train_tokens_per_second": 8518.48
    },
    {
      "epoch": 0.9833692431995691,
      "grad_norm": 0.9918969742871736,
      "learning_rate": 5.131131437815318e-06,
      "loss": 0.40102739334106446,
      "num_input_tokens_seen": 239179392,
      "step": 14605,
      "train_runtime": 28077.7186,
      "train_tokens_per_second": 8518.477
    },
    {
      "epoch": 0.9837058981955292,
      "grad_norm": 0.885360191273499,
      "learning_rate": 5.128488246928287e-06,
      "loss": 0.40128579139709475,
      "num_input_tokens_seen": 239261312,
      "step": 14610,
      "train_runtime": 28087.22,
      "train_tokens_per_second": 8518.512
    },
    {
      "epoch": 0.9840425531914894,
      "grad_norm": 1.2020907922303223,
      "learning_rate": 5.125845020109925e-06,
      "loss": 0.4022219181060791,
      "num_input_tokens_seen": 239343232,
      "step": 14615,
      "train_runtime": 28096.7808,
      "train_tokens_per_second": 8518.529
    },
    {
      "epoch": 0.9843792081874495,
      "grad_norm": 1.1877898837836867,
      "learning_rate": 5.123201758099406e-06,
      "loss": 0.4085413932800293,
      "num_input_tokens_seen": 239425152,
      "step": 14620,
      "train_runtime": 28106.4035,
      "train_tokens_per_second": 8518.527
    },
    {
      "epoch": 0.9847158631834096,
      "grad_norm": 1.234122911458847,
      "learning_rate": 5.12055846163591e-06,
      "loss": 0.3797164440155029,
      "num_input_tokens_seen": 239506784,
      "step": 14625,
      "train_runtime": 28116.3227,
      "train_tokens_per_second": 8518.425
    },
    {
      "epoch": 0.9850525181793698,
      "grad_norm": 0.9892106277541576,
      "learning_rate": 5.117915131458623e-06,
      "loss": 0.35406179428100587,
      "num_input_tokens_seen": 239588704,
      "step": 14630,
      "train_runtime": 28125.9379,
      "train_tokens_per_second": 8518.425
    },
    {
      "epoch": 0.9853891731753299,
      "grad_norm": 1.061868133779673,
      "learning_rate": 5.115271768306745e-06,
      "loss": 0.3778548240661621,
      "num_input_tokens_seen": 239670624,
      "step": 14635,
      "train_runtime": 28135.5408,
      "train_tokens_per_second": 8518.43
    },
    {
      "epoch": 0.9857258281712901,
      "grad_norm": 1.0031642668778111,
      "learning_rate": 5.112628372919485e-06,
      "loss": 0.4055961608886719,
      "num_input_tokens_seen": 239752544,
      "step": 14640,
      "train_runtime": 28145.0474,
      "train_tokens_per_second": 8518.463
    },
    {
      "epoch": 0.9860624831672502,
      "grad_norm": 0.9227198048442129,
      "learning_rate": 5.1099849460360595e-06,
      "loss": 0.3848163604736328,
      "num_input_tokens_seen": 239834464,
      "step": 14645,
      "train_runtime": 28154.6655,
      "train_tokens_per_second": 8518.463
    },
    {
      "epoch": 0.9863991381632103,
      "grad_norm": 0.9540527526452767,
      "learning_rate": 5.107341488395695e-06,
      "loss": 0.36407768726348877,
      "num_input_tokens_seen": 239916384,
      "step": 14650,
      "train_runtime": 28164.2757,
      "train_tokens_per_second": 8518.465
    },
    {
      "epoch": 0.9867357931591705,
      "grad_norm": 1.1483043471779735,
      "learning_rate": 5.104698000737626e-06,
      "loss": 0.3872018575668335,
      "num_input_tokens_seen": 239998304,
      "step": 14655,
      "train_runtime": 28173.8231,
      "train_tokens_per_second": 8518.486
    },
    {
      "epoch": 0.9870724481551306,
      "grad_norm": 1.1357276636469316,
      "learning_rate": 5.102054483801094e-06,
      "loss": 0.4128662109375,
      "num_input_tokens_seen": 240080224,
      "step": 14660,
      "train_runtime": 28183.3853,
      "train_tokens_per_second": 8518.502
    },
    {
      "epoch": 0.9874091031510908,
      "grad_norm": 1.0165292126932153,
      "learning_rate": 5.099410938325351e-06,
      "loss": 0.422945499420166,
      "num_input_tokens_seen": 240162144,
      "step": 14665,
      "train_runtime": 28193.0012,
      "train_tokens_per_second": 8518.502
    },
    {
      "epoch": 0.9877457581470509,
      "grad_norm": 1.0525909824310191,
      "learning_rate": 5.096767365049657e-06,
      "loss": 0.35707836151123046,
      "num_input_tokens_seen": 240244064,
      "step": 14670,
      "train_runtime": 28202.5724,
      "train_tokens_per_second": 8518.516
    },
    {
      "epoch": 0.988082413143011,
      "grad_norm": 1.1298606515659415,
      "learning_rate": 5.094123764713277e-06,
      "loss": 0.3234975814819336,
      "num_input_tokens_seen": 240325984,
      "step": 14675,
      "train_runtime": 28212.1431,
      "train_tokens_per_second": 8518.53
    },
    {
      "epoch": 0.9884190681389712,
      "grad_norm": 1.0127023723286002,
      "learning_rate": 5.091480138055487e-06,
      "loss": 0.36692450046539304,
      "num_input_tokens_seen": 240407904,
      "step": 14680,
      "train_runtime": 28221.7528,
      "train_tokens_per_second": 8518.532
    },
    {
      "epoch": 0.9887557231349313,
      "grad_norm": 1.0972313799835374,
      "learning_rate": 5.088836485815568e-06,
      "loss": 0.4012859344482422,
      "num_input_tokens_seen": 240489664,
      "step": 14685,
      "train_runtime": 28231.3111,
      "train_tokens_per_second": 8518.544
    },
    {
      "epoch": 0.9890923781308915,
      "grad_norm": 1.0882859303098469,
      "learning_rate": 5.086192808732809e-06,
      "loss": 0.38886146545410155,
      "num_input_tokens_seen": 240571584,
      "step": 14690,
      "train_runtime": 28240.8927,
      "train_tokens_per_second": 8518.555
    },
    {
      "epoch": 0.9894290331268516,
      "grad_norm": 1.2117006199629168,
      "learning_rate": 5.083549107546505e-06,
      "loss": 0.4080349445343018,
      "num_input_tokens_seen": 240653504,
      "step": 14695,
      "train_runtime": 28250.5249,
      "train_tokens_per_second": 8518.55
    },
    {
      "epoch": 0.9897656881228117,
      "grad_norm": 1.0738081952505516,
      "learning_rate": 5.080905382995961e-06,
      "loss": 0.40523519515991213,
      "num_input_tokens_seen": 240735424,
      "step": 14700,
      "train_runtime": 28260.1443,
      "train_tokens_per_second": 8518.549
    },
    {
      "epoch": 0.9901023431187719,
      "grad_norm": 0.9755467065394909,
      "learning_rate": 5.0782616358204835e-06,
      "loss": 0.4101409912109375,
      "num_input_tokens_seen": 240817344,
      "step": 14705,
      "train_runtime": 28269.6554,
      "train_tokens_per_second": 8518.581
    },
    {
      "epoch": 0.990438998114732,
      "grad_norm": 0.9772964937169322,
      "learning_rate": 5.075617866759389e-06,
      "loss": 0.3540593385696411,
      "num_input_tokens_seen": 240899264,
      "step": 14710,
      "train_runtime": 28279.2746,
      "train_tokens_per_second": 8518.58
    },
    {
      "epoch": 0.9907756531106922,
      "grad_norm": 1.0072388966813695,
      "learning_rate": 5.0729740765519995e-06,
      "loss": 0.42945079803466796,
      "num_input_tokens_seen": 240981184,
      "step": 14715,
      "train_runtime": 28288.887,
      "train_tokens_per_second": 8518.581
    },
    {
      "epoch": 0.9911123081066523,
      "grad_norm": 1.0808633244791694,
      "learning_rate": 5.070330265937643e-06,
      "loss": 0.3833989858627319,
      "num_input_tokens_seen": 241063104,
      "step": 14720,
      "train_runtime": 28298.3943,
      "train_tokens_per_second": 8518.614
    },
    {
      "epoch": 0.9914489631026124,
      "grad_norm": 1.166038968645016,
      "learning_rate": 5.067686435655651e-06,
      "loss": 0.40504064559936526,
      "num_input_tokens_seen": 241145024,
      "step": 14725,
      "train_runtime": 28308.0391,
      "train_tokens_per_second": 8518.606
    },
    {
      "epoch": 0.9917856180985726,
      "grad_norm": 0.9380105778480834,
      "learning_rate": 5.065042586445363e-06,
      "loss": 0.4027279853820801,
      "num_input_tokens_seen": 241226944,
      "step": 14730,
      "train_runtime": 28317.6617,
      "train_tokens_per_second": 8518.604
    },
    {
      "epoch": 0.9921222730945327,
      "grad_norm": 0.9987062690806032,
      "learning_rate": 5.0623987190461245e-06,
      "loss": 0.32041015625,
      "num_input_tokens_seen": 241308864,
      "step": 14735,
      "train_runtime": 28327.1645,
      "train_tokens_per_second": 8518.638
    },
    {
      "epoch": 0.9924589280904929,
      "grad_norm": 1.1572505464252514,
      "learning_rate": 5.059754834197281e-06,
      "loss": 0.4043837547302246,
      "num_input_tokens_seen": 241390784,
      "step": 14740,
      "train_runtime": 28336.768,
      "train_tokens_per_second": 8518.642
    },
    {
      "epoch": 0.992795583086453,
      "grad_norm": 1.0081490834232067,
      "learning_rate": 5.05711093263819e-06,
      "loss": 0.33425002098083495,
      "num_input_tokens_seen": 241472704,
      "step": 14745,
      "train_runtime": 28346.382,
      "train_tokens_per_second": 8518.643
    },
    {
      "epoch": 0.9931322380824131,
      "grad_norm": 1.1648548825066325,
      "learning_rate": 5.054467015108208e-06,
      "loss": 0.3988921642303467,
      "num_input_tokens_seen": 241554624,
      "step": 14750,
      "train_runtime": 28355.9365,
      "train_tokens_per_second": 8518.661
    },
    {
      "epoch": 0.9934688930783733,
      "grad_norm": 1.0423992535670183,
      "learning_rate": 5.051823082346701e-06,
      "loss": 0.4244471549987793,
      "num_input_tokens_seen": 241636544,
      "step": 14755,
      "train_runtime": 28365.4932,
      "train_tokens_per_second": 8518.679
    },
    {
      "epoch": 0.9938055480743334,
      "grad_norm": 1.5731444684441274,
      "learning_rate": 5.049179135093033e-06,
      "loss": 0.4119290351867676,
      "num_input_tokens_seen": 241718464,
      "step": 14760,
      "train_runtime": 28375.1193,
      "train_tokens_per_second": 8518.677
    },
    {
      "epoch": 0.9941422030702936,
      "grad_norm": 1.0653683107936933,
      "learning_rate": 5.046535174086575e-06,
      "loss": 0.4040102958679199,
      "num_input_tokens_seen": 241800384,
      "step": 14765,
      "train_runtime": 28384.6334,
      "train_tokens_per_second": 8518.707
    },
    {
      "epoch": 0.9944788580662537,
      "grad_norm": 1.1269675453997625,
      "learning_rate": 5.043891200066706e-06,
      "loss": 0.39289541244506837,
      "num_input_tokens_seen": 241882304,
      "step": 14770,
      "train_runtime": 28394.1996,
      "train_tokens_per_second": 8518.722
    },
    {
      "epoch": 0.9948155130622138,
      "grad_norm": 1.1454160001100302,
      "learning_rate": 5.041247213772802e-06,
      "loss": 0.3556215286254883,
      "num_input_tokens_seen": 241964224,
      "step": 14775,
      "train_runtime": 28403.8231,
      "train_tokens_per_second": 8518.72
    },
    {
      "epoch": 0.995152168058174,
      "grad_norm": 1.254691740302571,
      "learning_rate": 5.038603215944247e-06,
      "loss": 0.3981162548065186,
      "num_input_tokens_seen": 242045904,
      "step": 14780,
      "train_runtime": 28413.9398,
      "train_tokens_per_second": 8518.562
    },
    {
      "epoch": 0.9954888230541341,
      "grad_norm": 0.9532657985838747,
      "learning_rate": 5.0359592073204254e-06,
      "loss": 0.341628360748291,
      "num_input_tokens_seen": 242127264,
      "step": 14785,
      "train_runtime": 28423.7471,
      "train_tokens_per_second": 8518.485
    },
    {
      "epoch": 0.9958254780500942,
      "grad_norm": 1.1331515771230503,
      "learning_rate": 5.033315188640725e-06,
      "loss": 0.36317033767700196,
      "num_input_tokens_seen": 242209184,
      "step": 14790,
      "train_runtime": 28433.6521,
      "train_tokens_per_second": 8518.399
    },
    {
      "epoch": 0.9961621330460544,
      "grad_norm": 1.0371519624128136,
      "learning_rate": 5.03067116064454e-06,
      "loss": 0.351039457321167,
      "num_input_tokens_seen": 242291104,
      "step": 14795,
      "train_runtime": 28443.2665,
      "train_tokens_per_second": 8518.399
    },
    {
      "epoch": 0.9964987880420145,
      "grad_norm": 0.9414825721823017,
      "learning_rate": 5.028027124071262e-06,
      "loss": 0.40517354011535645,
      "num_input_tokens_seen": 242373024,
      "step": 14800,
      "train_runtime": 28452.7674,
      "train_tokens_per_second": 8518.434
    },
    {
      "epoch": 0.9968354430379747,
      "grad_norm": 1.141183301699909,
      "learning_rate": 5.025383079660289e-06,
      "loss": 0.35578439235687254,
      "num_input_tokens_seen": 242454944,
      "step": 14805,
      "train_runtime": 28462.391,
      "train_tokens_per_second": 8518.432
    },
    {
      "epoch": 0.9971720980339348,
      "grad_norm": 1.0953021854395024,
      "learning_rate": 5.0227390281510166e-06,
      "loss": 0.39389677047729493,
      "num_input_tokens_seen": 242536864,
      "step": 14810,
      "train_runtime": 28472.33,
      "train_tokens_per_second": 8518.336
    },
    {
      "epoch": 0.997508753029895,
      "grad_norm": 0.8840850712797221,
      "learning_rate": 5.020094970282849e-06,
      "loss": 0.3725303888320923,
      "num_input_tokens_seen": 242617832,
      "step": 14815,
      "train_runtime": 28482.0402,
      "train_tokens_per_second": 8518.274
    },
    {
      "epoch": 0.9978454080258551,
      "grad_norm": 0.866057851604808,
      "learning_rate": 5.017450906795187e-06,
      "loss": 0.35928323268890383,
      "num_input_tokens_seen": 242699752,
      "step": 14820,
      "train_runtime": 28491.6089,
      "train_tokens_per_second": 8518.289
    },
    {
      "epoch": 0.9981820630218152,
      "grad_norm": 0.8695636226727816,
      "learning_rate": 5.014806838427434e-06,
      "loss": 0.34984917640686036,
      "num_input_tokens_seen": 242781672,
      "step": 14825,
      "train_runtime": 28501.1945,
      "train_tokens_per_second": 8518.298
    },
    {
      "epoch": 0.9985187180177754,
      "grad_norm": 0.983965665333885,
      "learning_rate": 5.0121627659189964e-06,
      "loss": 0.39064669609069824,
      "num_input_tokens_seen": 242863592,
      "step": 14830,
      "train_runtime": 28510.7004,
      "train_tokens_per_second": 8518.331
    },
    {
      "epoch": 0.9988553730137355,
      "grad_norm": 1.1230593053108266,
      "learning_rate": 5.009518690009281e-06,
      "loss": 0.3707993507385254,
      "num_input_tokens_seen": 242945512,
      "step": 14835,
      "train_runtime": 28520.2673,
      "train_tokens_per_second": 8518.346
    },
    {
      "epoch": 0.9991920280096956,
      "grad_norm": 1.227234947245006,
      "learning_rate": 5.006874611437692e-06,
      "loss": 0.39643919467926025,
      "num_input_tokens_seen": 243027432,
      "step": 14840,
      "train_runtime": 28529.8596,
      "train_tokens_per_second": 8518.354
    },
    {
      "epoch": 0.9995286830056558,
      "grad_norm": 1.0391584988773863,
      "learning_rate": 5.004230530943641e-06,
      "loss": 0.40418319702148436,
      "num_input_tokens_seen": 243109352,
      "step": 14845,
      "train_runtime": 28539.4649,
      "train_tokens_per_second": 8518.357
    },
    {
      "epoch": 0.9998653380016159,
      "grad_norm": 0.8538054656951661,
      "learning_rate": 5.001586449266536e-06,
      "loss": 0.3173196792602539,
      "num_input_tokens_seen": 243191272,
      "step": 14850,
      "train_runtime": 28548.9712,
      "train_tokens_per_second": 8518.39
    },
    {
      "epoch": 1.000201992997576,
      "grad_norm": 0.9198183192887484,
      "learning_rate": 4.998942367145787e-06,
      "loss": 0.3430043935775757,
      "num_input_tokens_seen": 243272768,
      "step": 14855,
      "train_runtime": 28601.4108,
      "train_tokens_per_second": 8505.621
    },
    {
      "epoch": 1.0005386479935363,
      "grad_norm": 0.9934858065659632,
      "learning_rate": 4.996298285320798e-06,
      "loss": 0.3082250118255615,
      "num_input_tokens_seen": 243354688,
      "step": 14860,
      "train_runtime": 28610.9712,
      "train_tokens_per_second": 8505.642
    },
    {
      "epoch": 1.0008753029894963,
      "grad_norm": 0.9677681068060889,
      "learning_rate": 4.993654204530983e-06,
      "loss": 0.27727174758911133,
      "num_input_tokens_seen": 243436608,
      "step": 14865,
      "train_runtime": 28620.8162,
      "train_tokens_per_second": 8505.579
    },
    {
      "epoch": 1.0012119579854566,
      "grad_norm": 0.9138854309695523,
      "learning_rate": 4.99101012551575e-06,
      "loss": 0.29577538967132566,
      "num_input_tokens_seen": 243518528,
      "step": 14870,
      "train_runtime": 28630.413,
      "train_tokens_per_second": 8505.589
    },
    {
      "epoch": 1.0015486129814166,
      "grad_norm": 1.1144688353830279,
      "learning_rate": 4.9883660490145075e-06,
      "loss": 0.30466599464416505,
      "num_input_tokens_seen": 243600448,
      "step": 14875,
      "train_runtime": 28639.88,
      "train_tokens_per_second": 8505.638
    },
    {
      "epoch": 1.0018852679773769,
      "grad_norm": 0.9897705963591507,
      "learning_rate": 4.985721975766661e-06,
      "loss": 0.29644973278045655,
      "num_input_tokens_seen": 243682368,
      "step": 14880,
      "train_runtime": 28649.3489,
      "train_tokens_per_second": 8505.686
    },
    {
      "epoch": 1.002221922973337,
      "grad_norm": 1.1886879653154492,
      "learning_rate": 4.983077906511619e-06,
      "loss": 0.33289284706115724,
      "num_input_tokens_seen": 243764288,
      "step": 14885,
      "train_runtime": 28658.9988,
      "train_tokens_per_second": 8505.681
    },
    {
      "epoch": 1.0025585779692971,
      "grad_norm": 1.0795782037884032,
      "learning_rate": 4.980433841988786e-06,
      "loss": 0.3000157833099365,
      "num_input_tokens_seen": 243846208,
      "step": 14890,
      "train_runtime": 28668.6423,
      "train_tokens_per_second": 8505.677
    },
    {
      "epoch": 1.0028952329652572,
      "grad_norm": 1.0403600207068866,
      "learning_rate": 4.977789782937568e-06,
      "loss": 0.31911990642547605,
      "num_input_tokens_seen": 243928128,
      "step": 14895,
      "train_runtime": 28678.2257,
      "train_tokens_per_second": 8505.691
    },
    {
      "epoch": 1.0032318879612174,
      "grad_norm": 1.1822428208701095,
      "learning_rate": 4.975145730097364e-06,
      "loss": 0.30403666496276854,
      "num_input_tokens_seen": 244010048,
      "step": 14900,
      "train_runtime": 28687.9158,
      "train_tokens_per_second": 8505.674
    },
    {
      "epoch": 1.0035685429571775,
      "grad_norm": 0.989095418911667,
      "learning_rate": 4.97250168420758e-06,
      "loss": 0.29598116874694824,
      "num_input_tokens_seen": 244091968,
      "step": 14905,
      "train_runtime": 28697.5871,
      "train_tokens_per_second": 8505.662
    },
    {
      "epoch": 1.0039051979531377,
      "grad_norm": 1.1553768834620386,
      "learning_rate": 4.96985764600761e-06,
      "loss": 0.29841980934143064,
      "num_input_tokens_seen": 244173888,
      "step": 14910,
      "train_runtime": 28707.2266,
      "train_tokens_per_second": 8505.659
    },
    {
      "epoch": 1.0042418529490977,
      "grad_norm": 1.068535152969411,
      "learning_rate": 4.967213616236855e-06,
      "loss": 0.30892395973205566,
      "num_input_tokens_seen": 244255768,
      "step": 14915,
      "train_runtime": 28717.4801,
      "train_tokens_per_second": 8505.474
    },
    {
      "epoch": 1.004578507945058,
      "grad_norm": 1.1162036034525482,
      "learning_rate": 4.964569595634706e-06,
      "loss": 0.3292184114456177,
      "num_input_tokens_seen": 244337688,
      "step": 14920,
      "train_runtime": 28727.0912,
      "train_tokens_per_second": 8505.48
    },
    {
      "epoch": 1.004915162941018,
      "grad_norm": 0.989724502984292,
      "learning_rate": 4.961925584940559e-06,
      "loss": 0.32766008377075195,
      "num_input_tokens_seen": 244419608,
      "step": 14925,
      "train_runtime": 28736.6244,
      "train_tokens_per_second": 8505.509
    },
    {
      "epoch": 1.0052518179369783,
      "grad_norm": 0.9568126676179121,
      "learning_rate": 4.959281584893799e-06,
      "loss": 0.2724815607070923,
      "num_input_tokens_seen": 244501528,
      "step": 14930,
      "train_runtime": 28746.2849,
      "train_tokens_per_second": 8505.5
    },
    {
      "epoch": 1.0055884729329383,
      "grad_norm": 1.1379119594676175,
      "learning_rate": 4.956637596233814e-06,
      "loss": 0.29915552139282225,
      "num_input_tokens_seen": 244583448,
      "step": 14935,
      "train_runtime": 28756.0024,
      "train_tokens_per_second": 8505.475
    },
    {
      "epoch": 1.0059251279288985,
      "grad_norm": 1.110112634591764,
      "learning_rate": 4.953993619699986e-06,
      "loss": 0.3051766395568848,
      "num_input_tokens_seen": 244665368,
      "step": 14940,
      "train_runtime": 28765.7038,
      "train_tokens_per_second": 8505.454
    },
    {
      "epoch": 1.0062617829248586,
      "grad_norm": 1.0647939450552517,
      "learning_rate": 4.951349656031698e-06,
      "loss": 0.31187744140625,
      "num_input_tokens_seen": 244747288,
      "step": 14945,
      "train_runtime": 28775.3575,
      "train_tokens_per_second": 8505.447
    },
    {
      "epoch": 1.0065984379208188,
      "grad_norm": 1.1642560877862913,
      "learning_rate": 4.948705705968322e-06,
      "loss": 0.32361111640930174,
      "num_input_tokens_seen": 244829208,
      "step": 14950,
      "train_runtime": 28785.0629,
      "train_tokens_per_second": 8505.425
    },
    {
      "epoch": 1.0069350929167789,
      "grad_norm": 1.0126562277406892,
      "learning_rate": 4.946061770249233e-06,
      "loss": 0.28750219345092776,
      "num_input_tokens_seen": 244911128,
      "step": 14955,
      "train_runtime": 28794.8514,
      "train_tokens_per_second": 8505.379
    },
    {
      "epoch": 1.007271747912739,
      "grad_norm": 1.1524614970152902,
      "learning_rate": 4.9434178496137955e-06,
      "loss": 0.3647796154022217,
      "num_input_tokens_seen": 244993048,
      "step": 14960,
      "train_runtime": 28804.5463,
      "train_tokens_per_second": 8505.36
    },
    {
      "epoch": 1.0076084029086991,
      "grad_norm": 1.0382432822350334,
      "learning_rate": 4.940773944801378e-06,
      "loss": 0.29200408458709715,
      "num_input_tokens_seen": 245074968,
      "step": 14965,
      "train_runtime": 28814.2308,
      "train_tokens_per_second": 8505.345
    },
    {
      "epoch": 1.0079450579046594,
      "grad_norm": 1.1678647880491064,
      "learning_rate": 4.938130056551337e-06,
      "loss": 0.2899768352508545,
      "num_input_tokens_seen": 245156888,
      "step": 14970,
      "train_runtime": 28823.9029,
      "train_tokens_per_second": 8505.333
    },
    {
      "epoch": 1.0082817129006194,
      "grad_norm": 1.217842395757097,
      "learning_rate": 4.9354861856030305e-06,
      "loss": 0.32858591079711913,
      "num_input_tokens_seen": 245237712,
      "step": 14975,
      "train_runtime": 28833.7439,
      "train_tokens_per_second": 8505.233
    },
    {
      "epoch": 1.0086183678965797,
      "grad_norm": 1.0795063435040329,
      "learning_rate": 4.932842332695803e-06,
      "loss": 0.28708682060241697,
      "num_input_tokens_seen": 245319632,
      "step": 14980,
      "train_runtime": 28843.4482,
      "train_tokens_per_second": 8505.212
    },
    {
      "epoch": 1.0089550228925397,
      "grad_norm": 1.0993852838058127,
      "learning_rate": 4.9301984985690055e-06,
      "loss": 0.30554773807525637,
      "num_input_tokens_seen": 245401552,
      "step": 14985,
      "train_runtime": 28853.1645,
      "train_tokens_per_second": 8505.187
    },
    {
      "epoch": 1.0092916778885,
      "grad_norm": 1.0068793989680624,
      "learning_rate": 4.927554683961974e-06,
      "loss": 0.3000239372253418,
      "num_input_tokens_seen": 245483472,
      "step": 14990,
      "train_runtime": 28862.8147,
      "train_tokens_per_second": 8505.181
    },
    {
      "epoch": 1.00962833288446,
      "grad_norm": 1.050932218459084,
      "learning_rate": 4.924910889614046e-06,
      "loss": 0.2782566070556641,
      "num_input_tokens_seen": 245565392,
      "step": 14995,
      "train_runtime": 28872.368,
      "train_tokens_per_second": 8505.204
    },
    {
      "epoch": 1.0099649878804202,
      "grad_norm": 1.0659133887399925,
      "learning_rate": 4.922267116264546e-06,
      "loss": 0.28423609733581545,
      "num_input_tokens_seen": 245647312,
      "step": 15000,
      "train_runtime": 28882.0122,
      "train_tokens_per_second": 8505.201
    },
    {
      "epoch": 1.0103016428763802,
      "grad_norm": 1.1432623117146123,
      "learning_rate": 4.919623364652801e-06,
      "loss": 0.2970276832580566,
      "num_input_tokens_seen": 245729232,
      "step": 15005,
      "train_runtime": 28891.7222,
      "train_tokens_per_second": 8505.178
    },
    {
      "epoch": 1.0106382978723405,
      "grad_norm": 1.2547517231181256,
      "learning_rate": 4.916979635518124e-06,
      "loss": 0.3210721015930176,
      "num_input_tokens_seen": 245810912,
      "step": 15010,
      "train_runtime": 28902.0073,
      "train_tokens_per_second": 8504.977
    },
    {
      "epoch": 1.0109749528683005,
      "grad_norm": 1.3325602971780874,
      "learning_rate": 4.914335929599828e-06,
      "loss": 0.31528310775756835,
      "num_input_tokens_seen": 245892832,
      "step": 15015,
      "train_runtime": 28911.5359,
      "train_tokens_per_second": 8505.008
    },
    {
      "epoch": 1.0113116078642608,
      "grad_norm": 1.004781449561045,
      "learning_rate": 4.9116922476372155e-06,
      "loss": 0.28768353462219237,
      "num_input_tokens_seen": 245974752,
      "step": 15020,
      "train_runtime": 28921.2401,
      "train_tokens_per_second": 8504.986
    },
    {
      "epoch": 1.0116482628602208,
      "grad_norm": 1.1057930967202363,
      "learning_rate": 4.909048590369586e-06,
      "loss": 0.3125351905822754,
      "num_input_tokens_seen": 246056672,
      "step": 15025,
      "train_runtime": 28930.8694,
      "train_tokens_per_second": 8504.987
    },
    {
      "epoch": 1.011984917856181,
      "grad_norm": 1.1356136360008697,
      "learning_rate": 4.906404958536227e-06,
      "loss": 0.29378437995910645,
      "num_input_tokens_seen": 246138592,
      "step": 15030,
      "train_runtime": 28940.4808,
      "train_tokens_per_second": 8504.993
    },
    {
      "epoch": 1.012321572852141,
      "grad_norm": 1.1008177425044423,
      "learning_rate": 4.903761352876422e-06,
      "loss": 0.2957316875457764,
      "num_input_tokens_seen": 246220336,
      "step": 15035,
      "train_runtime": 28950.4594,
      "train_tokens_per_second": 8504.885
    },
    {
      "epoch": 1.0126582278481013,
      "grad_norm": 0.9972256027286434,
      "learning_rate": 4.9011177741294485e-06,
      "loss": 0.28055319786071775,
      "num_input_tokens_seen": 246302256,
      "step": 15040,
      "train_runtime": 28960.0694,
      "train_tokens_per_second": 8504.892
    },
    {
      "epoch": 1.0129948828440614,
      "grad_norm": 1.0865383878916963,
      "learning_rate": 4.898474223034576e-06,
      "loss": 0.3120269775390625,
      "num_input_tokens_seen": 246384176,
      "step": 15045,
      "train_runtime": 28969.7377,
      "train_tokens_per_second": 8504.881
    },
    {
      "epoch": 1.0133315378400216,
      "grad_norm": 0.9674978696319632,
      "learning_rate": 4.895830700331061e-06,
      "loss": 0.27366535663604735,
      "num_input_tokens_seen": 246466096,
      "step": 15050,
      "train_runtime": 28979.4061,
      "train_tokens_per_second": 8504.871
    },
    {
      "epoch": 1.0136681928359816,
      "grad_norm": 1.1485973117598631,
      "learning_rate": 4.893187206758158e-06,
      "loss": 0.2975769996643066,
      "num_input_tokens_seen": 246548016,
      "step": 15055,
      "train_runtime": 28989.037,
      "train_tokens_per_second": 8504.871
    },
    {
      "epoch": 1.014004847831942,
      "grad_norm": 1.1442044545315677,
      "learning_rate": 4.890543743055115e-06,
      "loss": 0.29758217334747317,
      "num_input_tokens_seen": 246629936,
      "step": 15060,
      "train_runtime": 28998.5697,
      "train_tokens_per_second": 8504.9
    },
    {
      "epoch": 1.014341502827902,
      "grad_norm": 1.0722890649479637,
      "learning_rate": 4.887900309961163e-06,
      "loss": 0.29640219211578367,
      "num_input_tokens_seen": 246711856,
      "step": 15065,
      "train_runtime": 29008.2257,
      "train_tokens_per_second": 8504.893
    },
    {
      "epoch": 1.0146781578238622,
      "grad_norm": 1.063424583300341,
      "learning_rate": 4.885256908215532e-06,
      "loss": 0.30859060287475587,
      "num_input_tokens_seen": 246793776,
      "step": 15070,
      "train_runtime": 29017.8416,
      "train_tokens_per_second": 8504.898
    },
    {
      "epoch": 1.0150148128198222,
      "grad_norm": 1.1673180590808414,
      "learning_rate": 4.88261353855744e-06,
      "loss": 0.29080021381378174,
      "num_input_tokens_seen": 246875696,
      "step": 15075,
      "train_runtime": 29027.3454,
      "train_tokens_per_second": 8504.935
    },
    {
      "epoch": 1.0153514678157824,
      "grad_norm": 1.1056883346986202,
      "learning_rate": 4.8799702017261e-06,
      "loss": 0.26958742141723635,
      "num_input_tokens_seen": 246956888,
      "step": 15080,
      "train_runtime": 29037.0576,
      "train_tokens_per_second": 8504.887
    },
    {
      "epoch": 1.0156881228117425,
      "grad_norm": 1.3451989748710833,
      "learning_rate": 4.877326898460709e-06,
      "loss": 0.32477195262908937,
      "num_input_tokens_seen": 247038808,
      "step": 15085,
      "train_runtime": 29046.5614,
      "train_tokens_per_second": 8504.924
    },
    {
      "epoch": 1.0160247778077027,
      "grad_norm": 0.8743217269510607,
      "learning_rate": 4.87468362950046e-06,
      "loss": 0.2609305143356323,
      "num_input_tokens_seen": 247120224,
      "step": 15090,
      "train_runtime": 29056.4063,
      "train_tokens_per_second": 8504.845
    },
    {
      "epoch": 1.0163614328036628,
      "grad_norm": 1.2334706263913742,
      "learning_rate": 4.872040395584533e-06,
      "loss": 0.31082754135131835,
      "num_input_tokens_seen": 247202144,
      "step": 15095,
      "train_runtime": 29065.9207,
      "train_tokens_per_second": 8504.879
    },
    {
      "epoch": 1.016698087799623,
      "grad_norm": 1.083544417122358,
      "learning_rate": 4.869397197452104e-06,
      "loss": 0.2676072597503662,
      "num_input_tokens_seen": 247284064,
      "step": 15100,
      "train_runtime": 29075.4274,
      "train_tokens_per_second": 8504.916
    },
    {
      "epoch": 1.017034742795583,
      "grad_norm": 1.012374002250196,
      "learning_rate": 4.866754035842329e-06,
      "loss": 0.28214704990386963,
      "num_input_tokens_seen": 247365984,
      "step": 15105,
      "train_runtime": 29084.9284,
      "train_tokens_per_second": 8504.954
    },
    {
      "epoch": 1.0173713977915433,
      "grad_norm": 1.0528554054757822,
      "learning_rate": 4.8641109114943666e-06,
      "loss": 0.287906551361084,
      "num_input_tokens_seen": 247447904,
      "step": 15110,
      "train_runtime": 29094.4367,
      "train_tokens_per_second": 8504.99
    },
    {
      "epoch": 1.0177080527875033,
      "grad_norm": 1.1321386917774412,
      "learning_rate": 4.861467825147352e-06,
      "loss": 0.29657688140869143,
      "num_input_tokens_seen": 247529824,
      "step": 15115,
      "train_runtime": 29103.9568,
      "train_tokens_per_second": 8505.023
    },
    {
      "epoch": 1.0180447077834636,
      "grad_norm": 1.1809549767058223,
      "learning_rate": 4.858824777540419e-06,
      "loss": 0.2824713706970215,
      "num_input_tokens_seen": 247611744,
      "step": 15120,
      "train_runtime": 29113.6395,
      "train_tokens_per_second": 8505.008
    },
    {
      "epoch": 1.0183813627794236,
      "grad_norm": 1.2605701666548623,
      "learning_rate": 4.856181769412686e-06,
      "loss": 0.31162662506103517,
      "num_input_tokens_seen": 247693664,
      "step": 15125,
      "train_runtime": 29123.2714,
      "train_tokens_per_second": 8505.008
    },
    {
      "epoch": 1.0187180177753838,
      "grad_norm": 0.9482861979368643,
      "learning_rate": 4.8535388015032644e-06,
      "loss": 0.28868398666381834,
      "num_input_tokens_seen": 247775584,
      "step": 15130,
      "train_runtime": 29132.8969,
      "train_tokens_per_second": 8505.01
    },
    {
      "epoch": 1.0190546727713439,
      "grad_norm": 1.1179860552458112,
      "learning_rate": 4.850895874551248e-06,
      "loss": 0.30797505378723145,
      "num_input_tokens_seen": 247857504,
      "step": 15135,
      "train_runtime": 29142.5668,
      "train_tokens_per_second": 8504.999
    },
    {
      "epoch": 1.0193913277673041,
      "grad_norm": 1.1869422549867645,
      "learning_rate": 4.848252989295726e-06,
      "loss": 0.3119257926940918,
      "num_input_tokens_seen": 247939424,
      "step": 15140,
      "train_runtime": 29152.2241,
      "train_tokens_per_second": 8504.992
    },
    {
      "epoch": 1.0197279827632642,
      "grad_norm": 0.9585773373125603,
      "learning_rate": 4.845610146475771e-06,
      "loss": 0.2820000171661377,
      "num_input_tokens_seen": 248021344,
      "step": 15145,
      "train_runtime": 29161.8881,
      "train_tokens_per_second": 8504.982
    },
    {
      "epoch": 1.0200646377592244,
      "grad_norm": 1.0280795451022549,
      "learning_rate": 4.842967346830447e-06,
      "loss": 0.27739155292510986,
      "num_input_tokens_seen": 248103264,
      "step": 15150,
      "train_runtime": 29171.5255,
      "train_tokens_per_second": 8504.981
    },
    {
      "epoch": 1.0204012927551844,
      "grad_norm": 1.1561428372444162,
      "learning_rate": 4.840324591098803e-06,
      "loss": 0.2948404312133789,
      "num_input_tokens_seen": 248185184,
      "step": 15155,
      "train_runtime": 29181.0444,
      "train_tokens_per_second": 8505.014
    },
    {
      "epoch": 1.0207379477511447,
      "grad_norm": 0.9188367793194586,
      "learning_rate": 4.83768188001988e-06,
      "loss": 0.2959731101989746,
      "num_input_tokens_seen": 248267104,
      "step": 15160,
      "train_runtime": 29190.7383,
      "train_tokens_per_second": 8504.996
    },
    {
      "epoch": 1.0210746027471047,
      "grad_norm": 0.9782571922297375,
      "learning_rate": 4.835039214332699e-06,
      "loss": 0.260086727142334,
      "num_input_tokens_seen": 248349024,
      "step": 15165,
      "train_runtime": 29200.4096,
      "train_tokens_per_second": 8504.984
    },
    {
      "epoch": 1.021411257743065,
      "grad_norm": 1.1028824310546004,
      "learning_rate": 4.832396594776277e-06,
      "loss": 0.28785114288330077,
      "num_input_tokens_seen": 248430944,
      "step": 15170,
      "train_runtime": 29210.0404,
      "train_tokens_per_second": 8504.985
    },
    {
      "epoch": 1.021747912739025,
      "grad_norm": 0.972720889241645,
      "learning_rate": 4.82975402208961e-06,
      "loss": 0.3025206089019775,
      "num_input_tokens_seen": 248512864,
      "step": 15175,
      "train_runtime": 29219.6882,
      "train_tokens_per_second": 8504.98
    },
    {
      "epoch": 1.0220845677349852,
      "grad_norm": 1.0479126393486191,
      "learning_rate": 4.827111497011692e-06,
      "loss": 0.2695987939834595,
      "num_input_tokens_seen": 248594784,
      "step": 15180,
      "train_runtime": 29229.1843,
      "train_tokens_per_second": 8505.02
    },
    {
      "epoch": 1.0224212227309453,
      "grad_norm": 1.1581153606039916,
      "learning_rate": 4.82446902028149e-06,
      "loss": 0.29395418167114257,
      "num_input_tokens_seen": 248676704,
      "step": 15185,
      "train_runtime": 29238.6866,
      "train_tokens_per_second": 8505.057
    },
    {
      "epoch": 1.0227578777269055,
      "grad_norm": 1.4828752610848457,
      "learning_rate": 4.821826592637968e-06,
      "loss": 0.3156465768814087,
      "num_input_tokens_seen": 248758624,
      "step": 15190,
      "train_runtime": 29248.3185,
      "train_tokens_per_second": 8505.057
    },
    {
      "epoch": 1.0230945327228655,
      "grad_norm": 0.9418844275087901,
      "learning_rate": 4.8191842148200694e-06,
      "loss": 0.29228479862213136,
      "num_input_tokens_seen": 248840544,
      "step": 15195,
      "train_runtime": 29257.9671,
      "train_tokens_per_second": 8505.052
    },
    {
      "epoch": 1.0234311877188258,
      "grad_norm": 1.0595499690702865,
      "learning_rate": 4.8165418875667315e-06,
      "loss": 0.28883774280548097,
      "num_input_tokens_seen": 248922464,
      "step": 15200,
      "train_runtime": 29267.665,
      "train_tokens_per_second": 8505.033
    },
    {
      "epoch": 1.0237678427147858,
      "grad_norm": 1.0528583250424988,
      "learning_rate": 4.813899611616867e-06,
      "loss": 0.30303876399993895,
      "num_input_tokens_seen": 249004384,
      "step": 15205,
      "train_runtime": 29277.3811,
      "train_tokens_per_second": 8505.009
    },
    {
      "epoch": 1.024104497710746,
      "grad_norm": 0.9977980414697997,
      "learning_rate": 4.8112573877093855e-06,
      "loss": 0.2958951473236084,
      "num_input_tokens_seen": 249086304,
      "step": 15210,
      "train_runtime": 29287.0705,
      "train_tokens_per_second": 8504.992
    },
    {
      "epoch": 1.024441152706706,
      "grad_norm": 1.1174999836343384,
      "learning_rate": 4.808615216583171e-06,
      "loss": 0.31137025356292725,
      "num_input_tokens_seen": 249168224,
      "step": 15215,
      "train_runtime": 29296.6902,
      "train_tokens_per_second": 8504.996
    },
    {
      "epoch": 1.0247778077026664,
      "grad_norm": 1.4627181707805645,
      "learning_rate": 4.805973098977102e-06,
      "loss": 0.31546597480773925,
      "num_input_tokens_seen": 249250144,
      "step": 15220,
      "train_runtime": 29306.3831,
      "train_tokens_per_second": 8504.978
    },
    {
      "epoch": 1.0251144626986264,
      "grad_norm": 0.9584736367068487,
      "learning_rate": 4.803331035630037e-06,
      "loss": 0.3137208938598633,
      "num_input_tokens_seen": 249332064,
      "step": 15225,
      "train_runtime": 29316.1146,
      "train_tokens_per_second": 8504.949
    },
    {
      "epoch": 1.0254511176945866,
      "grad_norm": 1.2137057675377336,
      "learning_rate": 4.800689027280822e-06,
      "loss": 0.2843333721160889,
      "num_input_tokens_seen": 249413984,
      "step": 15230,
      "train_runtime": 29325.6172,
      "train_tokens_per_second": 8504.987
    },
    {
      "epoch": 1.0257877726905467,
      "grad_norm": 0.9505375969639557,
      "learning_rate": 4.798047074668283e-06,
      "loss": 0.2921628952026367,
      "num_input_tokens_seen": 249495904,
      "step": 15235,
      "train_runtime": 29335.112,
      "train_tokens_per_second": 8505.026
    },
    {
      "epoch": 1.026124427686507,
      "grad_norm": 0.7495005078577603,
      "learning_rate": 4.795405178531236e-06,
      "loss": 0.2692469835281372,
      "num_input_tokens_seen": 249577304,
      "step": 15240,
      "train_runtime": 29344.6054,
      "train_tokens_per_second": 8505.049
    },
    {
      "epoch": 1.026461082682467,
      "grad_norm": 1.0868312293038278,
      "learning_rate": 4.792763339608478e-06,
      "loss": 0.29685709476470945,
      "num_input_tokens_seen": 249659224,
      "step": 15245,
      "train_runtime": 29354.1049,
      "train_tokens_per_second": 8505.087
    },
    {
      "epoch": 1.0267977376784272,
      "grad_norm": 1.0225927869013773,
      "learning_rate": 4.790121558638793e-06,
      "loss": 0.2693136930465698,
      "num_input_tokens_seen": 249741144,
      "step": 15250,
      "train_runtime": 29363.6007,
      "train_tokens_per_second": 8505.127
    },
    {
      "epoch": 1.0271343926743872,
      "grad_norm": 0.9972805192375636,
      "learning_rate": 4.787479836360944e-06,
      "loss": 0.30147316455841067,
      "num_input_tokens_seen": 249822792,
      "step": 15255,
      "train_runtime": 29373.5042,
      "train_tokens_per_second": 8505.039
    },
    {
      "epoch": 1.0274710476703475,
      "grad_norm": 1.1302534758069,
      "learning_rate": 4.784838173513682e-06,
      "loss": 0.29442734718322755,
      "num_input_tokens_seen": 249904712,
      "step": 15260,
      "train_runtime": 29383.0132,
      "train_tokens_per_second": 8505.074
    },
    {
      "epoch": 1.0278077026663075,
      "grad_norm": 1.1750824911772069,
      "learning_rate": 4.782196570835737e-06,
      "loss": 0.27294206619262695,
      "num_input_tokens_seen": 249986632,
      "step": 15265,
      "train_runtime": 29392.6919,
      "train_tokens_per_second": 8505.061
    },
    {
      "epoch": 1.0281443576622677,
      "grad_norm": 1.0456960169621574,
      "learning_rate": 4.779555029065827e-06,
      "loss": 0.2788179874420166,
      "num_input_tokens_seen": 250068552,
      "step": 15270,
      "train_runtime": 29402.3339,
      "train_tokens_per_second": 8505.058
    },
    {
      "epoch": 1.0284810126582278,
      "grad_norm": 1.1991312692431655,
      "learning_rate": 4.776913548942649e-06,
      "loss": 0.28983402252197266,
      "num_input_tokens_seen": 250150472,
      "step": 15275,
      "train_runtime": 29412.0259,
      "train_tokens_per_second": 8505.041
    },
    {
      "epoch": 1.028817667654188,
      "grad_norm": 1.2628800660408306,
      "learning_rate": 4.774272131204889e-06,
      "loss": 0.27174346446990966,
      "num_input_tokens_seen": 250232392,
      "step": 15280,
      "train_runtime": 29421.7073,
      "train_tokens_per_second": 8505.026
    },
    {
      "epoch": 1.029154322650148,
      "grad_norm": 1.1099146481805142,
      "learning_rate": 4.7716307765912045e-06,
      "loss": 0.30734236240386964,
      "num_input_tokens_seen": 250314312,
      "step": 15285,
      "train_runtime": 29431.2093,
      "train_tokens_per_second": 8505.064
    },
    {
      "epoch": 1.0294909776461083,
      "grad_norm": 1.1475310992334833,
      "learning_rate": 4.768989485840246e-06,
      "loss": 0.310355019569397,
      "num_input_tokens_seen": 250396232,
      "step": 15290,
      "train_runtime": 29440.8183,
      "train_tokens_per_second": 8505.07
    },
    {
      "epoch": 1.0298276326420683,
      "grad_norm": 1.134154529571949,
      "learning_rate": 4.766348259690641e-06,
      "loss": 0.3194678068161011,
      "num_input_tokens_seen": 250478152,
      "step": 15295,
      "train_runtime": 29450.5316,
      "train_tokens_per_second": 8505.047
    },
    {
      "epoch": 1.0301642876380286,
      "grad_norm": 1.1798483988597663,
      "learning_rate": 4.763707098881002e-06,
      "loss": 0.3203343391418457,
      "num_input_tokens_seen": 250560072,
      "step": 15300,
      "train_runtime": 29460.1551,
      "train_tokens_per_second": 8505.049
    },
    {
      "epoch": 1.0305009426339886,
      "grad_norm": 1.123066328589066,
      "learning_rate": 4.761066004149917e-06,
      "loss": 0.3033395051956177,
      "num_input_tokens_seen": 250641992,
      "step": 15305,
      "train_runtime": 29469.7141,
      "train_tokens_per_second": 8505.07
    },
    {
      "epoch": 1.0308375976299489,
      "grad_norm": 1.2893206839683307,
      "learning_rate": 4.758424976235964e-06,
      "loss": 0.3165580749511719,
      "num_input_tokens_seen": 250723912,
      "step": 15310,
      "train_runtime": 29479.282,
      "train_tokens_per_second": 8505.089
    },
    {
      "epoch": 1.031174252625909,
      "grad_norm": 1.108699789822247,
      "learning_rate": 4.755784015877694e-06,
      "loss": 0.3259037733078003,
      "num_input_tokens_seen": 250805832,
      "step": 15315,
      "train_runtime": 29488.9166,
      "train_tokens_per_second": 8505.088
    },
    {
      "epoch": 1.0315109076218691,
      "grad_norm": 1.0321406422060098,
      "learning_rate": 4.753143123813646e-06,
      "loss": 0.3084686756134033,
      "num_input_tokens_seen": 250887752,
      "step": 15320,
      "train_runtime": 29498.4706,
      "train_tokens_per_second": 8505.111
    },
    {
      "epoch": 1.0318475626178292,
      "grad_norm": 1.130327396487325,
      "learning_rate": 4.7505023007823365e-06,
      "loss": 0.2904776096343994,
      "num_input_tokens_seen": 250969672,
      "step": 15325,
      "train_runtime": 29508.057,
      "train_tokens_per_second": 8505.124
    },
    {
      "epoch": 1.0321842176137894,
      "grad_norm": 1.009815546885119,
      "learning_rate": 4.747861547522264e-06,
      "loss": 0.2995340347290039,
      "num_input_tokens_seen": 251051368,
      "step": 15330,
      "train_runtime": 29517.7972,
      "train_tokens_per_second": 8505.085
    },
    {
      "epoch": 1.0325208726097495,
      "grad_norm": 1.007061653984028,
      "learning_rate": 4.745220864771904e-06,
      "loss": 0.2754704713821411,
      "num_input_tokens_seen": 251133288,
      "step": 15335,
      "train_runtime": 29527.5787,
      "train_tokens_per_second": 8505.042
    },
    {
      "epoch": 1.0328575276057097,
      "grad_norm": 0.9974584269341483,
      "learning_rate": 4.742580253269718e-06,
      "loss": 0.28981566429138184,
      "num_input_tokens_seen": 251215208,
      "step": 15340,
      "train_runtime": 29537.0836,
      "train_tokens_per_second": 8505.078
    },
    {
      "epoch": 1.0331941826016697,
      "grad_norm": 0.8421762120143823,
      "learning_rate": 4.739939713754141e-06,
      "loss": 0.2644367218017578,
      "num_input_tokens_seen": 251296568,
      "step": 15345,
      "train_runtime": 29547.0441,
      "train_tokens_per_second": 8504.965
    },
    {
      "epoch": 1.03353083759763,
      "grad_norm": 1.1313588348851995,
      "learning_rate": 4.737299246963597e-06,
      "loss": 0.27610068321228026,
      "num_input_tokens_seen": 251378488,
      "step": 15350,
      "train_runtime": 29556.6939,
      "train_tokens_per_second": 8504.96
    },
    {
      "epoch": 1.03386749259359,
      "grad_norm": 1.041016768695226,
      "learning_rate": 4.734658853636479e-06,
      "loss": 0.2754704475402832,
      "num_input_tokens_seen": 251460408,
      "step": 15355,
      "train_runtime": 29566.1929,
      "train_tokens_per_second": 8504.998
    },
    {
      "epoch": 1.0342041475895503,
      "grad_norm": 1.0497269868379397,
      "learning_rate": 4.732018534511167e-06,
      "loss": 0.3103634834289551,
      "num_input_tokens_seen": 251542208,
      "step": 15360,
      "train_runtime": 29576.5148,
      "train_tokens_per_second": 8504.795
    },
    {
      "epoch": 1.0345408025855103,
      "grad_norm": 1.0179037749117918,
      "learning_rate": 4.729378290326017e-06,
      "loss": 0.2703082084655762,
      "num_input_tokens_seen": 251624128,
      "step": 15365,
      "train_runtime": 29586.024,
      "train_tokens_per_second": 8504.831
    },
    {
      "epoch": 1.0348774575814705,
      "grad_norm": 0.9192898432047028,
      "learning_rate": 4.726738121819365e-06,
      "loss": 0.2728667974472046,
      "num_input_tokens_seen": 251706048,
      "step": 15370,
      "train_runtime": 29595.5301,
      "train_tokens_per_second": 8504.867
    },
    {
      "epoch": 1.0352141125774306,
      "grad_norm": 1.1791978058330848,
      "learning_rate": 4.7240980297295255e-06,
      "loss": 0.25916118621826173,
      "num_input_tokens_seen": 251787888,
      "step": 15375,
      "train_runtime": 29605.752,
      "train_tokens_per_second": 8504.695
    },
    {
      "epoch": 1.0355507675733908,
      "grad_norm": 0.9563875819141544,
      "learning_rate": 4.721458014794795e-06,
      "loss": 0.3009016990661621,
      "num_input_tokens_seen": 251869808,
      "step": 15380,
      "train_runtime": 29615.4163,
      "train_tokens_per_second": 8504.686
    },
    {
      "epoch": 1.0358874225693508,
      "grad_norm": 1.292647137219822,
      "learning_rate": 4.718818077753439e-06,
      "loss": 0.34977548122406005,
      "num_input_tokens_seen": 251951728,
      "step": 15385,
      "train_runtime": 29624.9328,
      "train_tokens_per_second": 8504.719
    },
    {
      "epoch": 1.036224077565311,
      "grad_norm": 1.023141926652569,
      "learning_rate": 4.716178219343712e-06,
      "loss": 0.29911463260650634,
      "num_input_tokens_seen": 252033648,
      "step": 15390,
      "train_runtime": 29634.4368,
      "train_tokens_per_second": 8504.756
    },
    {
      "epoch": 1.0365607325612711,
      "grad_norm": 1.021232139296396,
      "learning_rate": 4.713538440303841e-06,
      "loss": 0.29764699935913086,
      "num_input_tokens_seen": 252115568,
      "step": 15395,
      "train_runtime": 29643.9436,
      "train_tokens_per_second": 8504.792
    },
    {
      "epoch": 1.0368973875572314,
      "grad_norm": 1.158168889468369,
      "learning_rate": 4.7108987413720335e-06,
      "loss": 0.29134724140167234,
      "num_input_tokens_seen": 252197488,
      "step": 15400,
      "train_runtime": 29653.4404,
      "train_tokens_per_second": 8504.831
    },
    {
      "epoch": 1.0372340425531914,
      "grad_norm": 1.06809931031655,
      "learning_rate": 4.708259123286469e-06,
      "loss": 0.30393381118774415,
      "num_input_tokens_seen": 252279408,
      "step": 15405,
      "train_runtime": 29663.472,
      "train_tokens_per_second": 8504.716
    },
    {
      "epoch": 1.0375706975491517,
      "grad_norm": 1.1564610385217673,
      "learning_rate": 4.7056195867853105e-06,
      "loss": 0.2978277921676636,
      "num_input_tokens_seen": 252361328,
      "step": 15410,
      "train_runtime": 29672.9716,
      "train_tokens_per_second": 8504.754
    },
    {
      "epoch": 1.0379073525451117,
      "grad_norm": 1.099541520538992,
      "learning_rate": 4.702980132606696e-06,
      "loss": 0.30000350475311277,
      "num_input_tokens_seen": 252443248,
      "step": 15415,
      "train_runtime": 29682.4667,
      "train_tokens_per_second": 8504.793
    },
    {
      "epoch": 1.038244007541072,
      "grad_norm": 1.1147425860287172,
      "learning_rate": 4.700340761488741e-06,
      "loss": 0.2778669834136963,
      "num_input_tokens_seen": 252525168,
      "step": 15420,
      "train_runtime": 29691.9604,
      "train_tokens_per_second": 8504.833
    },
    {
      "epoch": 1.038580662537032,
      "grad_norm": 1.0404521582334674,
      "learning_rate": 4.6977014741695345e-06,
      "loss": 0.27747135162353515,
      "num_input_tokens_seen": 252607088,
      "step": 15425,
      "train_runtime": 29701.4566,
      "train_tokens_per_second": 8504.872
    },
    {
      "epoch": 1.0389173175329922,
      "grad_norm": 1.0702788012705449,
      "learning_rate": 4.69506227138715e-06,
      "loss": 0.3099741220474243,
      "num_input_tokens_seen": 252689008,
      "step": 15430,
      "train_runtime": 29710.9575,
      "train_tokens_per_second": 8504.91
    },
    {
      "epoch": 1.0392539725289522,
      "grad_norm": 1.171807949455024,
      "learning_rate": 4.6924231538796265e-06,
      "loss": 0.3172988653182983,
      "num_input_tokens_seen": 252770928,
      "step": 15435,
      "train_runtime": 29720.5609,
      "train_tokens_per_second": 8504.918
    },
    {
      "epoch": 1.0395906275249125,
      "grad_norm": 1.2061329333897757,
      "learning_rate": 4.6897841223849885e-06,
      "loss": 0.2749769687652588,
      "num_input_tokens_seen": 252852848,
      "step": 15440,
      "train_runtime": 29730.2285,
      "train_tokens_per_second": 8504.908
    },
    {
      "epoch": 1.0399272825208725,
      "grad_norm": 1.0982254952618897,
      "learning_rate": 4.68714517764123e-06,
      "loss": 0.2903334856033325,
      "num_input_tokens_seen": 252934768,
      "step": 15445,
      "train_runtime": 29740.0358,
      "train_tokens_per_second": 8504.858
    },
    {
      "epoch": 1.0402639375168328,
      "grad_norm": 0.8734979351165268,
      "learning_rate": 4.684506320386327e-06,
      "loss": 0.26625823974609375,
      "num_input_tokens_seen": 253016688,
      "step": 15450,
      "train_runtime": 29749.5702,
      "train_tokens_per_second": 8504.885
    },
    {
      "epoch": 1.0406005925127928,
      "grad_norm": 0.9253552853827903,
      "learning_rate": 4.6818675513582235e-06,
      "loss": 0.2852307319641113,
      "num_input_tokens_seen": 253098608,
      "step": 15455,
      "train_runtime": 29759.284,
      "train_tokens_per_second": 8504.862
    },
    {
      "epoch": 1.040937247508753,
      "grad_norm": 1.2092892081502202,
      "learning_rate": 4.679228871294846e-06,
      "loss": 0.27957653999328613,
      "num_input_tokens_seen": 253180528,
      "step": 15460,
      "train_runtime": 29768.9651,
      "train_tokens_per_second": 8504.848
    },
    {
      "epoch": 1.041273902504713,
      "grad_norm": 0.940637288287169,
      "learning_rate": 4.67659028093409e-06,
      "loss": 0.2736029148101807,
      "num_input_tokens_seen": 253262448,
      "step": 15465,
      "train_runtime": 29778.6706,
      "train_tokens_per_second": 8504.827
    },
    {
      "epoch": 1.0416105575006733,
      "grad_norm": 1.3145637093160425,
      "learning_rate": 4.673951781013834e-06,
      "loss": 0.3238328456878662,
      "num_input_tokens_seen": 253344368,
      "step": 15470,
      "train_runtime": 29788.335,
      "train_tokens_per_second": 8504.818
    },
    {
      "epoch": 1.0419472124966334,
      "grad_norm": 1.0574862615529035,
      "learning_rate": 4.67131337227192e-06,
      "loss": 0.2913242816925049,
      "num_input_tokens_seen": 253426288,
      "step": 15475,
      "train_runtime": 29798.0206,
      "train_tokens_per_second": 8504.803
    },
    {
      "epoch": 1.0422838674925936,
      "grad_norm": 1.0055604983113728,
      "learning_rate": 4.668675055446175e-06,
      "loss": 0.2834345579147339,
      "num_input_tokens_seen": 253508208,
      "step": 15480,
      "train_runtime": 29807.6885,
      "train_tokens_per_second": 8504.793
    },
    {
      "epoch": 1.0426205224885536,
      "grad_norm": 1.171100459441688,
      "learning_rate": 4.666036831274392e-06,
      "loss": 0.27343761920928955,
      "num_input_tokens_seen": 253590128,
      "step": 15485,
      "train_runtime": 29817.2118,
      "train_tokens_per_second": 8504.824
    },
    {
      "epoch": 1.0429571774845139,
      "grad_norm": 1.2436976190722648,
      "learning_rate": 4.663398700494346e-06,
      "loss": 0.31651260852813723,
      "num_input_tokens_seen": 253672048,
      "step": 15490,
      "train_runtime": 29826.8346,
      "train_tokens_per_second": 8504.826
    },
    {
      "epoch": 1.043293832480474,
      "grad_norm": 1.2725818624939422,
      "learning_rate": 4.66076066384378e-06,
      "loss": 0.330286431312561,
      "num_input_tokens_seen": 253753968,
      "step": 15495,
      "train_runtime": 29836.489,
      "train_tokens_per_second": 8504.82
    },
    {
      "epoch": 1.0436304874764342,
      "grad_norm": 0.7627968382343293,
      "learning_rate": 4.658122722060413e-06,
      "loss": 0.28867030143737793,
      "num_input_tokens_seen": 253835888,
      "step": 15500,
      "train_runtime": 29845.9983,
      "train_tokens_per_second": 8504.855
    },
    {
      "epoch": 1.0439671424723942,
      "grad_norm": 1.1255644044012434,
      "learning_rate": 4.655484875881935e-06,
      "loss": 0.3260690689086914,
      "num_input_tokens_seen": 253917808,
      "step": 15505,
      "train_runtime": 29855.5184,
      "train_tokens_per_second": 8504.887
    },
    {
      "epoch": 1.0443037974683544,
      "grad_norm": 0.978217513425425,
      "learning_rate": 4.652847126046014e-06,
      "loss": 0.2981850624084473,
      "num_input_tokens_seen": 253999696,
      "step": 15510,
      "train_runtime": 29865.1927,
      "train_tokens_per_second": 8504.874
    },
    {
      "epoch": 1.0446404524643145,
      "grad_norm": 1.1541816920537986,
      "learning_rate": 4.650209473290286e-06,
      "loss": 0.3166623592376709,
      "num_input_tokens_seen": 254081616,
      "step": 15515,
      "train_runtime": 29874.7937,
      "train_tokens_per_second": 8504.883
    },
    {
      "epoch": 1.0449771074602747,
      "grad_norm": 1.1396888456321765,
      "learning_rate": 4.647571918352366e-06,
      "loss": 0.30245540142059324,
      "num_input_tokens_seen": 254163536,
      "step": 15520,
      "train_runtime": 29884.2953,
      "train_tokens_per_second": 8504.92
    },
    {
      "epoch": 1.0453137624562348,
      "grad_norm": 1.2153617189149317,
      "learning_rate": 4.644934461969833e-06,
      "loss": 0.3096425533294678,
      "num_input_tokens_seen": 254245456,
      "step": 15525,
      "train_runtime": 29893.939,
      "train_tokens_per_second": 8504.917
    },
    {
      "epoch": 1.045650417452195,
      "grad_norm": 0.9329395093388921,
      "learning_rate": 4.642297104880247e-06,
      "loss": 0.26487011909484864,
      "num_input_tokens_seen": 254327376,
      "step": 15530,
      "train_runtime": 29903.5932,
      "train_tokens_per_second": 8504.91
    },
    {
      "epoch": 1.045987072448155,
      "grad_norm": 1.0726034345320663,
      "learning_rate": 4.639659847821133e-06,
      "loss": 0.2815439224243164,
      "num_input_tokens_seen": 254409296,
      "step": 15535,
      "train_runtime": 29913.2512,
      "train_tokens_per_second": 8504.903
    },
    {
      "epoch": 1.0463237274441153,
      "grad_norm": 1.2744734047304656,
      "learning_rate": 4.6370226915299945e-06,
      "loss": 0.31378300189971925,
      "num_input_tokens_seen": 254491040,
      "step": 15540,
      "train_runtime": 29923.4156,
      "train_tokens_per_second": 8504.746
    },
    {
      "epoch": 1.0466603824400753,
      "grad_norm": 1.0979054124001622,
      "learning_rate": 4.634385636744302e-06,
      "loss": 0.2885329246520996,
      "num_input_tokens_seen": 254572960,
      "step": 15545,
      "train_runtime": 29933.0675,
      "train_tokens_per_second": 8504.74
    },
    {
      "epoch": 1.0469970374360356,
      "grad_norm": 1.1978039762287405,
      "learning_rate": 4.631748684201503e-06,
      "loss": 0.3285683155059814,
      "num_input_tokens_seen": 254654848,
      "step": 15550,
      "train_runtime": 29943.2804,
      "train_tokens_per_second": 8504.574
    },
    {
      "epoch": 1.0473336924319956,
      "grad_norm": 1.004416895724003,
      "learning_rate": 4.629111834639008e-06,
      "loss": 0.27324185371398924,
      "num_input_tokens_seen": 254736768,
      "step": 15555,
      "train_runtime": 29952.8901,
      "train_tokens_per_second": 8504.581
    },
    {
      "epoch": 1.0476703474279558,
      "grad_norm": 1.1341823535524085,
      "learning_rate": 4.626475088794206e-06,
      "loss": 0.28508682250976564,
      "num_input_tokens_seen": 254818688,
      "step": 15560,
      "train_runtime": 29962.3976,
      "train_tokens_per_second": 8504.616
    },
    {
      "epoch": 1.0480070024239159,
      "grad_norm": 1.2499836993800653,
      "learning_rate": 4.623838447404454e-06,
      "loss": 0.3165287971496582,
      "num_input_tokens_seen": 254900608,
      "step": 15565,
      "train_runtime": 29971.9309,
      "train_tokens_per_second": 8504.644
    },
    {
      "epoch": 1.0483436574198761,
      "grad_norm": 1.0525297151628423,
      "learning_rate": 4.621201911207082e-06,
      "loss": 0.3139209270477295,
      "num_input_tokens_seen": 254982528,
      "step": 15570,
      "train_runtime": 29981.6499,
      "train_tokens_per_second": 8504.62
    },
    {
      "epoch": 1.0486803124158361,
      "grad_norm": 0.9080733807796952,
      "learning_rate": 4.618565480939387e-06,
      "loss": 0.3166510105133057,
      "num_input_tokens_seen": 255064448,
      "step": 15575,
      "train_runtime": 29991.3148,
      "train_tokens_per_second": 8504.61
    },
    {
      "epoch": 1.0490169674117964,
      "grad_norm": 1.1179994943894527,
      "learning_rate": 4.6159291573386395e-06,
      "loss": 0.3052693843841553,
      "num_input_tokens_seen": 255146368,
      "step": 15580,
      "train_runtime": 30000.9672,
      "train_tokens_per_second": 8504.605
    },
    {
      "epoch": 1.0493536224077564,
      "grad_norm": 1.185990356320347,
      "learning_rate": 4.613292941142077e-06,
      "loss": 0.31412539482116697,
      "num_input_tokens_seen": 255227896,
      "step": 15585,
      "train_runtime": 30010.9126,
      "train_tokens_per_second": 8504.503
    },
    {
      "epoch": 1.0496902774037167,
      "grad_norm": 1.1297577072642342,
      "learning_rate": 4.610656833086912e-06,
      "loss": 0.2958220481872559,
      "num_input_tokens_seen": 255309808,
      "step": 15590,
      "train_runtime": 30020.4292,
      "train_tokens_per_second": 8504.536
    },
    {
      "epoch": 1.0500269323996767,
      "grad_norm": 1.129885075652163,
      "learning_rate": 4.60802083391032e-06,
      "loss": 0.3271326065063477,
      "num_input_tokens_seen": 255391728,
      "step": 15595,
      "train_runtime": 30030.0016,
      "train_tokens_per_second": 8504.553
    },
    {
      "epoch": 1.050363587395637,
      "grad_norm": 1.143996733066,
      "learning_rate": 4.605384944349454e-06,
      "loss": 0.27104547023773196,
      "num_input_tokens_seen": 255473648,
      "step": 15600,
      "train_runtime": 30039.5799,
      "train_tokens_per_second": 8504.568
    },
    {
      "epoch": 1.050700242391597,
      "grad_norm": 1.0267956092602195,
      "learning_rate": 4.602749165141429e-06,
      "loss": 0.26870384216308596,
      "num_input_tokens_seen": 255555568,
      "step": 15605,
      "train_runtime": 30049.2096,
      "train_tokens_per_second": 8504.569
    },
    {
      "epoch": 1.0510368973875572,
      "grad_norm": 1.1100636754390243,
      "learning_rate": 4.600113497023333e-06,
      "loss": 0.3137278318405151,
      "num_input_tokens_seen": 255637488,
      "step": 15610,
      "train_runtime": 30058.7489,
      "train_tokens_per_second": 8504.595
    },
    {
      "epoch": 1.0513735523835175,
      "grad_norm": 0.9917163114424822,
      "learning_rate": 4.597477940732222e-06,
      "loss": 0.3123222827911377,
      "num_input_tokens_seen": 255719408,
      "step": 15615,
      "train_runtime": 30068.4006,
      "train_tokens_per_second": 8504.59
    },
    {
      "epoch": 1.0517102073794775,
      "grad_norm": 1.1250234619995447,
      "learning_rate": 4.594842497005124e-06,
      "loss": 0.30327744483947755,
      "num_input_tokens_seen": 255801328,
      "step": 15620,
      "train_runtime": 30077.9018,
      "train_tokens_per_second": 8504.627
    },
    {
      "epoch": 1.0520468623754375,
      "grad_norm": 0.9330324483672136,
      "learning_rate": 4.5922071665790264e-06,
      "loss": 0.29355247020721437,
      "num_input_tokens_seen": 255883248,
      "step": 15625,
      "train_runtime": 30087.3999,
      "train_tokens_per_second": 8504.665
    },
    {
      "epoch": 1.0523835173713978,
      "grad_norm": 0.9471856779020001,
      "learning_rate": 4.589571950190897e-06,
      "loss": 0.3084597587585449,
      "num_input_tokens_seen": 255965168,
      "step": 15630,
      "train_runtime": 30096.9486,
      "train_tokens_per_second": 8504.688
    },
    {
      "epoch": 1.052720172367358,
      "grad_norm": 1.0640255357883188,
      "learning_rate": 4.586936848577661e-06,
      "loss": 0.29934868812561033,
      "num_input_tokens_seen": 256047088,
      "step": 15635,
      "train_runtime": 30106.5999,
      "train_tokens_per_second": 8504.683
    },
    {
      "epoch": 1.053056827363318,
      "grad_norm": 0.8042135266723359,
      "learning_rate": 4.584301862476219e-06,
      "loss": 0.3069495677947998,
      "num_input_tokens_seen": 256129008,
      "step": 15640,
      "train_runtime": 30116.2883,
      "train_tokens_per_second": 8504.667
    },
    {
      "epoch": 1.053393482359278,
      "grad_norm": 1.0670901247890712,
      "learning_rate": 4.581666992623435e-06,
      "loss": 0.2867729187011719,
      "num_input_tokens_seen": 256210928,
      "step": 15645,
      "train_runtime": 30125.9229,
      "train_tokens_per_second": 8504.667
    },
    {
      "epoch": 1.0537301373552383,
      "grad_norm": 1.0767674129140536,
      "learning_rate": 4.579032239756144e-06,
      "loss": 0.2890630722045898,
      "num_input_tokens_seen": 256292848,
      "step": 15650,
      "train_runtime": 30135.5514,
      "train_tokens_per_second": 8504.668
    },
    {
      "epoch": 1.0540667923511986,
      "grad_norm": 0.8991879122465177,
      "learning_rate": 4.576397604611143e-06,
      "loss": 0.2681386709213257,
      "num_input_tokens_seen": 256374768,
      "step": 15655,
      "train_runtime": 30145.1998,
      "train_tokens_per_second": 8504.663
    },
    {
      "epoch": 1.0544034473471586,
      "grad_norm": 1.2278137829914635,
      "learning_rate": 4.573763087925202e-06,
      "loss": 0.3228842496871948,
      "num_input_tokens_seen": 256456688,
      "step": 15660,
      "train_runtime": 30154.8852,
      "train_tokens_per_second": 8504.648
    },
    {
      "epoch": 1.0547401023431189,
      "grad_norm": 1.1871466575224479,
      "learning_rate": 4.571128690435053e-06,
      "loss": 0.3223149538040161,
      "num_input_tokens_seen": 256538608,
      "step": 15665,
      "train_runtime": 30164.5727,
      "train_tokens_per_second": 8504.633
    },
    {
      "epoch": 1.055076757339079,
      "grad_norm": 1.2835039312966345,
      "learning_rate": 4.568494412877401e-06,
      "loss": 0.30426223278045655,
      "num_input_tokens_seen": 256620528,
      "step": 15670,
      "train_runtime": 30174.149,
      "train_tokens_per_second": 8504.648
    },
    {
      "epoch": 1.0554134123350392,
      "grad_norm": 1.0875857540871061,
      "learning_rate": 4.565860255988908e-06,
      "loss": 0.2981568336486816,
      "num_input_tokens_seen": 256702400,
      "step": 15675,
      "train_runtime": 30184.197,
      "train_tokens_per_second": 8504.53
    },
    {
      "epoch": 1.0557500673309992,
      "grad_norm": 1.0018910332707611,
      "learning_rate": 4.563226220506212e-06,
      "loss": 0.28299698829650877,
      "num_input_tokens_seen": 256784320,
      "step": 15680,
      "train_runtime": 30193.6972,
      "train_tokens_per_second": 8504.567
    },
    {
      "epoch": 1.0560867223269594,
      "grad_norm": 0.849483885148106,
      "learning_rate": 4.5605923071659085e-06,
      "loss": 0.2851670742034912,
      "num_input_tokens_seen": 256866240,
      "step": 15685,
      "train_runtime": 30203.1897,
      "train_tokens_per_second": 8504.606
    },
    {
      "epoch": 1.0564233773229195,
      "grad_norm": 1.0930230547949753,
      "learning_rate": 4.557958516704566e-06,
      "loss": 0.2798621416091919,
      "num_input_tokens_seen": 256948160,
      "step": 15690,
      "train_runtime": 30212.7999,
      "train_tokens_per_second": 8504.613
    },
    {
      "epoch": 1.0567600323188797,
      "grad_norm": 0.9859562379169785,
      "learning_rate": 4.5553248498587144e-06,
      "loss": 0.3054790019989014,
      "num_input_tokens_seen": 257030080,
      "step": 15695,
      "train_runtime": 30222.3656,
      "train_tokens_per_second": 8504.631
    },
    {
      "epoch": 1.0570966873148397,
      "grad_norm": 1.123663625684198,
      "learning_rate": 4.552691307364853e-06,
      "loss": 0.28694615364074705,
      "num_input_tokens_seen": 257112000,
      "step": 15700,
      "train_runtime": 30231.8706,
      "train_tokens_per_second": 8504.667
    },
    {
      "epoch": 1.0574333423108,
      "grad_norm": 1.274705333161985,
      "learning_rate": 4.5500578899594375e-06,
      "loss": 0.29702072143554686,
      "num_input_tokens_seen": 257193920,
      "step": 15705,
      "train_runtime": 30241.5817,
      "train_tokens_per_second": 8504.645
    },
    {
      "epoch": 1.05776999730676,
      "grad_norm": 0.9242010209659047,
      "learning_rate": 4.5474245983789e-06,
      "loss": 0.3143962621688843,
      "num_input_tokens_seen": 257275840,
      "step": 15710,
      "train_runtime": 30251.1883,
      "train_tokens_per_second": 8504.652
    },
    {
      "epoch": 1.0581066523027203,
      "grad_norm": 0.9815744625378056,
      "learning_rate": 4.54479143335963e-06,
      "loss": 0.2723519802093506,
      "num_input_tokens_seen": 257357760,
      "step": 15715,
      "train_runtime": 30260.9238,
      "train_tokens_per_second": 8504.623
    },
    {
      "epoch": 1.0584433072986803,
      "grad_norm": 1.160701377530122,
      "learning_rate": 4.5421583956379855e-06,
      "loss": 0.2934859275817871,
      "num_input_tokens_seen": 257439680,
      "step": 15720,
      "train_runtime": 30270.4486,
      "train_tokens_per_second": 8504.654
    },
    {
      "epoch": 1.0587799622946406,
      "grad_norm": 1.0154957038693808,
      "learning_rate": 4.539525485950284e-06,
      "loss": 0.2787100553512573,
      "num_input_tokens_seen": 257521600,
      "step": 15725,
      "train_runtime": 30280.1429,
      "train_tokens_per_second": 8504.636
    },
    {
      "epoch": 1.0591166172906006,
      "grad_norm": 1.1529136287534785,
      "learning_rate": 4.536892705032815e-06,
      "loss": 0.31363487243652344,
      "num_input_tokens_seen": 257603520,
      "step": 15730,
      "train_runtime": 30289.8662,
      "train_tokens_per_second": 8504.611
    },
    {
      "epoch": 1.0594532722865608,
      "grad_norm": 1.0417203187985544,
      "learning_rate": 4.534260053621821e-06,
      "loss": 0.30051631927490235,
      "num_input_tokens_seen": 257685440,
      "step": 15735,
      "train_runtime": 30299.5504,
      "train_tokens_per_second": 8504.596
    },
    {
      "epoch": 1.0597899272825209,
      "grad_norm": 0.9830866070928939,
      "learning_rate": 4.531627532453519e-06,
      "loss": 0.2678525924682617,
      "num_input_tokens_seen": 257767360,
      "step": 15740,
      "train_runtime": 30309.1835,
      "train_tokens_per_second": 8504.596
    },
    {
      "epoch": 1.0601265822784811,
      "grad_norm": 1.0794958624730988,
      "learning_rate": 4.5289951422640826e-06,
      "loss": 0.321579122543335,
      "num_input_tokens_seen": 257849280,
      "step": 15745,
      "train_runtime": 30318.7301,
      "train_tokens_per_second": 8504.62
    },
    {
      "epoch": 1.0604632372744411,
      "grad_norm": 1.5455104643101327,
      "learning_rate": 4.526362883789655e-06,
      "loss": 0.30257525444030764,
      "num_input_tokens_seen": 257931200,
      "step": 15750,
      "train_runtime": 30328.3987,
      "train_tokens_per_second": 8504.61
    },
    {
      "epoch": 1.0607998922704014,
      "grad_norm": 1.265452021413695,
      "learning_rate": 4.5237307577663345e-06,
      "loss": 0.3371888160705566,
      "num_input_tokens_seen": 258013120,
      "step": 15755,
      "train_runtime": 30338.062,
      "train_tokens_per_second": 8504.601
    },
    {
      "epoch": 1.0611365472663614,
      "grad_norm": 0.9574998314224704,
      "learning_rate": 4.52109876493019e-06,
      "loss": 0.3089162349700928,
      "num_input_tokens_seen": 258095040,
      "step": 15760,
      "train_runtime": 30347.7718,
      "train_tokens_per_second": 8504.58
    },
    {
      "epoch": 1.0614732022623217,
      "grad_norm": 1.0000318272899709,
      "learning_rate": 4.518466906017249e-06,
      "loss": 0.3015407085418701,
      "num_input_tokens_seen": 258176960,
      "step": 15765,
      "train_runtime": 30357.4854,
      "train_tokens_per_second": 8504.557
    },
    {
      "epoch": 1.0618098572582817,
      "grad_norm": 1.0632660205259867,
      "learning_rate": 4.515835181763503e-06,
      "loss": 0.2754048347473145,
      "num_input_tokens_seen": 258258880,
      "step": 15770,
      "train_runtime": 30367.1972,
      "train_tokens_per_second": 8504.535
    },
    {
      "epoch": 1.062146512254242,
      "grad_norm": 1.1168425885319204,
      "learning_rate": 4.513203592904903e-06,
      "loss": 0.31572372913360597,
      "num_input_tokens_seen": 258340800,
      "step": 15775,
      "train_runtime": 30376.8985,
      "train_tokens_per_second": 8504.515
    },
    {
      "epoch": 1.062483167250202,
      "grad_norm": 1.2461790310892817,
      "learning_rate": 4.510572140177368e-06,
      "loss": 0.3054159164428711,
      "num_input_tokens_seen": 258422720,
      "step": 15780,
      "train_runtime": 30386.6057,
      "train_tokens_per_second": 8504.494
    },
    {
      "epoch": 1.0628198222461622,
      "grad_norm": 1.1419556121936367,
      "learning_rate": 4.507940824316771e-06,
      "loss": 0.32837395668029784,
      "num_input_tokens_seen": 258504640,
      "step": 15785,
      "train_runtime": 30396.2108,
      "train_tokens_per_second": 8504.502
    },
    {
      "epoch": 1.0631564772421223,
      "grad_norm": 1.0645379853133887,
      "learning_rate": 4.505309646058954e-06,
      "loss": 0.29581942558288576,
      "num_input_tokens_seen": 258586560,
      "step": 15790,
      "train_runtime": 30405.9138,
      "train_tokens_per_second": 8504.482
    },
    {
      "epoch": 1.0634931322380825,
      "grad_norm": 1.1870509205728719,
      "learning_rate": 4.502678606139716e-06,
      "loss": 0.2936953067779541,
      "num_input_tokens_seen": 258668480,
      "step": 15795,
      "train_runtime": 30415.5819,
      "train_tokens_per_second": 8504.473
    },
    {
      "epoch": 1.0638297872340425,
      "grad_norm": 0.9996301113056432,
      "learning_rate": 4.500047705294823e-06,
      "loss": 0.2909511089324951,
      "num_input_tokens_seen": 258750400,
      "step": 15800,
      "train_runtime": 30425.2764,
      "train_tokens_per_second": 8504.455
    },
    {
      "epoch": 1.0641664422300028,
      "grad_norm": 1.0628013233389382,
      "learning_rate": 4.497416944259993e-06,
      "loss": 0.2741884708404541,
      "num_input_tokens_seen": 258832320,
      "step": 15805,
      "train_runtime": 30434.9563,
      "train_tokens_per_second": 8504.442
    },
    {
      "epoch": 1.0645030972259628,
      "grad_norm": 1.1019615875115865,
      "learning_rate": 4.494786323770912e-06,
      "loss": 0.30904312133789064,
      "num_input_tokens_seen": 258914240,
      "step": 15810,
      "train_runtime": 30444.6163,
      "train_tokens_per_second": 8504.434
    },
    {
      "epoch": 1.064839752221923,
      "grad_norm": 1.0478773446839325,
      "learning_rate": 4.492155844563224e-06,
      "loss": 0.27628302574157715,
      "num_input_tokens_seen": 258996160,
      "step": 15815,
      "train_runtime": 30454.2583,
      "train_tokens_per_second": 8504.432
    },
    {
      "epoch": 1.065176407217883,
      "grad_norm": 1.055838779309148,
      "learning_rate": 4.489525507372537e-06,
      "loss": 0.30629501342773435,
      "num_input_tokens_seen": 259078080,
      "step": 15820,
      "train_runtime": 30463.9547,
      "train_tokens_per_second": 8504.414
    },
    {
      "epoch": 1.0655130622138433,
      "grad_norm": 1.023766651583277,
      "learning_rate": 4.486895312934411e-06,
      "loss": 0.30488927364349366,
      "num_input_tokens_seen": 259160000,
      "step": 15825,
      "train_runtime": 30473.6452,
      "train_tokens_per_second": 8504.398
    },
    {
      "epoch": 1.0658497172098034,
      "grad_norm": 1.2782782255817224,
      "learning_rate": 4.484265261984378e-06,
      "loss": 0.33880412578582764,
      "num_input_tokens_seen": 259241920,
      "step": 15830,
      "train_runtime": 30483.2921,
      "train_tokens_per_second": 8504.394
    },
    {
      "epoch": 1.0661863722057636,
      "grad_norm": 0.9944702193438444,
      "learning_rate": 4.4816353552579165e-06,
      "loss": 0.26977508068084716,
      "num_input_tokens_seen": 259323840,
      "step": 15835,
      "train_runtime": 30492.922,
      "train_tokens_per_second": 8504.395
    },
    {
      "epoch": 1.0665230272017237,
      "grad_norm": 1.0601564443169607,
      "learning_rate": 4.479005593490477e-06,
      "loss": 0.3082322835922241,
      "num_input_tokens_seen": 259405760,
      "step": 15840,
      "train_runtime": 30502.4156,
      "train_tokens_per_second": 8504.433
    },
    {
      "epoch": 1.066859682197684,
      "grad_norm": 1.1414123232405962,
      "learning_rate": 4.476375977417461e-06,
      "loss": 0.2983654499053955,
      "num_input_tokens_seen": 259487680,
      "step": 15845,
      "train_runtime": 30511.9094,
      "train_tokens_per_second": 8504.472
    },
    {
      "epoch": 1.067196337193644,
      "grad_norm": 1.2056105296583117,
      "learning_rate": 4.473746507774235e-06,
      "loss": 0.29499292373657227,
      "num_input_tokens_seen": 259569600,
      "step": 15850,
      "train_runtime": 30521.4159,
      "train_tokens_per_second": 8504.507
    },
    {
      "epoch": 1.0675329921896042,
      "grad_norm": 0.9774110542888093,
      "learning_rate": 4.471117185296117e-06,
      "loss": 0.2633699655532837,
      "num_input_tokens_seen": 259651520,
      "step": 15855,
      "train_runtime": 30531.0375,
      "train_tokens_per_second": 8504.51
    },
    {
      "epoch": 1.0678696471855642,
      "grad_norm": 1.0136265087684007,
      "learning_rate": 4.468488010718394e-06,
      "loss": 0.2704366683959961,
      "num_input_tokens_seen": 259733440,
      "step": 15860,
      "train_runtime": 30540.5791,
      "train_tokens_per_second": 8504.536
    },
    {
      "epoch": 1.0682063021815245,
      "grad_norm": 1.1046142251435822,
      "learning_rate": 4.465858984776302e-06,
      "loss": 0.3071028232574463,
      "num_input_tokens_seen": 259815360,
      "step": 15865,
      "train_runtime": 30550.1589,
      "train_tokens_per_second": 8504.55
    },
    {
      "epoch": 1.0685429571774845,
      "grad_norm": 1.0235282009525508,
      "learning_rate": 4.463230108205044e-06,
      "loss": 0.3084611177444458,
      "num_input_tokens_seen": 259897280,
      "step": 15870,
      "train_runtime": 30559.7689,
      "train_tokens_per_second": 8504.556
    },
    {
      "epoch": 1.0688796121734447,
      "grad_norm": 1.080661398525172,
      "learning_rate": 4.460601381739772e-06,
      "loss": 0.3041506052017212,
      "num_input_tokens_seen": 259979200,
      "step": 15875,
      "train_runtime": 30569.3632,
      "train_tokens_per_second": 8504.567
    },
    {
      "epoch": 1.0692162671694048,
      "grad_norm": 1.3411507774397982,
      "learning_rate": 4.457972806115607e-06,
      "loss": 0.32751874923706054,
      "num_input_tokens_seen": 260061120,
      "step": 15880,
      "train_runtime": 30578.8691,
      "train_tokens_per_second": 8504.602
    },
    {
      "epoch": 1.069552922165365,
      "grad_norm": 1.0658936094604483,
      "learning_rate": 4.455344382067616e-06,
      "loss": 0.2899141550064087,
      "num_input_tokens_seen": 260143040,
      "step": 15885,
      "train_runtime": 30588.4984,
      "train_tokens_per_second": 8504.603
    },
    {
      "epoch": 1.069889577161325,
      "grad_norm": 1.0610100723822502,
      "learning_rate": 4.452716110330832e-06,
      "loss": 0.32257914543151855,
      "num_input_tokens_seen": 260224960,
      "step": 15890,
      "train_runtime": 30598.0294,
      "train_tokens_per_second": 8504.631
    },
    {
      "epoch": 1.0702262321572853,
      "grad_norm": 0.9720747882823504,
      "learning_rate": 4.450087991640242e-06,
      "loss": 0.30805397033691406,
      "num_input_tokens_seen": 260306432,
      "step": 15895,
      "train_runtime": 30607.9958,
      "train_tokens_per_second": 8504.524
    },
    {
      "epoch": 1.0705628871532453,
      "grad_norm": 1.0306169253825281,
      "learning_rate": 4.447460026730794e-06,
      "loss": 0.2878419876098633,
      "num_input_tokens_seen": 260388352,
      "step": 15900,
      "train_runtime": 30617.706,
      "train_tokens_per_second": 8504.502
    },
    {
      "epoch": 1.0708995421492056,
      "grad_norm": 1.3906067547134653,
      "learning_rate": 4.444832216337384e-06,
      "loss": 0.3074807643890381,
      "num_input_tokens_seen": 260470080,
      "step": 15905,
      "train_runtime": 30627.8698,
      "train_tokens_per_second": 8504.349
    },
    {
      "epoch": 1.0712361971451656,
      "grad_norm": 1.2471330291391747,
      "learning_rate": 4.442204561194877e-06,
      "loss": 0.2848317861557007,
      "num_input_tokens_seen": 260552000,
      "step": 15910,
      "train_runtime": 30637.5259,
      "train_tokens_per_second": 8504.342
    },
    {
      "epoch": 1.0715728521411259,
      "grad_norm": 1.078884738500063,
      "learning_rate": 4.439577062038085e-06,
      "loss": 0.2851496696472168,
      "num_input_tokens_seen": 260633920,
      "step": 15915,
      "train_runtime": 30647.1936,
      "train_tokens_per_second": 8504.332
    },
    {
      "epoch": 1.0719095071370859,
      "grad_norm": 1.1430753039066486,
      "learning_rate": 4.436949719601782e-06,
      "loss": 0.3014174699783325,
      "num_input_tokens_seen": 260715840,
      "step": 15920,
      "train_runtime": 30656.8412,
      "train_tokens_per_second": 8504.328
    },
    {
      "epoch": 1.0722461621330461,
      "grad_norm": 1.1356014394712532,
      "learning_rate": 4.434322534620692e-06,
      "loss": 0.3022881269454956,
      "num_input_tokens_seen": 260796568,
      "step": 15925,
      "train_runtime": 30666.4566,
      "train_tokens_per_second": 8504.294
    },
    {
      "epoch": 1.0725828171290062,
      "grad_norm": 1.1521371827323084,
      "learning_rate": 4.431695507829504e-06,
      "loss": 0.25545530319213866,
      "num_input_tokens_seen": 260878488,
      "step": 15930,
      "train_runtime": 30675.965,
      "train_tokens_per_second": 8504.329
    },
    {
      "epoch": 1.0729194721249664,
      "grad_norm": 1.083254646767001,
      "learning_rate": 4.429068639962853e-06,
      "loss": 0.3149627447128296,
      "num_input_tokens_seen": 260959896,
      "step": 15935,
      "train_runtime": 30685.7492,
      "train_tokens_per_second": 8504.27
    },
    {
      "epoch": 1.0732561271209264,
      "grad_norm": 0.8655903901982223,
      "learning_rate": 4.426441931755337e-06,
      "loss": 0.28327672481536864,
      "num_input_tokens_seen": 261041816,
      "step": 15940,
      "train_runtime": 30695.4177,
      "train_tokens_per_second": 8504.26
    },
    {
      "epoch": 1.0735927821168867,
      "grad_norm": 0.9239859614437073,
      "learning_rate": 4.4238153839415045e-06,
      "loss": 0.29706945419311526,
      "num_input_tokens_seen": 261123736,
      "step": 15945,
      "train_runtime": 30705.082,
      "train_tokens_per_second": 8504.251
    },
    {
      "epoch": 1.0739294371128467,
      "grad_norm": 1.0253914769543855,
      "learning_rate": 4.421188997255865e-06,
      "loss": 0.3020064353942871,
      "num_input_tokens_seen": 261205656,
      "step": 15950,
      "train_runtime": 30714.7193,
      "train_tokens_per_second": 8504.25
    },
    {
      "epoch": 1.074266092108807,
      "grad_norm": 1.2154297618360088,
      "learning_rate": 4.418562772432874e-06,
      "loss": 0.3180800676345825,
      "num_input_tokens_seen": 261287576,
      "step": 15955,
      "train_runtime": 30724.395,
      "train_tokens_per_second": 8504.238
    },
    {
      "epoch": 1.074602747104767,
      "grad_norm": 1.2352814939309331,
      "learning_rate": 4.415936710206952e-06,
      "loss": 0.3027245044708252,
      "num_input_tokens_seen": 261369496,
      "step": 15960,
      "train_runtime": 30734.0783,
      "train_tokens_per_second": 8504.224
    },
    {
      "epoch": 1.0749394021007272,
      "grad_norm": 1.1900569671486319,
      "learning_rate": 4.413310811312465e-06,
      "loss": 0.2899316787719727,
      "num_input_tokens_seen": 261451416,
      "step": 15965,
      "train_runtime": 30743.6966,
      "train_tokens_per_second": 8504.228
    },
    {
      "epoch": 1.0752760570966873,
      "grad_norm": 0.9596582383106192,
      "learning_rate": 4.41068507648374e-06,
      "loss": 0.27515349388122556,
      "num_input_tokens_seen": 261533336,
      "step": 15970,
      "train_runtime": 30753.2266,
      "train_tokens_per_second": 8504.257
    },
    {
      "epoch": 1.0756127120926475,
      "grad_norm": 1.0287623475219583,
      "learning_rate": 4.408059506455053e-06,
      "loss": 0.2976978778839111,
      "num_input_tokens_seen": 261615256,
      "step": 15975,
      "train_runtime": 30762.9326,
      "train_tokens_per_second": 8504.237
    },
    {
      "epoch": 1.0759493670886076,
      "grad_norm": 0.9016726530299874,
      "learning_rate": 4.40543410196064e-06,
      "loss": 0.26392326354980467,
      "num_input_tokens_seen": 261697176,
      "step": 15980,
      "train_runtime": 30772.5612,
      "train_tokens_per_second": 8504.238
    },
    {
      "epoch": 1.0762860220845678,
      "grad_norm": 1.089320931760244,
      "learning_rate": 4.402808863734681e-06,
      "loss": 0.32344305515289307,
      "num_input_tokens_seen": 261779096,
      "step": 15985,
      "train_runtime": 30782.3428,
      "train_tokens_per_second": 8504.197
    },
    {
      "epoch": 1.0766226770805278,
      "grad_norm": 2.1672989746248525,
      "learning_rate": 4.40018379251132e-06,
      "loss": 0.29990859031677247,
      "num_input_tokens_seen": 261861016,
      "step": 15990,
      "train_runtime": 30791.8456,
      "train_tokens_per_second": 8504.233
    },
    {
      "epoch": 1.076959332076488,
      "grad_norm": 1.2816060197326025,
      "learning_rate": 4.397558889024649e-06,
      "loss": 0.3063052177429199,
      "num_input_tokens_seen": 261942832,
      "step": 15995,
      "train_runtime": 30801.9103,
      "train_tokens_per_second": 8504.11
    },
    {
      "epoch": 1.0772959870724481,
      "grad_norm": 0.9630182046224102,
      "learning_rate": 4.394934154008715e-06,
      "loss": 0.3004565477371216,
      "num_input_tokens_seen": 262024752,
      "step": 16000,
      "train_runtime": 30811.4355,
      "train_tokens_per_second": 8504.14
    },
    {
      "epoch": 1.0776326420684084,
      "grad_norm": 1.0138219563955535,
      "learning_rate": 4.392309588197514e-06,
      "loss": 0.30215456485748293,
      "num_input_tokens_seen": 262106672,
      "step": 16005,
      "train_runtime": 30821.0441,
      "train_tokens_per_second": 8504.146
    },
    {
      "epoch": 1.0779692970643684,
      "grad_norm": 0.9335599885543524,
      "learning_rate": 4.389685192324999e-06,
      "loss": 0.2691475868225098,
      "num_input_tokens_seen": 262188592,
      "step": 16010,
      "train_runtime": 30830.7476,
      "train_tokens_per_second": 8504.127
    },
    {
      "epoch": 1.0783059520603286,
      "grad_norm": 1.0213202177961467,
      "learning_rate": 4.387060967125073e-06,
      "loss": 0.29409046173095704,
      "num_input_tokens_seen": 262270288,
      "step": 16015,
      "train_runtime": 30840.9345,
      "train_tokens_per_second": 8503.967
    },
    {
      "epoch": 1.0786426070562887,
      "grad_norm": 1.1679427115672356,
      "learning_rate": 4.384436913331596e-06,
      "loss": 0.33565165996551516,
      "num_input_tokens_seen": 262352208,
      "step": 16020,
      "train_runtime": 30850.5677,
      "train_tokens_per_second": 8503.967
    },
    {
      "epoch": 1.078979262052249,
      "grad_norm": 1.0727333973587858,
      "learning_rate": 4.38181303167837e-06,
      "loss": 0.29699656963348386,
      "num_input_tokens_seen": 262434128,
      "step": 16025,
      "train_runtime": 30860.0654,
      "train_tokens_per_second": 8504.004
    },
    {
      "epoch": 1.079315917048209,
      "grad_norm": 1.181397888527305,
      "learning_rate": 4.379189322899159e-06,
      "loss": 0.2838745594024658,
      "num_input_tokens_seen": 262516048,
      "step": 16030,
      "train_runtime": 30869.5813,
      "train_tokens_per_second": 8504.037
    },
    {
      "epoch": 1.0796525720441692,
      "grad_norm": 1.144020329704365,
      "learning_rate": 4.376565787727676e-06,
      "loss": 0.2837462186813354,
      "num_input_tokens_seen": 262597968,
      "step": 16035,
      "train_runtime": 30879.0896,
      "train_tokens_per_second": 8504.071
    },
    {
      "epoch": 1.0799892270401292,
      "grad_norm": 0.9970849372966765,
      "learning_rate": 4.373942426897581e-06,
      "loss": 0.29268341064453124,
      "num_input_tokens_seen": 262679888,
      "step": 16040,
      "train_runtime": 30888.7156,
      "train_tokens_per_second": 8504.073
    },
    {
      "epoch": 1.0803258820360895,
      "grad_norm": 1.041698134413779,
      "learning_rate": 4.37131924114249e-06,
      "loss": 0.29241721630096434,
      "num_input_tokens_seen": 262761808,
      "step": 16045,
      "train_runtime": 30898.3701,
      "train_tokens_per_second": 8504.067
    },
    {
      "epoch": 1.0806625370320495,
      "grad_norm": 1.0884717944408586,
      "learning_rate": 4.368696231195968e-06,
      "loss": 0.31699633598327637,
      "num_input_tokens_seen": 262843728,
      "step": 16050,
      "train_runtime": 30908.0083,
      "train_tokens_per_second": 8504.066
    },
    {
      "epoch": 1.0809991920280098,
      "grad_norm": 1.2656416662433378,
      "learning_rate": 4.366073397791534e-06,
      "loss": 0.27146134376525877,
      "num_input_tokens_seen": 262925648,
      "step": 16055,
      "train_runtime": 30917.5335,
      "train_tokens_per_second": 8504.095
    },
    {
      "epoch": 1.0813358470239698,
      "grad_norm": 1.1967531108891865,
      "learning_rate": 4.363450741662651e-06,
      "loss": 0.26724107265472413,
      "num_input_tokens_seen": 263007568,
      "step": 16060,
      "train_runtime": 30927.1063,
      "train_tokens_per_second": 8504.112
    },
    {
      "epoch": 1.08167250201993,
      "grad_norm": 1.1068751544374378,
      "learning_rate": 4.360828263542739e-06,
      "loss": 0.28857152462005614,
      "num_input_tokens_seen": 263089488,
      "step": 16065,
      "train_runtime": 30936.664,
      "train_tokens_per_second": 8504.132
    },
    {
      "epoch": 1.08200915701589,
      "grad_norm": 1.1992107513005275,
      "learning_rate": 4.358205964165164e-06,
      "loss": 0.29805865287780764,
      "num_input_tokens_seen": 263171408,
      "step": 16070,
      "train_runtime": 30946.2774,
      "train_tokens_per_second": 8504.138
    },
    {
      "epoch": 1.0823458120118503,
      "grad_norm": 1.1899544314139663,
      "learning_rate": 4.355583844263247e-06,
      "loss": 0.31063218116760255,
      "num_input_tokens_seen": 263253328,
      "step": 16075,
      "train_runtime": 30955.9452,
      "train_tokens_per_second": 8504.128
    },
    {
      "epoch": 1.0826824670078103,
      "grad_norm": 0.9789029324762066,
      "learning_rate": 4.352961904570251e-06,
      "loss": 0.288391637802124,
      "num_input_tokens_seen": 263335056,
      "step": 16080,
      "train_runtime": 30966.0704,
      "train_tokens_per_second": 8503.987
    },
    {
      "epoch": 1.0830191220037706,
      "grad_norm": 0.9902846865284451,
      "learning_rate": 4.350340145819397e-06,
      "loss": 0.2976391792297363,
      "num_input_tokens_seen": 263416976,
      "step": 16085,
      "train_runtime": 30975.5633,
      "train_tokens_per_second": 8504.025
    },
    {
      "epoch": 1.0833557769997306,
      "grad_norm": 1.134787196512875,
      "learning_rate": 4.347718568743847e-06,
      "loss": 0.2947996139526367,
      "num_input_tokens_seen": 263498896,
      "step": 16090,
      "train_runtime": 30985.1869,
      "train_tokens_per_second": 8504.028
    },
    {
      "epoch": 1.0836924319956909,
      "grad_norm": 1.103716046722558,
      "learning_rate": 4.34509717407672e-06,
      "loss": 0.3263574123382568,
      "num_input_tokens_seen": 263580816,
      "step": 16095,
      "train_runtime": 30994.7722,
      "train_tokens_per_second": 8504.041
    },
    {
      "epoch": 1.084029086991651,
      "grad_norm": 1.0639331885978558,
      "learning_rate": 4.342475962551081e-06,
      "loss": 0.28209874629974363,
      "num_input_tokens_seen": 263662736,
      "step": 16100,
      "train_runtime": 31004.2728,
      "train_tokens_per_second": 8504.077
    },
    {
      "epoch": 1.0843657419876112,
      "grad_norm": 0.9822223367769962,
      "learning_rate": 4.339854934899943e-06,
      "loss": 0.28464493751525877,
      "num_input_tokens_seen": 263744656,
      "step": 16105,
      "train_runtime": 31013.7706,
      "train_tokens_per_second": 8504.115
    },
    {
      "epoch": 1.0847023969835712,
      "grad_norm": 0.8302014066553635,
      "learning_rate": 4.337234091856265e-06,
      "loss": 0.25967166423797605,
      "num_input_tokens_seen": 263826576,
      "step": 16110,
      "train_runtime": 31023.2714,
      "train_tokens_per_second": 8504.151
    },
    {
      "epoch": 1.0850390519795314,
      "grad_norm": 1.1060902134653454,
      "learning_rate": 4.334613434152961e-06,
      "loss": 0.3104947566986084,
      "num_input_tokens_seen": 263908496,
      "step": 16115,
      "train_runtime": 31032.7706,
      "train_tokens_per_second": 8504.187
    },
    {
      "epoch": 1.0853757069754915,
      "grad_norm": 1.3014530384560294,
      "learning_rate": 4.331992962522888e-06,
      "loss": 0.3263056755065918,
      "num_input_tokens_seen": 263990416,
      "step": 16120,
      "train_runtime": 31042.2727,
      "train_tokens_per_second": 8504.223
    },
    {
      "epoch": 1.0857123619714517,
      "grad_norm": 1.1283808787684366,
      "learning_rate": 4.329372677698854e-06,
      "loss": 0.29098508358001707,
      "num_input_tokens_seen": 264072336,
      "step": 16125,
      "train_runtime": 31051.7669,
      "train_tokens_per_second": 8504.261
    },
    {
      "epoch": 1.0860490169674117,
      "grad_norm": 1.2351109181151645,
      "learning_rate": 4.3267525804136114e-06,
      "loss": 0.2959153175354004,
      "num_input_tokens_seen": 264154256,
      "step": 16130,
      "train_runtime": 31061.2694,
      "train_tokens_per_second": 8504.297
    },
    {
      "epoch": 1.086385671963372,
      "grad_norm": 1.0555884337204318,
      "learning_rate": 4.324132671399864e-06,
      "loss": 0.291260814666748,
      "num_input_tokens_seen": 264236176,
      "step": 16135,
      "train_runtime": 31070.777,
      "train_tokens_per_second": 8504.331
    },
    {
      "epoch": 1.086722326959332,
      "grad_norm": 1.1666374268120168,
      "learning_rate": 4.321512951390258e-06,
      "loss": 0.31822335720062256,
      "num_input_tokens_seen": 264318096,
      "step": 16140,
      "train_runtime": 31080.2937,
      "train_tokens_per_second": 8504.363
    },
    {
      "epoch": 1.0870589819552923,
      "grad_norm": 1.2309442576014968,
      "learning_rate": 4.318893421117393e-06,
      "loss": 0.2625856399536133,
      "num_input_tokens_seen": 264400016,
      "step": 16145,
      "train_runtime": 31089.9985,
      "train_tokens_per_second": 8504.343
    },
    {
      "epoch": 1.0873956369512523,
      "grad_norm": 0.9367577010549993,
      "learning_rate": 4.31627408131381e-06,
      "loss": 0.27561235427856445,
      "num_input_tokens_seen": 264481936,
      "step": 16150,
      "train_runtime": 31099.6071,
      "train_tokens_per_second": 8504.35
    },
    {
      "epoch": 1.0877322919472125,
      "grad_norm": 1.0412587234003663,
      "learning_rate": 4.313654932712001e-06,
      "loss": 0.26760919094085694,
      "num_input_tokens_seen": 264563856,
      "step": 16155,
      "train_runtime": 31109.2616,
      "train_tokens_per_second": 8504.344
    },
    {
      "epoch": 1.0880689469431726,
      "grad_norm": 0.8587264507556973,
      "learning_rate": 4.3110359760444e-06,
      "loss": 0.3098461151123047,
      "num_input_tokens_seen": 264645776,
      "step": 16160,
      "train_runtime": 31118.7608,
      "train_tokens_per_second": 8504.38
    },
    {
      "epoch": 1.0884056019391328,
      "grad_norm": 0.9241135102249675,
      "learning_rate": 4.308417212043392e-06,
      "loss": 0.2980842590332031,
      "num_input_tokens_seen": 264727696,
      "step": 16165,
      "train_runtime": 31128.2638,
      "train_tokens_per_second": 8504.416
    },
    {
      "epoch": 1.0887422569350929,
      "grad_norm": 0.9981901763993773,
      "learning_rate": 4.305798641441304e-06,
      "loss": 0.27417731285095215,
      "num_input_tokens_seen": 264809616,
      "step": 16170,
      "train_runtime": 31137.7606,
      "train_tokens_per_second": 8504.453
    },
    {
      "epoch": 1.089078911931053,
      "grad_norm": 1.1383130738598237,
      "learning_rate": 4.303180264970416e-06,
      "loss": 0.2914299964904785,
      "num_input_tokens_seen": 264891536,
      "step": 16175,
      "train_runtime": 31147.2599,
      "train_tokens_per_second": 8504.489
    },
    {
      "epoch": 1.0894155669270131,
      "grad_norm": 0.9479716871472125,
      "learning_rate": 4.3005620833629415e-06,
      "loss": 0.30765376091003416,
      "num_input_tokens_seen": 264973456,
      "step": 16180,
      "train_runtime": 31156.7546,
      "train_tokens_per_second": 8504.527
    },
    {
      "epoch": 1.0897522219229734,
      "grad_norm": 1.1942446812797503,
      "learning_rate": 4.297944097351053e-06,
      "loss": 0.2907488584518433,
      "num_input_tokens_seen": 265055376,
      "step": 16185,
      "train_runtime": 31166.2556,
      "train_tokens_per_second": 8504.563
    },
    {
      "epoch": 1.0900888769189334,
      "grad_norm": 1.0202964276497015,
      "learning_rate": 4.295326307666856e-06,
      "loss": 0.3314064979553223,
      "num_input_tokens_seen": 265137296,
      "step": 16190,
      "train_runtime": 31175.7576,
      "train_tokens_per_second": 8504.598
    },
    {
      "epoch": 1.0904255319148937,
      "grad_norm": 1.1167077373125827,
      "learning_rate": 4.292708715042411e-06,
      "loss": 0.2896484136581421,
      "num_input_tokens_seen": 265218208,
      "step": 16195,
      "train_runtime": 31185.4237,
      "train_tokens_per_second": 8504.557
    },
    {
      "epoch": 1.0907621869108537,
      "grad_norm": 1.0177281481621887,
      "learning_rate": 4.290091320209718e-06,
      "loss": 0.2748218536376953,
      "num_input_tokens_seen": 265300128,
      "step": 16200,
      "train_runtime": 31194.9505,
      "train_tokens_per_second": 8504.586
    },
    {
      "epoch": 1.091098841906814,
      "grad_norm": 0.9326770990539867,
      "learning_rate": 4.287474123900726e-06,
      "loss": 0.29551217555999754,
      "num_input_tokens_seen": 265382048,
      "step": 16205,
      "train_runtime": 31204.4838,
      "train_tokens_per_second": 8504.613
    },
    {
      "epoch": 1.091435496902774,
      "grad_norm": 1.0373862941665466,
      "learning_rate": 4.284857126847321e-06,
      "loss": 0.3091291904449463,
      "num_input_tokens_seen": 265463584,
      "step": 16210,
      "train_runtime": 31214.1676,
      "train_tokens_per_second": 8504.586
    },
    {
      "epoch": 1.0917721518987342,
      "grad_norm": 1.003417099482249,
      "learning_rate": 4.28224032978134e-06,
      "loss": 0.28601109981536865,
      "num_input_tokens_seen": 265545504,
      "step": 16215,
      "train_runtime": 31223.8293,
      "train_tokens_per_second": 8504.578
    },
    {
      "epoch": 1.0921088068946943,
      "grad_norm": 1.215758384012362,
      "learning_rate": 4.279623733434561e-06,
      "loss": 0.32459197044372556,
      "num_input_tokens_seen": 265627424,
      "step": 16220,
      "train_runtime": 31233.3364,
      "train_tokens_per_second": 8504.613
    },
    {
      "epoch": 1.0924454618906545,
      "grad_norm": 1.2225279355883165,
      "learning_rate": 4.277007338538711e-06,
      "loss": 0.2876670122146606,
      "num_input_tokens_seen": 265709344,
      "step": 16225,
      "train_runtime": 31242.8504,
      "train_tokens_per_second": 8504.645
    },
    {
      "epoch": 1.0927821168866145,
      "grad_norm": 0.975452883793185,
      "learning_rate": 4.27439114582545e-06,
      "loss": 0.27702739238739016,
      "num_input_tokens_seen": 265791264,
      "step": 16230,
      "train_runtime": 31252.3497,
      "train_tokens_per_second": 8504.681
    },
    {
      "epoch": 1.0931187718825748,
      "grad_norm": 1.1942459433938304,
      "learning_rate": 4.271775156026395e-06,
      "loss": 0.3010574817657471,
      "num_input_tokens_seen": 265873184,
      "step": 16235,
      "train_runtime": 31261.8422,
      "train_tokens_per_second": 8504.719
    },
    {
      "epoch": 1.0934554268785348,
      "grad_norm": 1.0504107418226634,
      "learning_rate": 4.269159369873093e-06,
      "loss": 0.29325597286224364,
      "num_input_tokens_seen": 265954984,
      "step": 16240,
      "train_runtime": 31271.8789,
      "train_tokens_per_second": 8504.605
    },
    {
      "epoch": 1.093792081874495,
      "grad_norm": 0.9833649326836915,
      "learning_rate": 4.266543788097043e-06,
      "loss": 0.2876462936401367,
      "num_input_tokens_seen": 266036904,
      "step": 16245,
      "train_runtime": 31281.3897,
      "train_tokens_per_second": 8504.638
    },
    {
      "epoch": 1.094128736870455,
      "grad_norm": 1.4580996893236577,
      "learning_rate": 4.263928411429683e-06,
      "loss": 0.32999043464660643,
      "num_input_tokens_seen": 266118712,
      "step": 16250,
      "train_runtime": 31291.5348,
      "train_tokens_per_second": 8504.495
    },
    {
      "epoch": 1.0944653918664153,
      "grad_norm": 1.168522475678685,
      "learning_rate": 4.261313240602398e-06,
      "loss": 0.30488123893737795,
      "num_input_tokens_seen": 266200632,
      "step": 16255,
      "train_runtime": 31301.223,
      "train_tokens_per_second": 8504.48
    },
    {
      "epoch": 1.0948020468623754,
      "grad_norm": 1.1966451406195056,
      "learning_rate": 4.2586982763465064e-06,
      "loss": 0.287630033493042,
      "num_input_tokens_seen": 266282552,
      "step": 16260,
      "train_runtime": 31310.8691,
      "train_tokens_per_second": 8504.477
    },
    {
      "epoch": 1.0951387018583356,
      "grad_norm": 1.083126315182361,
      "learning_rate": 4.256083519393281e-06,
      "loss": 0.29693899154663084,
      "num_input_tokens_seen": 266364472,
      "step": 16265,
      "train_runtime": 31320.3632,
      "train_tokens_per_second": 8504.514
    },
    {
      "epoch": 1.0954753568542956,
      "grad_norm": 1.0634196008643244,
      "learning_rate": 4.253468970473925e-06,
      "loss": 0.3023137807846069,
      "num_input_tokens_seen": 266446392,
      "step": 16270,
      "train_runtime": 31329.9601,
      "train_tokens_per_second": 8504.524
    },
    {
      "epoch": 1.095812011850256,
      "grad_norm": 0.92480612122469,
      "learning_rate": 4.250854630319593e-06,
      "loss": 0.29716603755950927,
      "num_input_tokens_seen": 266528312,
      "step": 16275,
      "train_runtime": 31339.4593,
      "train_tokens_per_second": 8504.56
    },
    {
      "epoch": 1.096148666846216,
      "grad_norm": 1.080146849368891,
      "learning_rate": 4.248240499661373e-06,
      "loss": 0.28090567588806153,
      "num_input_tokens_seen": 266610232,
      "step": 16280,
      "train_runtime": 31349.1025,
      "train_tokens_per_second": 8504.557
    },
    {
      "epoch": 1.0964853218421762,
      "grad_norm": 0.9973094371396768,
      "learning_rate": 4.245626579230302e-06,
      "loss": 0.28320088386535647,
      "num_input_tokens_seen": 266692152,
      "step": 16285,
      "train_runtime": 31358.7279,
      "train_tokens_per_second": 8504.559
    },
    {
      "epoch": 1.0968219768381362,
      "grad_norm": 1.4197341197327529,
      "learning_rate": 4.243012869757351e-06,
      "loss": 0.30205450057983396,
      "num_input_tokens_seen": 266774072,
      "step": 16290,
      "train_runtime": 31368.3495,
      "train_tokens_per_second": 8504.562
    },
    {
      "epoch": 1.0971586318340965,
      "grad_norm": 1.1490156481840688,
      "learning_rate": 4.240399371973438e-06,
      "loss": 0.28776934146881106,
      "num_input_tokens_seen": 266855992,
      "step": 16295,
      "train_runtime": 31377.852,
      "train_tokens_per_second": 8504.597
    },
    {
      "epoch": 1.0974952868300565,
      "grad_norm": 1.069942851903785,
      "learning_rate": 4.2377860866094165e-06,
      "loss": 0.2949162483215332,
      "num_input_tokens_seen": 266937912,
      "step": 16300,
      "train_runtime": 31387.4638,
      "train_tokens_per_second": 8504.603
    },
    {
      "epoch": 1.0978319418260167,
      "grad_norm": 1.1791259868610953,
      "learning_rate": 4.235173014396088e-06,
      "loss": 0.267609977722168,
      "num_input_tokens_seen": 267019832,
      "step": 16305,
      "train_runtime": 31397.1196,
      "train_tokens_per_second": 8504.596
    },
    {
      "epoch": 1.0981685968219768,
      "grad_norm": 1.1316873424638032,
      "learning_rate": 4.2325601560641854e-06,
      "loss": 0.3253532886505127,
      "num_input_tokens_seen": 267101752,
      "step": 16310,
      "train_runtime": 31406.8088,
      "train_tokens_per_second": 8504.581
    },
    {
      "epoch": 1.098505251817937,
      "grad_norm": 1.0774402242650767,
      "learning_rate": 4.229947512344389e-06,
      "loss": 0.2851902484893799,
      "num_input_tokens_seen": 267183672,
      "step": 16315,
      "train_runtime": 31416.5115,
      "train_tokens_per_second": 8504.562
    },
    {
      "epoch": 1.098841906813897,
      "grad_norm": 1.2548800696972389,
      "learning_rate": 4.227335083967314e-06,
      "loss": 0.3136876583099365,
      "num_input_tokens_seen": 267265592,
      "step": 16320,
      "train_runtime": 31426.1892,
      "train_tokens_per_second": 8504.55
    },
    {
      "epoch": 1.0991785618098573,
      "grad_norm": 0.9676059503516043,
      "learning_rate": 4.224722871663522e-06,
      "loss": 0.2762033462524414,
      "num_input_tokens_seen": 267347512,
      "step": 16325,
      "train_runtime": 31435.838,
      "train_tokens_per_second": 8504.545
    },
    {
      "epoch": 1.0995152168058173,
      "grad_norm": 1.0628768928644534,
      "learning_rate": 4.222110876163503e-06,
      "loss": 0.32217726707458494,
      "num_input_tokens_seen": 267429432,
      "step": 16330,
      "train_runtime": 31445.3842,
      "train_tokens_per_second": 8504.569
    },
    {
      "epoch": 1.0998518718017776,
      "grad_norm": 1.3386536963753788,
      "learning_rate": 4.2194990981977e-06,
      "loss": 0.3114830493927002,
      "num_input_tokens_seen": 267511352,
      "step": 16335,
      "train_runtime": 31454.9572,
      "train_tokens_per_second": 8504.585
    },
    {
      "epoch": 1.1001885267977376,
      "grad_norm": 1.0290533614220552,
      "learning_rate": 4.216887538496482e-06,
      "loss": 0.27106313705444335,
      "num_input_tokens_seen": 267593272,
      "step": 16340,
      "train_runtime": 31464.5971,
      "train_tokens_per_second": 8504.583
    },
    {
      "epoch": 1.1005251817936978,
      "grad_norm": 1.1143679586293131,
      "learning_rate": 4.2142761977901695e-06,
      "loss": 0.3103271484375,
      "num_input_tokens_seen": 267675192,
      "step": 16345,
      "train_runtime": 31474.2238,
      "train_tokens_per_second": 8504.584
    },
    {
      "epoch": 1.1008618367896579,
      "grad_norm": 1.2509942092523465,
      "learning_rate": 4.21166507680901e-06,
      "loss": 0.30484209060668943,
      "num_input_tokens_seen": 267757112,
      "step": 16350,
      "train_runtime": 31483.8761,
      "train_tokens_per_second": 8504.579
    },
    {
      "epoch": 1.1011984917856181,
      "grad_norm": 1.240756483932799,
      "learning_rate": 4.209054176283202e-06,
      "loss": 0.2854678869247437,
      "num_input_tokens_seen": 267839032,
      "step": 16355,
      "train_runtime": 31493.3719,
      "train_tokens_per_second": 8504.616
    },
    {
      "epoch": 1.1015351467815782,
      "grad_norm": 1.1899561671713372,
      "learning_rate": 4.206443496942868e-06,
      "loss": 0.2918886423110962,
      "num_input_tokens_seen": 267920520,
      "step": 16360,
      "train_runtime": 31503.2361,
      "train_tokens_per_second": 8504.54
    },
    {
      "epoch": 1.1018718017775384,
      "grad_norm": 1.139401739915148,
      "learning_rate": 4.203833039518081e-06,
      "loss": 0.28115866184234617,
      "num_input_tokens_seen": 268002440,
      "step": 16365,
      "train_runtime": 31512.748,
      "train_tokens_per_second": 8504.572
    },
    {
      "epoch": 1.1022084567734984,
      "grad_norm": 1.028388152569662,
      "learning_rate": 4.201222804738845e-06,
      "loss": 0.29680986404418946,
      "num_input_tokens_seen": 268084360,
      "step": 16370,
      "train_runtime": 31522.3616,
      "train_tokens_per_second": 8504.577
    },
    {
      "epoch": 1.1025451117694587,
      "grad_norm": 1.1679940154023742,
      "learning_rate": 4.198612793335107e-06,
      "loss": 0.29271886348724363,
      "num_input_tokens_seen": 268166280,
      "step": 16375,
      "train_runtime": 31532.0123,
      "train_tokens_per_second": 8504.572
    },
    {
      "epoch": 1.1028817667654187,
      "grad_norm": 1.0660526424958354,
      "learning_rate": 4.196003006036743e-06,
      "loss": 0.309922456741333,
      "num_input_tokens_seen": 268248200,
      "step": 16380,
      "train_runtime": 31541.8437,
      "train_tokens_per_second": 8504.519
    },
    {
      "epoch": 1.103218421761379,
      "grad_norm": 1.1635605127656454,
      "learning_rate": 4.193393443573576e-06,
      "loss": 0.2947725772857666,
      "num_input_tokens_seen": 268330120,
      "step": 16385,
      "train_runtime": 31551.3643,
      "train_tokens_per_second": 8504.549
    },
    {
      "epoch": 1.103555076757339,
      "grad_norm": 1.0815893170340656,
      "learning_rate": 4.19078410667536e-06,
      "loss": 0.31111328601837157,
      "num_input_tokens_seen": 268412040,
      "step": 16390,
      "train_runtime": 31560.8616,
      "train_tokens_per_second": 8504.585
    },
    {
      "epoch": 1.1038917317532992,
      "grad_norm": 0.8745162733971532,
      "learning_rate": 4.188174996071788e-06,
      "loss": 0.2784985065460205,
      "num_input_tokens_seen": 268493520,
      "step": 16395,
      "train_runtime": 31570.9941,
      "train_tokens_per_second": 8504.437
    },
    {
      "epoch": 1.1042283867492593,
      "grad_norm": 0.9034393741553398,
      "learning_rate": 4.18556611249249e-06,
      "loss": 0.2819284677505493,
      "num_input_tokens_seen": 268575440,
      "step": 16400,
      "train_runtime": 31580.5731,
      "train_tokens_per_second": 8504.451
    },
    {
      "epoch": 1.1045650417452195,
      "grad_norm": 1.2802955625339287,
      "learning_rate": 4.182957456667032e-06,
      "loss": 0.31667139530181887,
      "num_input_tokens_seen": 268657360,
      "step": 16405,
      "train_runtime": 31590.0752,
      "train_tokens_per_second": 8504.486
    },
    {
      "epoch": 1.1049016967411796,
      "grad_norm": 0.9309488714435469,
      "learning_rate": 4.1803490293249145e-06,
      "loss": 0.3240525245666504,
      "num_input_tokens_seen": 268739280,
      "step": 16410,
      "train_runtime": 31599.5935,
      "train_tokens_per_second": 8504.517
    },
    {
      "epoch": 1.1052383517371398,
      "grad_norm": 1.1582645412296717,
      "learning_rate": 4.1777408311955785e-06,
      "loss": 0.2969407558441162,
      "num_input_tokens_seen": 268820880,
      "step": 16415,
      "train_runtime": 31609.2286,
      "train_tokens_per_second": 8504.506
    },
    {
      "epoch": 1.1055750067330998,
      "grad_norm": 0.9920526612592973,
      "learning_rate": 4.175132863008395e-06,
      "loss": 0.3149813175201416,
      "num_input_tokens_seen": 268902800,
      "step": 16420,
      "train_runtime": 31618.8948,
      "train_tokens_per_second": 8504.497
    },
    {
      "epoch": 1.10591166172906,
      "grad_norm": 1.2347890205422891,
      "learning_rate": 4.17252512549268e-06,
      "loss": 0.29873361587524416,
      "num_input_tokens_seen": 268984056,
      "step": 16425,
      "train_runtime": 31628.774,
      "train_tokens_per_second": 8504.41
    },
    {
      "epoch": 1.10624831672502,
      "grad_norm": 1.2604703075279573,
      "learning_rate": 4.169917619377671e-06,
      "loss": 0.27913479804992675,
      "num_input_tokens_seen": 269065976,
      "step": 16430,
      "train_runtime": 31638.3263,
      "train_tokens_per_second": 8504.431
    },
    {
      "epoch": 1.1065849717209804,
      "grad_norm": 1.386381756322484,
      "learning_rate": 4.167310345392557e-06,
      "loss": 0.31276302337646483,
      "num_input_tokens_seen": 269147896,
      "step": 16435,
      "train_runtime": 31647.893,
      "train_tokens_per_second": 8504.449
    },
    {
      "epoch": 1.1069216267169404,
      "grad_norm": 1.1617375504304241,
      "learning_rate": 4.1647033042664455e-06,
      "loss": 0.33061654567718507,
      "num_input_tokens_seen": 269229816,
      "step": 16440,
      "train_runtime": 31657.5242,
      "train_tokens_per_second": 8504.45
    },
    {
      "epoch": 1.1072582817129006,
      "grad_norm": 1.089577300379171,
      "learning_rate": 4.162096496728394e-06,
      "loss": 0.30886335372924806,
      "num_input_tokens_seen": 269311736,
      "step": 16445,
      "train_runtime": 31667.1723,
      "train_tokens_per_second": 8504.445
    },
    {
      "epoch": 1.1075949367088607,
      "grad_norm": 1.0451327238668653,
      "learning_rate": 4.1594899235073835e-06,
      "loss": 0.28856501579284666,
      "num_input_tokens_seen": 269393656,
      "step": 16450,
      "train_runtime": 31676.6979,
      "train_tokens_per_second": 8504.474
    },
    {
      "epoch": 1.107931591704821,
      "grad_norm": 1.2297598616726373,
      "learning_rate": 4.156883585332338e-06,
      "loss": 0.3070847034454346,
      "num_input_tokens_seen": 269475576,
      "step": 16455,
      "train_runtime": 31686.2462,
      "train_tokens_per_second": 8504.497
    },
    {
      "epoch": 1.108268246700781,
      "grad_norm": 0.9762497670987921,
      "learning_rate": 4.154277482932107e-06,
      "loss": 0.29737257957458496,
      "num_input_tokens_seen": 269557496,
      "step": 16460,
      "train_runtime": 31695.8244,
      "train_tokens_per_second": 8504.511
    },
    {
      "epoch": 1.1086049016967412,
      "grad_norm": 1.0633966826473478,
      "learning_rate": 4.151671617035482e-06,
      "loss": 0.281425929069519,
      "num_input_tokens_seen": 269638128,
      "step": 16465,
      "train_runtime": 31705.5543,
      "train_tokens_per_second": 8504.445
    },
    {
      "epoch": 1.1089415566927012,
      "grad_norm": 0.9141495387833567,
      "learning_rate": 4.149065988371184e-06,
      "loss": 0.2852363109588623,
      "num_input_tokens_seen": 269720048,
      "step": 16470,
      "train_runtime": 31715.1775,
      "train_tokens_per_second": 8504.447
    },
    {
      "epoch": 1.1092782116886615,
      "grad_norm": 1.00330157458748,
      "learning_rate": 4.14646059766787e-06,
      "loss": 0.28204290866851806,
      "num_input_tokens_seen": 269801968,
      "step": 16475,
      "train_runtime": 31724.6786,
      "train_tokens_per_second": 8504.482
    },
    {
      "epoch": 1.1096148666846215,
      "grad_norm": 1.058209164030456,
      "learning_rate": 4.143855445654127e-06,
      "loss": 0.33531899452209474,
      "num_input_tokens_seen": 269883888,
      "step": 16480,
      "train_runtime": 31734.2991,
      "train_tokens_per_second": 8504.486
    },
    {
      "epoch": 1.1099515216805818,
      "grad_norm": 1.2144824875771774,
      "learning_rate": 4.14125053305848e-06,
      "loss": 0.27022390365600585,
      "num_input_tokens_seen": 269965808,
      "step": 16485,
      "train_runtime": 31743.9068,
      "train_tokens_per_second": 8504.492
    },
    {
      "epoch": 1.1102881766765418,
      "grad_norm": 0.9473463824260276,
      "learning_rate": 4.1386458606093805e-06,
      "loss": 0.2951659679412842,
      "num_input_tokens_seen": 270047728,
      "step": 16490,
      "train_runtime": 31753.4087,
      "train_tokens_per_second": 8504.527
    },
    {
      "epoch": 1.110624831672502,
      "grad_norm": 1.2074794466681216,
      "learning_rate": 4.1360414290352214e-06,
      "loss": 0.30239861011505126,
      "num_input_tokens_seen": 270129648,
      "step": 16495,
      "train_runtime": 31763.0391,
      "train_tokens_per_second": 8504.528
    },
    {
      "epoch": 1.110961486668462,
      "grad_norm": 1.2875141725381976,
      "learning_rate": 4.1334372390643205e-06,
      "loss": 0.28926901817321776,
      "num_input_tokens_seen": 270211568,
      "step": 16500,
      "train_runtime": 31772.6618,
      "train_tokens_per_second": 8504.53
    },
    {
      "epoch": 1.1112981416644223,
      "grad_norm": 1.0306125164398843,
      "learning_rate": 4.130833291424934e-06,
      "loss": 0.3100718021392822,
      "num_input_tokens_seen": 270293488,
      "step": 16505,
      "train_runtime": 31782.2117,
      "train_tokens_per_second": 8504.553
    },
    {
      "epoch": 1.1116347966603823,
      "grad_norm": 1.2265564258369617,
      "learning_rate": 4.128229586845244e-06,
      "loss": 0.2749967098236084,
      "num_input_tokens_seen": 270375408,
      "step": 16510,
      "train_runtime": 31791.7146,
      "train_tokens_per_second": 8504.587
    },
    {
      "epoch": 1.1119714516563426,
      "grad_norm": 1.1077920088063602,
      "learning_rate": 4.125626126053372e-06,
      "loss": 0.2792144775390625,
      "num_input_tokens_seen": 270456944,
      "step": 16515,
      "train_runtime": 31801.6441,
      "train_tokens_per_second": 8504.496
    },
    {
      "epoch": 1.1123081066523026,
      "grad_norm": 1.1202626859760512,
      "learning_rate": 4.123022909777364e-06,
      "loss": 0.3200608491897583,
      "num_input_tokens_seen": 270538864,
      "step": 16520,
      "train_runtime": 31811.1943,
      "train_tokens_per_second": 8504.518
    },
    {
      "epoch": 1.1126447616482629,
      "grad_norm": 0.920395445877053,
      "learning_rate": 4.120419938745205e-06,
      "loss": 0.2941126346588135,
      "num_input_tokens_seen": 270620784,
      "step": 16525,
      "train_runtime": 31820.7889,
      "train_tokens_per_second": 8504.528
    },
    {
      "epoch": 1.112981416644223,
      "grad_norm": 0.9732704806807981,
      "learning_rate": 4.1178172136848035e-06,
      "loss": 0.28786444664001465,
      "num_input_tokens_seen": 270702368,
      "step": 16530,
      "train_runtime": 31830.8346,
      "train_tokens_per_second": 8504.407
    },
    {
      "epoch": 1.1133180716401831,
      "grad_norm": 1.0987106427594697,
      "learning_rate": 4.115214735324007e-06,
      "loss": 0.27983100414276124,
      "num_input_tokens_seen": 270784288,
      "step": 16535,
      "train_runtime": 31840.3917,
      "train_tokens_per_second": 8504.427
    },
    {
      "epoch": 1.1136547266361432,
      "grad_norm": 0.9571145450713633,
      "learning_rate": 4.112612504390586e-06,
      "loss": 0.26424570083618165,
      "num_input_tokens_seen": 270866208,
      "step": 16540,
      "train_runtime": 31849.9619,
      "train_tokens_per_second": 8504.444
    },
    {
      "epoch": 1.1139913816321034,
      "grad_norm": 1.0200167155070556,
      "learning_rate": 4.11001052161225e-06,
      "loss": 0.31450986862182617,
      "num_input_tokens_seen": 270948128,
      "step": 16545,
      "train_runtime": 31859.5838,
      "train_tokens_per_second": 8504.447
    },
    {
      "epoch": 1.1143280366280637,
      "grad_norm": 1.2516638795825372,
      "learning_rate": 4.107408787716632e-06,
      "loss": 0.3177192211151123,
      "num_input_tokens_seen": 271030048,
      "step": 16550,
      "train_runtime": 31869.2079,
      "train_tokens_per_second": 8504.449
    },
    {
      "epoch": 1.1146646916240237,
      "grad_norm": 1.031582573850077,
      "learning_rate": 4.104807303431303e-06,
      "loss": 0.2848636627197266,
      "num_input_tokens_seen": 271111968,
      "step": 16555,
      "train_runtime": 31878.7167,
      "train_tokens_per_second": 8504.482
    },
    {
      "epoch": 1.1150013466199837,
      "grad_norm": 0.9631504548662486,
      "learning_rate": 4.102206069483754e-06,
      "loss": 0.2988715410232544,
      "num_input_tokens_seen": 271193888,
      "step": 16560,
      "train_runtime": 31888.3345,
      "train_tokens_per_second": 8504.486
    },
    {
      "epoch": 1.115338001615944,
      "grad_norm": 1.0968369607143529,
      "learning_rate": 4.099605086601417e-06,
      "loss": 0.26802849769592285,
      "num_input_tokens_seen": 271275808,
      "step": 16565,
      "train_runtime": 31897.9539,
      "train_tokens_per_second": 8504.489
    },
    {
      "epoch": 1.1156746566119042,
      "grad_norm": 1.1725661309684903,
      "learning_rate": 4.097004355511645e-06,
      "loss": 0.2834800958633423,
      "num_input_tokens_seen": 271357728,
      "step": 16570,
      "train_runtime": 31907.5414,
      "train_tokens_per_second": 8504.501
    },
    {
      "epoch": 1.1160113116078643,
      "grad_norm": 1.0709775570822582,
      "learning_rate": 4.094403876941728e-06,
      "loss": 0.2624353408813477,
      "num_input_tokens_seen": 271439648,
      "step": 16575,
      "train_runtime": 31917.1442,
      "train_tokens_per_second": 8504.509
    },
    {
      "epoch": 1.1163479666038243,
      "grad_norm": 1.0395585549145827,
      "learning_rate": 4.0918036516188755e-06,
      "loss": 0.303822660446167,
      "num_input_tokens_seen": 271521568,
      "step": 16580,
      "train_runtime": 31926.7767,
      "train_tokens_per_second": 8504.509
    },
    {
      "epoch": 1.1166846215997845,
      "grad_norm": 1.2041146762207438,
      "learning_rate": 4.0892036802702395e-06,
      "loss": 0.2932334661483765,
      "num_input_tokens_seen": 271603488,
      "step": 16585,
      "train_runtime": 31936.3247,
      "train_tokens_per_second": 8504.532
    },
    {
      "epoch": 1.1170212765957448,
      "grad_norm": 0.9141746629709618,
      "learning_rate": 4.086603963622887e-06,
      "loss": 0.25414214134216306,
      "num_input_tokens_seen": 271685408,
      "step": 16590,
      "train_runtime": 31945.8969,
      "train_tokens_per_second": 8504.548
    },
    {
      "epoch": 1.1173579315917048,
      "grad_norm": 1.0071874046705325,
      "learning_rate": 4.084004502403825e-06,
      "loss": 0.33689212799072266,
      "num_input_tokens_seen": 271766520,
      "step": 16595,
      "train_runtime": 31955.6859,
      "train_tokens_per_second": 8504.481
    },
    {
      "epoch": 1.1176945865876649,
      "grad_norm": 0.9626514988329286,
      "learning_rate": 4.081405297339982e-06,
      "loss": 0.25916323661804197,
      "num_input_tokens_seen": 271848440,
      "step": 16600,
      "train_runtime": 31965.3761,
      "train_tokens_per_second": 8504.466
    },
    {
      "epoch": 1.118031241583625,
      "grad_norm": 1.1554989420734296,
      "learning_rate": 4.0788063491582205e-06,
      "loss": 0.3103208065032959,
      "num_input_tokens_seen": 271930360,
      "step": 16605,
      "train_runtime": 31975.0256,
      "train_tokens_per_second": 8504.461
    },
    {
      "epoch": 1.1183678965795854,
      "grad_norm": 1.28090973162263,
      "learning_rate": 4.076207658585323e-06,
      "loss": 0.27756052017211913,
      "num_input_tokens_seen": 272012280,
      "step": 16610,
      "train_runtime": 31984.5315,
      "train_tokens_per_second": 8504.495
    },
    {
      "epoch": 1.1187045515755454,
      "grad_norm": 1.0864767135399662,
      "learning_rate": 4.07360922634801e-06,
      "loss": 0.28603711128234866,
      "num_input_tokens_seen": 272094200,
      "step": 16615,
      "train_runtime": 31994.1259,
      "train_tokens_per_second": 8504.505
    },
    {
      "epoch": 1.1190412065715054,
      "grad_norm": 1.0594624909103698,
      "learning_rate": 4.071011053172922e-06,
      "loss": 0.2781623363494873,
      "num_input_tokens_seen": 272176120,
      "step": 16620,
      "train_runtime": 32003.7567,
      "train_tokens_per_second": 8504.505
    },
    {
      "epoch": 1.1193778615674657,
      "grad_norm": 1.1212078806969432,
      "learning_rate": 4.068413139786631e-06,
      "loss": 0.28902387619018555,
      "num_input_tokens_seen": 272258040,
      "step": 16625,
      "train_runtime": 32013.3801,
      "train_tokens_per_second": 8504.508
    },
    {
      "epoch": 1.119714516563426,
      "grad_norm": 1.3395992135719565,
      "learning_rate": 4.065815486915635e-06,
      "loss": 0.28354549407958984,
      "num_input_tokens_seen": 272339960,
      "step": 16630,
      "train_runtime": 32023.044,
      "train_tokens_per_second": 8504.499
    },
    {
      "epoch": 1.120051171559386,
      "grad_norm": 1.1362867890335664,
      "learning_rate": 4.063218095286359e-06,
      "loss": 0.3191251754760742,
      "num_input_tokens_seen": 272421880,
      "step": 16635,
      "train_runtime": 32032.6484,
      "train_tokens_per_second": 8504.507
    },
    {
      "epoch": 1.120387826555346,
      "grad_norm": 1.09155864526682,
      "learning_rate": 4.0606209656251555e-06,
      "loss": 0.2700512409210205,
      "num_input_tokens_seen": 272503800,
      "step": 16640,
      "train_runtime": 32042.1645,
      "train_tokens_per_second": 8504.538
    },
    {
      "epoch": 1.1207244815513062,
      "grad_norm": 1.0973159988851304,
      "learning_rate": 4.058024098658304e-06,
      "loss": 0.29471578598022463,
      "num_input_tokens_seen": 272585720,
      "step": 16645,
      "train_runtime": 32051.8122,
      "train_tokens_per_second": 8504.534
    },
    {
      "epoch": 1.1210611365472665,
      "grad_norm": 0.9317251640770284,
      "learning_rate": 4.055427495112008e-06,
      "loss": 0.3014401435852051,
      "num_input_tokens_seen": 272667640,
      "step": 16650,
      "train_runtime": 32061.3475,
      "train_tokens_per_second": 8504.56
    },
    {
      "epoch": 1.1213977915432265,
      "grad_norm": 1.0354514720958283,
      "learning_rate": 4.052831155712403e-06,
      "loss": 0.27642486095428465,
      "num_input_tokens_seen": 272749560,
      "step": 16655,
      "train_runtime": 32070.8597,
      "train_tokens_per_second": 8504.591
    },
    {
      "epoch": 1.1217344465391865,
      "grad_norm": 0.9805570704208246,
      "learning_rate": 4.050235081185544e-06,
      "loss": 0.26367506980895994,
      "num_input_tokens_seen": 272831480,
      "step": 16660,
      "train_runtime": 32080.4924,
      "train_tokens_per_second": 8504.591
    },
    {
      "epoch": 1.1220711015351468,
      "grad_norm": 1.0460215903079857,
      "learning_rate": 4.047639272257415e-06,
      "loss": 0.2925943613052368,
      "num_input_tokens_seen": 272913400,
      "step": 16665,
      "train_runtime": 32090.0708,
      "train_tokens_per_second": 8504.606
    },
    {
      "epoch": 1.122407756531107,
      "grad_norm": 1.1134739431757033,
      "learning_rate": 4.045043729653927e-06,
      "loss": 0.27778286933898927,
      "num_input_tokens_seen": 272995320,
      "step": 16670,
      "train_runtime": 32099.6973,
      "train_tokens_per_second": 8504.607
    },
    {
      "epoch": 1.122744411527067,
      "grad_norm": 0.9988436496012199,
      "learning_rate": 4.0424484541009175e-06,
      "loss": 0.3080436706542969,
      "num_input_tokens_seen": 273077240,
      "step": 16675,
      "train_runtime": 32109.3787,
      "train_tokens_per_second": 8504.594
    },
    {
      "epoch": 1.1230810665230273,
      "grad_norm": 1.0877108270132974,
      "learning_rate": 4.039853446324141e-06,
      "loss": 0.26916773319244386,
      "num_input_tokens_seen": 273159160,
      "step": 16680,
      "train_runtime": 32119.0013,
      "train_tokens_per_second": 8504.597
    },
    {
      "epoch": 1.1234177215189873,
      "grad_norm": 0.9613109801216217,
      "learning_rate": 4.037258707049289e-06,
      "loss": 0.27469420433044434,
      "num_input_tokens_seen": 273241080,
      "step": 16685,
      "train_runtime": 32128.4986,
      "train_tokens_per_second": 8504.633
    },
    {
      "epoch": 1.1237543765149476,
      "grad_norm": 1.0114843553669945,
      "learning_rate": 4.034664237001969e-06,
      "loss": 0.2804504156112671,
      "num_input_tokens_seen": 273323000,
      "step": 16690,
      "train_runtime": 32138.1013,
      "train_tokens_per_second": 8504.641
    },
    {
      "epoch": 1.1240910315109076,
      "grad_norm": 1.2469733070771225,
      "learning_rate": 4.032070036907716e-06,
      "loss": 0.3208060026168823,
      "num_input_tokens_seen": 273404920,
      "step": 16695,
      "train_runtime": 32147.606,
      "train_tokens_per_second": 8504.674
    },
    {
      "epoch": 1.1244276865068679,
      "grad_norm": 1.1986739325608107,
      "learning_rate": 4.029476107491991e-06,
      "loss": 0.28627991676330566,
      "num_input_tokens_seen": 273486840,
      "step": 16700,
      "train_runtime": 32157.2059,
      "train_tokens_per_second": 8504.683
    },
    {
      "epoch": 1.124764341502828,
      "grad_norm": 1.048005538638437,
      "learning_rate": 4.0268824494801784e-06,
      "loss": 0.28406662940979005,
      "num_input_tokens_seen": 273568760,
      "step": 16705,
      "train_runtime": 32166.7563,
      "train_tokens_per_second": 8504.705
    },
    {
      "epoch": 1.1251009964987881,
      "grad_norm": 0.968593735067805,
      "learning_rate": 4.024289063597585e-06,
      "loss": 0.29394071102142333,
      "num_input_tokens_seen": 273650680,
      "step": 16710,
      "train_runtime": 32176.3137,
      "train_tokens_per_second": 8504.724
    },
    {
      "epoch": 1.1254376514947482,
      "grad_norm": 1.082630310097757,
      "learning_rate": 4.021695950569443e-06,
      "loss": 0.28331761360168456,
      "num_input_tokens_seen": 273732600,
      "step": 16715,
      "train_runtime": 32185.9478,
      "train_tokens_per_second": 8504.724
    },
    {
      "epoch": 1.1257743064907082,
      "grad_norm": 1.2283873367712403,
      "learning_rate": 4.01910311112091e-06,
      "loss": 0.3054789066314697,
      "num_input_tokens_seen": 273814520,
      "step": 16720,
      "train_runtime": 32195.5186,
      "train_tokens_per_second": 8504.74
    },
    {
      "epoch": 1.1261109614866684,
      "grad_norm": 1.1765240773546355,
      "learning_rate": 4.016510545977064e-06,
      "loss": 0.3222665786743164,
      "num_input_tokens_seen": 273896440,
      "step": 16725,
      "train_runtime": 32205.1281,
      "train_tokens_per_second": 8504.746
    },
    {
      "epoch": 1.1264476164826287,
      "grad_norm": 0.9451779918664058,
      "learning_rate": 4.013918255862908e-06,
      "loss": 0.29177882671356203,
      "num_input_tokens_seen": 273978320,
      "step": 16730,
      "train_runtime": 32215.3745,
      "train_tokens_per_second": 8504.583
    },
    {
      "epoch": 1.1267842714785887,
      "grad_norm": 0.9844849771900482,
      "learning_rate": 4.011326241503369e-06,
      "loss": 0.2624610424041748,
      "num_input_tokens_seen": 274060240,
      "step": 16735,
      "train_runtime": 32225.0291,
      "train_tokens_per_second": 8504.577
    },
    {
      "epoch": 1.127120926474549,
      "grad_norm": 1.1233332932375006,
      "learning_rate": 4.00873450362329e-06,
      "loss": 0.3252018690109253,
      "num_input_tokens_seen": 274142160,
      "step": 16740,
      "train_runtime": 32234.6386,
      "train_tokens_per_second": 8504.583
    },
    {
      "epoch": 1.127457581470509,
      "grad_norm": 1.001051715226029,
      "learning_rate": 4.0061430429474476e-06,
      "loss": 0.30056324005126955,
      "num_input_tokens_seen": 274223584,
      "step": 16745,
      "train_runtime": 32244.657,
      "train_tokens_per_second": 8504.466
    },
    {
      "epoch": 1.1277942364664693,
      "grad_norm": 1.0107515330886612,
      "learning_rate": 4.003551860200534e-06,
      "loss": 0.3057603597640991,
      "num_input_tokens_seen": 274305504,
      "step": 16750,
      "train_runtime": 32254.1752,
      "train_tokens_per_second": 8504.496
    },
    {
      "epoch": 1.1281308914624293,
      "grad_norm": 1.0751329633106679,
      "learning_rate": 4.000960956107167e-06,
      "loss": 0.31880176067352295,
      "num_input_tokens_seen": 274387424,
      "step": 16755,
      "train_runtime": 32263.7961,
      "train_tokens_per_second": 8504.499
    },
    {
      "epoch": 1.1284675464583895,
      "grad_norm": 1.0179429900665844,
      "learning_rate": 3.998370331391881e-06,
      "loss": 0.2737745761871338,
      "num_input_tokens_seen": 274469344,
      "step": 16760,
      "train_runtime": 32273.3879,
      "train_tokens_per_second": 8504.51
    },
    {
      "epoch": 1.1288042014543496,
      "grad_norm": 0.9448755548694692,
      "learning_rate": 3.995779986779139e-06,
      "loss": 0.3029641628265381,
      "num_input_tokens_seen": 274551264,
      "step": 16765,
      "train_runtime": 32282.9411,
      "train_tokens_per_second": 8504.531
    },
    {
      "epoch": 1.1291408564503098,
      "grad_norm": 1.0352317140639957,
      "learning_rate": 3.9931899229933204e-06,
      "loss": 0.2878798723220825,
      "num_input_tokens_seen": 274633184,
      "step": 16770,
      "train_runtime": 32292.5225,
      "train_tokens_per_second": 8504.544
    },
    {
      "epoch": 1.1294775114462698,
      "grad_norm": 0.9639326070575372,
      "learning_rate": 3.990600140758731e-06,
      "loss": 0.28133468627929686,
      "num_input_tokens_seen": 274715104,
      "step": 16775,
      "train_runtime": 32302.1451,
      "train_tokens_per_second": 8504.547
    },
    {
      "epoch": 1.12981416644223,
      "grad_norm": 1.0104635352468754,
      "learning_rate": 3.988010640799591e-06,
      "loss": 0.2915398120880127,
      "num_input_tokens_seen": 274797024,
      "step": 16780,
      "train_runtime": 32311.7007,
      "train_tokens_per_second": 8504.567
    },
    {
      "epoch": 1.1301508214381901,
      "grad_norm": 1.0499815878592103,
      "learning_rate": 3.985421423840051e-06,
      "loss": 0.3211040735244751,
      "num_input_tokens_seen": 274878944,
      "step": 16785,
      "train_runtime": 32321.2792,
      "train_tokens_per_second": 8504.581
    },
    {
      "epoch": 1.1304874764341504,
      "grad_norm": 1.179507415371332,
      "learning_rate": 3.982832490604173e-06,
      "loss": 0.28603343963623046,
      "num_input_tokens_seen": 274960864,
      "step": 16790,
      "train_runtime": 32330.8948,
      "train_tokens_per_second": 8504.586
    },
    {
      "epoch": 1.1308241314301104,
      "grad_norm": 1.3680666384571016,
      "learning_rate": 3.9802438418159465e-06,
      "loss": 0.2966575622558594,
      "num_input_tokens_seen": 275042784,
      "step": 16795,
      "train_runtime": 32340.5066,
      "train_tokens_per_second": 8504.591
    },
    {
      "epoch": 1.1311607864260707,
      "grad_norm": 1.0610329161164096,
      "learning_rate": 3.977655478199275e-06,
      "loss": 0.31412353515625,
      "num_input_tokens_seen": 275124704,
      "step": 16800,
      "train_runtime": 32350.0058,
      "train_tokens_per_second": 8504.626
    },
    {
      "epoch": 1.1314974414220307,
      "grad_norm": 1.0425644802739125,
      "learning_rate": 3.975067400477993e-06,
      "loss": 0.3232024908065796,
      "num_input_tokens_seen": 275205376,
      "step": 16805,
      "train_runtime": 32360.2811,
      "train_tokens_per_second": 8504.419
    },
    {
      "epoch": 1.131834096417991,
      "grad_norm": 1.191024729164976,
      "learning_rate": 3.972479609375842e-06,
      "loss": 0.2990230321884155,
      "num_input_tokens_seen": 275287296,
      "step": 16810,
      "train_runtime": 32369.8913,
      "train_tokens_per_second": 8504.424
    },
    {
      "epoch": 1.132170751413951,
      "grad_norm": 1.1284550289841955,
      "learning_rate": 3.969892105616492e-06,
      "loss": 0.3032519340515137,
      "num_input_tokens_seen": 275369216,
      "step": 16815,
      "train_runtime": 32379.4038,
      "train_tokens_per_second": 8504.456
    },
    {
      "epoch": 1.1325074064099112,
      "grad_norm": 1.1028593340837913,
      "learning_rate": 3.967304889923529e-06,
      "loss": 0.296938157081604,
      "num_input_tokens_seen": 275451136,
      "step": 16820,
      "train_runtime": 32389.0513,
      "train_tokens_per_second": 8504.452
    },
    {
      "epoch": 1.1328440614058712,
      "grad_norm": 1.106440181688713,
      "learning_rate": 3.964717963020463e-06,
      "loss": 0.29951369762420654,
      "num_input_tokens_seen": 275532656,
      "step": 16825,
      "train_runtime": 32399.0373,
      "train_tokens_per_second": 8504.347
    },
    {
      "epoch": 1.1331807164018315,
      "grad_norm": 1.18785335760852,
      "learning_rate": 3.962131325630715e-06,
      "loss": 0.28385961055755615,
      "num_input_tokens_seen": 275614576,
      "step": 16830,
      "train_runtime": 32408.591,
      "train_tokens_per_second": 8504.368
    },
    {
      "epoch": 1.1335173713977915,
      "grad_norm": 1.2145755141663455,
      "learning_rate": 3.959544978477634e-06,
      "loss": 0.31904067993164065,
      "num_input_tokens_seen": 275696496,
      "step": 16835,
      "train_runtime": 32418.1612,
      "train_tokens_per_second": 8504.384
    },
    {
      "epoch": 1.1338540263937518,
      "grad_norm": 0.9486510294363989,
      "learning_rate": 3.9569589222844805e-06,
      "loss": 0.2854208469390869,
      "num_input_tokens_seen": 275778416,
      "step": 16840,
      "train_runtime": 32427.6979,
      "train_tokens_per_second": 8504.409
    },
    {
      "epoch": 1.1341906813897118,
      "grad_norm": 1.0019940989241964,
      "learning_rate": 3.954373157774439e-06,
      "loss": 0.33161287307739257,
      "num_input_tokens_seen": 275860336,
      "step": 16845,
      "train_runtime": 32437.3169,
      "train_tokens_per_second": 8504.413
    },
    {
      "epoch": 1.134527336385672,
      "grad_norm": 1.0915368332231714,
      "learning_rate": 3.951787685670609e-06,
      "loss": 0.29764952659606936,
      "num_input_tokens_seen": 275942256,
      "step": 16850,
      "train_runtime": 32446.8227,
      "train_tokens_per_second": 8504.446
    },
    {
      "epoch": 1.134863991381632,
      "grad_norm": 1.0589599016597495,
      "learning_rate": 3.949202506696012e-06,
      "loss": 0.28556735515594484,
      "num_input_tokens_seen": 276024176,
      "step": 16855,
      "train_runtime": 32456.4408,
      "train_tokens_per_second": 8504.45
    },
    {
      "epoch": 1.1352006463775923,
      "grad_norm": 1.080712270571883,
      "learning_rate": 3.946617621573581e-06,
      "loss": 0.29979307651519777,
      "num_input_tokens_seen": 276106096,
      "step": 16860,
      "train_runtime": 32466.0452,
      "train_tokens_per_second": 8504.457
    },
    {
      "epoch": 1.1355373013735524,
      "grad_norm": 0.9692424319816458,
      "learning_rate": 3.944033031026175e-06,
      "loss": 0.27134392261505125,
      "num_input_tokens_seen": 276187696,
      "step": 16865,
      "train_runtime": 32475.9967,
      "train_tokens_per_second": 8504.364
    },
    {
      "epoch": 1.1358739563695126,
      "grad_norm": 1.2353893722366116,
      "learning_rate": 3.941448735776563e-06,
      "loss": 0.2746992826461792,
      "num_input_tokens_seen": 276269616,
      "step": 16870,
      "train_runtime": 32485.5539,
      "train_tokens_per_second": 8504.384
    },
    {
      "epoch": 1.1362106113654726,
      "grad_norm": 1.0309505632318494,
      "learning_rate": 3.938864736547439e-06,
      "loss": 0.30948772430419924,
      "num_input_tokens_seen": 276351536,
      "step": 16875,
      "train_runtime": 32495.1218,
      "train_tokens_per_second": 8504.401
    },
    {
      "epoch": 1.1365472663614329,
      "grad_norm": 0.8995085129603064,
      "learning_rate": 3.936281034061405e-06,
      "loss": 0.28262507915496826,
      "num_input_tokens_seen": 276433456,
      "step": 16880,
      "train_runtime": 32504.6314,
      "train_tokens_per_second": 8504.433
    },
    {
      "epoch": 1.136883921357393,
      "grad_norm": 1.150773299900184,
      "learning_rate": 3.93369762904099e-06,
      "loss": 0.3212784767150879,
      "num_input_tokens_seen": 276515376,
      "step": 16885,
      "train_runtime": 32514.1994,
      "train_tokens_per_second": 8504.45
    },
    {
      "epoch": 1.1372205763533532,
      "grad_norm": 1.0479119486892439,
      "learning_rate": 3.931114522208631e-06,
      "loss": 0.2872127294540405,
      "num_input_tokens_seen": 276597296,
      "step": 16890,
      "train_runtime": 32523.8122,
      "train_tokens_per_second": 8504.455
    },
    {
      "epoch": 1.1375572313493132,
      "grad_norm": 1.0551117412509115,
      "learning_rate": 3.928531714286689e-06,
      "loss": 0.28008179664611815,
      "num_input_tokens_seen": 276679216,
      "step": 16895,
      "train_runtime": 32533.418,
      "train_tokens_per_second": 8504.462
    },
    {
      "epoch": 1.1378938863452734,
      "grad_norm": 1.138779908676609,
      "learning_rate": 3.925949205997434e-06,
      "loss": 0.2973592519760132,
      "num_input_tokens_seen": 276761136,
      "step": 16900,
      "train_runtime": 32542.9232,
      "train_tokens_per_second": 8504.495
    },
    {
      "epoch": 1.1382305413412335,
      "grad_norm": 0.9414355190728608,
      "learning_rate": 3.923366998063062e-06,
      "loss": 0.2886681318283081,
      "num_input_tokens_seen": 276843056,
      "step": 16905,
      "train_runtime": 32552.5662,
      "train_tokens_per_second": 8504.493
    },
    {
      "epoch": 1.1385671963371937,
      "grad_norm": 1.1074606002917835,
      "learning_rate": 3.920785091205674e-06,
      "loss": 0.2793721675872803,
      "num_input_tokens_seen": 276924976,
      "step": 16910,
      "train_runtime": 32562.1176,
      "train_tokens_per_second": 8504.514
    },
    {
      "epoch": 1.1389038513331537,
      "grad_norm": 1.24063131263779,
      "learning_rate": 3.918203486147294e-06,
      "loss": 0.35880389213562014,
      "num_input_tokens_seen": 277006896,
      "step": 16915,
      "train_runtime": 32571.8559,
      "train_tokens_per_second": 8504.486
    },
    {
      "epoch": 1.139240506329114,
      "grad_norm": 1.3592422192004552,
      "learning_rate": 3.9156221836098594e-06,
      "loss": 0.3175477981567383,
      "num_input_tokens_seen": 277088816,
      "step": 16920,
      "train_runtime": 32581.4878,
      "train_tokens_per_second": 8504.486
    },
    {
      "epoch": 1.139577161325074,
      "grad_norm": 1.2504086728509234,
      "learning_rate": 3.913041184315224e-06,
      "loss": 0.2922720193862915,
      "num_input_tokens_seen": 277170736,
      "step": 16925,
      "train_runtime": 32591.0982,
      "train_tokens_per_second": 8504.492
    },
    {
      "epoch": 1.1399138163210343,
      "grad_norm": 1.0909328220483991,
      "learning_rate": 3.910460488985154e-06,
      "loss": 0.29398312568664553,
      "num_input_tokens_seen": 277252656,
      "step": 16930,
      "train_runtime": 32600.6479,
      "train_tokens_per_second": 8504.514
    },
    {
      "epoch": 1.1402504713169943,
      "grad_norm": 1.266339447225454,
      "learning_rate": 3.9078800983413355e-06,
      "loss": 0.30196175575256345,
      "num_input_tokens_seen": 277334576,
      "step": 16935,
      "train_runtime": 32610.2051,
      "train_tokens_per_second": 8504.533
    },
    {
      "epoch": 1.1405871263129546,
      "grad_norm": 1.0117838051027885,
      "learning_rate": 3.905300013105365e-06,
      "loss": 0.27072412967681886,
      "num_input_tokens_seen": 277416496,
      "step": 16940,
      "train_runtime": 32619.8422,
      "train_tokens_per_second": 8504.532
    },
    {
      "epoch": 1.1409237813089146,
      "grad_norm": 1.130821416712874,
      "learning_rate": 3.902720233998754e-06,
      "loss": 0.28881635665893557,
      "num_input_tokens_seen": 277498416,
      "step": 16945,
      "train_runtime": 32629.4154,
      "train_tokens_per_second": 8504.548
    },
    {
      "epoch": 1.1412604363048748,
      "grad_norm": 1.0014099886644057,
      "learning_rate": 3.90014076174293e-06,
      "loss": 0.28891098499298096,
      "num_input_tokens_seen": 277580336,
      "step": 16950,
      "train_runtime": 32638.9412,
      "train_tokens_per_second": 8504.575
    },
    {
      "epoch": 1.1415970913008349,
      "grad_norm": 0.984933795500556,
      "learning_rate": 3.897561597059237e-06,
      "loss": 0.3115857601165771,
      "num_input_tokens_seen": 277662256,
      "step": 16955,
      "train_runtime": 32648.549,
      "train_tokens_per_second": 8504.582
    },
    {
      "epoch": 1.1419337462967951,
      "grad_norm": 1.0535262287858724,
      "learning_rate": 3.894982740668927e-06,
      "loss": 0.27950048446655273,
      "num_input_tokens_seen": 277744176,
      "step": 16960,
      "train_runtime": 32658.1803,
      "train_tokens_per_second": 8504.582
    },
    {
      "epoch": 1.1422704012927551,
      "grad_norm": 1.2415376339268123,
      "learning_rate": 3.89240419329317e-06,
      "loss": 0.3361311435699463,
      "num_input_tokens_seen": 277826096,
      "step": 16965,
      "train_runtime": 32668.0071,
      "train_tokens_per_second": 8504.532
    },
    {
      "epoch": 1.1426070562887154,
      "grad_norm": 1.0703395830614237,
      "learning_rate": 3.889825955653046e-06,
      "loss": 0.29104540348052976,
      "num_input_tokens_seen": 277908016,
      "step": 16970,
      "train_runtime": 32677.634,
      "train_tokens_per_second": 8504.533
    },
    {
      "epoch": 1.1429437112846754,
      "grad_norm": 1.2853368733378927,
      "learning_rate": 3.887248028469558e-06,
      "loss": 0.284702205657959,
      "num_input_tokens_seen": 277989936,
      "step": 16975,
      "train_runtime": 32687.2756,
      "train_tokens_per_second": 8504.531
    },
    {
      "epoch": 1.1432803662806357,
      "grad_norm": 1.011994836188198,
      "learning_rate": 3.884670412463607e-06,
      "loss": 0.28913178443908694,
      "num_input_tokens_seen": 278071856,
      "step": 16980,
      "train_runtime": 32696.8014,
      "train_tokens_per_second": 8504.558
    },
    {
      "epoch": 1.1436170212765957,
      "grad_norm": 1.1945038822582505,
      "learning_rate": 3.8820931083560194e-06,
      "loss": 0.32308804988861084,
      "num_input_tokens_seen": 278153776,
      "step": 16985,
      "train_runtime": 32706.3985,
      "train_tokens_per_second": 8504.568
    },
    {
      "epoch": 1.143953676272556,
      "grad_norm": 1.140267210771112,
      "learning_rate": 3.879516116867531e-06,
      "loss": 0.26616194248199465,
      "num_input_tokens_seen": 278235696,
      "step": 16990,
      "train_runtime": 32715.9507,
      "train_tokens_per_second": 8504.588
    },
    {
      "epoch": 1.144290331268516,
      "grad_norm": 1.2236709294190662,
      "learning_rate": 3.876939438718786e-06,
      "loss": 0.30200991630554197,
      "num_input_tokens_seen": 278317616,
      "step": 16995,
      "train_runtime": 32725.6427,
      "train_tokens_per_second": 8504.573
    },
    {
      "epoch": 1.1446269862644762,
      "grad_norm": 0.9838872723071032,
      "learning_rate": 3.874363074630345e-06,
      "loss": 0.2790166616439819,
      "num_input_tokens_seen": 278399536,
      "step": 17000,
      "train_runtime": 32735.197,
      "train_tokens_per_second": 8504.593
    },
    {
      "epoch": 1.1449636412604363,
      "grad_norm": 1.1592282549510593,
      "learning_rate": 3.871787025322681e-06,
      "loss": 0.2872983455657959,
      "num_input_tokens_seen": 278481456,
      "step": 17005,
      "train_runtime": 32744.8865,
      "train_tokens_per_second": 8504.578
    },
    {
      "epoch": 1.1453002962563965,
      "grad_norm": 1.0452644534137951,
      "learning_rate": 3.86921129151618e-06,
      "loss": 0.3128769636154175,
      "num_input_tokens_seen": 278563376,
      "step": 17010,
      "train_runtime": 32754.403,
      "train_tokens_per_second": 8504.609
    },
    {
      "epoch": 1.1456369512523565,
      "grad_norm": 1.3590282860892826,
      "learning_rate": 3.866635873931133e-06,
      "loss": 0.31491594314575194,
      "num_input_tokens_seen": 278645296,
      "step": 17015,
      "train_runtime": 32764.0677,
      "train_tokens_per_second": 8504.6
    },
    {
      "epoch": 1.1459736062483168,
      "grad_norm": 1.1683159175255655,
      "learning_rate": 3.864060773287751e-06,
      "loss": 0.31587886810302734,
      "num_input_tokens_seen": 278726560,
      "step": 17020,
      "train_runtime": 32773.9168,
      "train_tokens_per_second": 8504.524
    },
    {
      "epoch": 1.1463102612442768,
      "grad_norm": 0.9845984898407623,
      "learning_rate": 3.861485990306151e-06,
      "loss": 0.2836310863494873,
      "num_input_tokens_seen": 278808480,
      "step": 17025,
      "train_runtime": 32783.6413,
      "train_tokens_per_second": 8504.5
    },
    {
      "epoch": 1.146646916240237,
      "grad_norm": 1.0994484789775294,
      "learning_rate": 3.858911525706365e-06,
      "loss": 0.28117642402648924,
      "num_input_tokens_seen": 278890400,
      "step": 17030,
      "train_runtime": 32793.2753,
      "train_tokens_per_second": 8504.5
    },
    {
      "epoch": 1.146983571236197,
      "grad_norm": 1.2256018992496853,
      "learning_rate": 3.85633738020833e-06,
      "loss": 0.3281055688858032,
      "num_input_tokens_seen": 278972320,
      "step": 17035,
      "train_runtime": 32802.9391,
      "train_tokens_per_second": 8504.492
    },
    {
      "epoch": 1.1473202262321573,
      "grad_norm": 1.1246319684188792,
      "learning_rate": 3.853763554531902e-06,
      "loss": 0.2881981372833252,
      "num_input_tokens_seen": 279054240,
      "step": 17040,
      "train_runtime": 32812.6272,
      "train_tokens_per_second": 8504.477
    },
    {
      "epoch": 1.1476568812281174,
      "grad_norm": 1.1599137822748642,
      "learning_rate": 3.851190049396839e-06,
      "loss": 0.2714566230773926,
      "num_input_tokens_seen": 279136160,
      "step": 17045,
      "train_runtime": 32822.2891,
      "train_tokens_per_second": 8504.47
    },
    {
      "epoch": 1.1479935362240776,
      "grad_norm": 1.0200185636265369,
      "learning_rate": 3.848616865522815e-06,
      "loss": 0.2978977680206299,
      "num_input_tokens_seen": 279218080,
      "step": 17050,
      "train_runtime": 32832.175,
      "train_tokens_per_second": 8504.404
    },
    {
      "epoch": 1.1483301912200377,
      "grad_norm": 1.0489747365019702,
      "learning_rate": 3.846044003629414e-06,
      "loss": 0.2973723173141479,
      "num_input_tokens_seen": 279300000,
      "step": 17055,
      "train_runtime": 32842.1058,
      "train_tokens_per_second": 8504.327
    },
    {
      "epoch": 1.148666846215998,
      "grad_norm": 0.946367655906018,
      "learning_rate": 3.843471464436128e-06,
      "loss": 0.2708949089050293,
      "num_input_tokens_seen": 279381456,
      "step": 17060,
      "train_runtime": 32852.0863,
      "train_tokens_per_second": 8504.223
    },
    {
      "epoch": 1.149003501211958,
      "grad_norm": 1.1310412438857462,
      "learning_rate": 3.840899248662358e-06,
      "loss": 0.28795781135559084,
      "num_input_tokens_seen": 279463376,
      "step": 17065,
      "train_runtime": 32861.668,
      "train_tokens_per_second": 8504.236
    },
    {
      "epoch": 1.1493401562079182,
      "grad_norm": 1.2885190911471507,
      "learning_rate": 3.838327357027419e-06,
      "loss": 0.3156806230545044,
      "num_input_tokens_seen": 279545296,
      "step": 17070,
      "train_runtime": 32871.3784,
      "train_tokens_per_second": 8504.216
    },
    {
      "epoch": 1.1496768112038782,
      "grad_norm": 1.2003858633127218,
      "learning_rate": 3.835755790250528e-06,
      "loss": 0.2914268493652344,
      "num_input_tokens_seen": 279627216,
      "step": 17075,
      "train_runtime": 32880.901,
      "train_tokens_per_second": 8504.244
    },
    {
      "epoch": 1.1500134661998385,
      "grad_norm": 0.8178137739781051,
      "learning_rate": 3.83318454905082e-06,
      "loss": 0.26856274604797364,
      "num_input_tokens_seen": 279709136,
      "step": 17080,
      "train_runtime": 32890.4759,
      "train_tokens_per_second": 8504.259
    },
    {
      "epoch": 1.1503501211957985,
      "grad_norm": 1.0673272498921849,
      "learning_rate": 3.830613634147331e-06,
      "loss": 0.29312644004821775,
      "num_input_tokens_seen": 279791056,
      "step": 17085,
      "train_runtime": 32900.0848,
      "train_tokens_per_second": 8504.266
    },
    {
      "epoch": 1.1506867761917587,
      "grad_norm": 1.032047202744455,
      "learning_rate": 3.82804304625901e-06,
      "loss": 0.316084623336792,
      "num_input_tokens_seen": 279872976,
      "step": 17090,
      "train_runtime": 32909.5941,
      "train_tokens_per_second": 8504.297
    },
    {
      "epoch": 1.1510234311877188,
      "grad_norm": 1.1275134462527587,
      "learning_rate": 3.825472786104715e-06,
      "loss": 0.29494571685791016,
      "num_input_tokens_seen": 279954896,
      "step": 17095,
      "train_runtime": 32919.215,
      "train_tokens_per_second": 8504.3
    },
    {
      "epoch": 1.151360086183679,
      "grad_norm": 1.1337621031689626,
      "learning_rate": 3.8229028544032095e-06,
      "loss": 0.31159367561340334,
      "num_input_tokens_seen": 280036816,
      "step": 17100,
      "train_runtime": 32928.7228,
      "train_tokens_per_second": 8504.333
    },
    {
      "epoch": 1.151696741179639,
      "grad_norm": 1.3556224398245793,
      "learning_rate": 3.8203332518731666e-06,
      "loss": 0.30308876037597654,
      "num_input_tokens_seen": 280118736,
      "step": 17105,
      "train_runtime": 32938.2234,
      "train_tokens_per_second": 8504.367
    },
    {
      "epoch": 1.1520333961755993,
      "grad_norm": 1.1764096028098607,
      "learning_rate": 3.8177639792331695e-06,
      "loss": 0.28809094429016113,
      "num_input_tokens_seen": 280200304,
      "step": 17110,
      "train_runtime": 32947.7288,
      "train_tokens_per_second": 8504.389
    },
    {
      "epoch": 1.1523700511715593,
      "grad_norm": 1.2212724497049674,
      "learning_rate": 3.815195037201704e-06,
      "loss": 0.31122004985809326,
      "num_input_tokens_seen": 280282224,
      "step": 17115,
      "train_runtime": 32957.2322,
      "train_tokens_per_second": 8504.422
    },
    {
      "epoch": 1.1527067061675196,
      "grad_norm": 1.259763958406621,
      "learning_rate": 3.8126264264971694e-06,
      "loss": 0.31268575191497805,
      "num_input_tokens_seen": 280364144,
      "step": 17120,
      "train_runtime": 32966.7412,
      "train_tokens_per_second": 8504.454
    },
    {
      "epoch": 1.1530433611634796,
      "grad_norm": 1.0881490221157,
      "learning_rate": 3.8100581478378673e-06,
      "loss": 0.2689759969711304,
      "num_input_tokens_seen": 280446064,
      "step": 17125,
      "train_runtime": 32976.2516,
      "train_tokens_per_second": 8504.486
    },
    {
      "epoch": 1.1533800161594399,
      "grad_norm": 1.1136311834429138,
      "learning_rate": 3.807490201942012e-06,
      "loss": 0.2656580448150635,
      "num_input_tokens_seen": 280527984,
      "step": 17130,
      "train_runtime": 32985.811,
      "train_tokens_per_second": 8504.505
    },
    {
      "epoch": 1.1537166711553999,
      "grad_norm": 1.189407134281799,
      "learning_rate": 3.804922589527717e-06,
      "loss": 0.2881657123565674,
      "num_input_tokens_seen": 280609904,
      "step": 17135,
      "train_runtime": 32995.4364,
      "train_tokens_per_second": 8504.507
    },
    {
      "epoch": 1.1540533261513601,
      "grad_norm": 1.325972687997251,
      "learning_rate": 3.8023553113130096e-06,
      "loss": 0.3271569013595581,
      "num_input_tokens_seen": 280691824,
      "step": 17140,
      "train_runtime": 33005.0408,
      "train_tokens_per_second": 8504.514
    },
    {
      "epoch": 1.1543899811473202,
      "grad_norm": 1.2097317456090793,
      "learning_rate": 3.799788368015821e-06,
      "loss": 0.281634783744812,
      "num_input_tokens_seen": 280773744,
      "step": 17145,
      "train_runtime": 33014.5484,
      "train_tokens_per_second": 8504.546
    },
    {
      "epoch": 1.1547266361432804,
      "grad_norm": 1.0039338426597098,
      "learning_rate": 3.7972217603539873e-06,
      "loss": 0.2977518320083618,
      "num_input_tokens_seen": 280855664,
      "step": 17150,
      "train_runtime": 33024.0493,
      "train_tokens_per_second": 8504.58
    },
    {
      "epoch": 1.1550632911392404,
      "grad_norm": 1.281882884056456,
      "learning_rate": 3.7946554890452524e-06,
      "loss": 0.2910843133926392,
      "num_input_tokens_seen": 280937584,
      "step": 17155,
      "train_runtime": 33033.5775,
      "train_tokens_per_second": 8504.607
    },
    {
      "epoch": 1.1553999461352007,
      "grad_norm": 1.1871855250252894,
      "learning_rate": 3.792089554807269e-06,
      "loss": 0.3069218635559082,
      "num_input_tokens_seen": 281019504,
      "step": 17160,
      "train_runtime": 33043.0859,
      "train_tokens_per_second": 8504.639
    },
    {
      "epoch": 1.1557366011311607,
      "grad_norm": 1.1818818060925562,
      "learning_rate": 3.789523958357587e-06,
      "loss": 0.2988304615020752,
      "num_input_tokens_seen": 281101424,
      "step": 17165,
      "train_runtime": 33052.7345,
      "train_tokens_per_second": 8504.634
    },
    {
      "epoch": 1.156073256127121,
      "grad_norm": 1.0803676758527136,
      "learning_rate": 3.7869587004136714e-06,
      "loss": 0.2774320602416992,
      "num_input_tokens_seen": 281183344,
      "step": 17170,
      "train_runtime": 33062.3576,
      "train_tokens_per_second": 8504.637
    },
    {
      "epoch": 1.156409911123081,
      "grad_norm": 1.0829928079308684,
      "learning_rate": 3.7843937816928864e-06,
      "loss": 0.2838351488113403,
      "num_input_tokens_seen": 281265264,
      "step": 17175,
      "train_runtime": 33071.9078,
      "train_tokens_per_second": 8504.658
    },
    {
      "epoch": 1.1567465661190413,
      "grad_norm": 1.1915826236037446,
      "learning_rate": 3.781829202912506e-06,
      "loss": 0.30068340301513674,
      "num_input_tokens_seen": 281347184,
      "step": 17180,
      "train_runtime": 33081.4844,
      "train_tokens_per_second": 8504.672
    },
    {
      "epoch": 1.1570832211150013,
      "grad_norm": 1.0685388883799691,
      "learning_rate": 3.7792649647897023e-06,
      "loss": 0.2676640510559082,
      "num_input_tokens_seen": 281429104,
      "step": 17185,
      "train_runtime": 33091.0955,
      "train_tokens_per_second": 8504.678
    },
    {
      "epoch": 1.1574198761109615,
      "grad_norm": 1.0395570989767775,
      "learning_rate": 3.7767010680415595e-06,
      "loss": 0.2702397346496582,
      "num_input_tokens_seen": 281511024,
      "step": 17190,
      "train_runtime": 33100.6018,
      "train_tokens_per_second": 8504.71
    },
    {
      "epoch": 1.1577565311069216,
      "grad_norm": 1.0864232919066752,
      "learning_rate": 3.774137513385061e-06,
      "loss": 0.2835197925567627,
      "num_input_tokens_seen": 281592944,
      "step": 17195,
      "train_runtime": 33110.106,
      "train_tokens_per_second": 8504.743
    },
    {
      "epoch": 1.1580931861028818,
      "grad_norm": 1.0477275223700364,
      "learning_rate": 3.7715743015371005e-06,
      "loss": 0.2944310665130615,
      "num_input_tokens_seen": 281674864,
      "step": 17200,
      "train_runtime": 33119.732,
      "train_tokens_per_second": 8504.745
    },
    {
      "epoch": 1.1584298410988418,
      "grad_norm": 1.06572478354561,
      "learning_rate": 3.769011433214467e-06,
      "loss": 0.27936749458312987,
      "num_input_tokens_seen": 281756784,
      "step": 17205,
      "train_runtime": 33129.3524,
      "train_tokens_per_second": 8504.748
    },
    {
      "epoch": 1.158766496094802,
      "grad_norm": 1.137345517200276,
      "learning_rate": 3.766448909133863e-06,
      "loss": 0.316347074508667,
      "num_input_tokens_seen": 281838704,
      "step": 17210,
      "train_runtime": 33138.8933,
      "train_tokens_per_second": 8504.771
    },
    {
      "epoch": 1.1591031510907621,
      "grad_norm": 1.154788535203004,
      "learning_rate": 3.7638867300118864e-06,
      "loss": 0.27007982730865476,
      "num_input_tokens_seen": 281920624,
      "step": 17215,
      "train_runtime": 33148.4603,
      "train_tokens_per_second": 8504.788
    },
    {
      "epoch": 1.1594398060867224,
      "grad_norm": 1.1154970630577874,
      "learning_rate": 3.7613248965650447e-06,
      "loss": 0.26854848861694336,
      "num_input_tokens_seen": 282002544,
      "step": 17220,
      "train_runtime": 33157.9798,
      "train_tokens_per_second": 8504.817
    },
    {
      "epoch": 1.1597764610826824,
      "grad_norm": 0.9482863564904987,
      "learning_rate": 3.758763409509745e-06,
      "loss": 0.27627608776092527,
      "num_input_tokens_seen": 282084464,
      "step": 17225,
      "train_runtime": 33167.6052,
      "train_tokens_per_second": 8504.819
    },
    {
      "epoch": 1.1601131160786426,
      "grad_norm": 1.370457249640007,
      "learning_rate": 3.7562022695623013e-06,
      "loss": 0.3225153684616089,
      "num_input_tokens_seen": 282166384,
      "step": 17230,
      "train_runtime": 33177.1455,
      "train_tokens_per_second": 8504.842
    },
    {
      "epoch": 1.1604497710746027,
      "grad_norm": 0.9567959565257712,
      "learning_rate": 3.7536414774389252e-06,
      "loss": 0.3076266527175903,
      "num_input_tokens_seen": 282248304,
      "step": 17235,
      "train_runtime": 33186.7085,
      "train_tokens_per_second": 8504.86
    },
    {
      "epoch": 1.160786426070563,
      "grad_norm": 1.2719029431537698,
      "learning_rate": 3.7510810338557363e-06,
      "loss": 0.2906172275543213,
      "num_input_tokens_seen": 282330224,
      "step": 17240,
      "train_runtime": 33196.307,
      "train_tokens_per_second": 8504.868
    },
    {
      "epoch": 1.161123081066523,
      "grad_norm": 1.0766507632029805,
      "learning_rate": 3.748520939528753e-06,
      "loss": 0.2787693977355957,
      "num_input_tokens_seen": 282412144,
      "step": 17245,
      "train_runtime": 33205.9298,
      "train_tokens_per_second": 8504.871
    },
    {
      "epoch": 1.1614597360624832,
      "grad_norm": 1.2671159387619606,
      "learning_rate": 3.7459611951739007e-06,
      "loss": 0.2980589628219604,
      "num_input_tokens_seen": 282494064,
      "step": 17250,
      "train_runtime": 33215.4832,
      "train_tokens_per_second": 8504.891
    },
    {
      "epoch": 1.1617963910584432,
      "grad_norm": 1.2087940445215566,
      "learning_rate": 3.7434018015069984e-06,
      "loss": 0.3100168704986572,
      "num_input_tokens_seen": 282575984,
      "step": 17255,
      "train_runtime": 33225.0731,
      "train_tokens_per_second": 8504.902
    },
    {
      "epoch": 1.1621330460544035,
      "grad_norm": 1.234132079680918,
      "learning_rate": 3.740842759243778e-06,
      "loss": 0.28036036491394045,
      "num_input_tokens_seen": 282657904,
      "step": 17260,
      "train_runtime": 33234.6905,
      "train_tokens_per_second": 8504.906
    },
    {
      "epoch": 1.1624697010503635,
      "grad_norm": 1.0624166039953233,
      "learning_rate": 3.738284069099862e-06,
      "loss": 0.25705971717834475,
      "num_input_tokens_seen": 282739824,
      "step": 17265,
      "train_runtime": 33244.31,
      "train_tokens_per_second": 8504.909
    },
    {
      "epoch": 1.1628063560463238,
      "grad_norm": 0.9820740159870744,
      "learning_rate": 3.735725731790785e-06,
      "loss": 0.2645743370056152,
      "num_input_tokens_seen": 282821744,
      "step": 17270,
      "train_runtime": 33253.813,
      "train_tokens_per_second": 8504.942
    },
    {
      "epoch": 1.1631430110422838,
      "grad_norm": 1.1311851508372377,
      "learning_rate": 3.733167748031974e-06,
      "loss": 0.2976576328277588,
      "num_input_tokens_seen": 282903664,
      "step": 17275,
      "train_runtime": 33263.4409,
      "train_tokens_per_second": 8504.943
    },
    {
      "epoch": 1.163479666038244,
      "grad_norm": 1.2152459222888514,
      "learning_rate": 3.7306101185387646e-06,
      "loss": 0.2660788059234619,
      "num_input_tokens_seen": 282985584,
      "step": 17280,
      "train_runtime": 33273.0544,
      "train_tokens_per_second": 8504.948
    },
    {
      "epoch": 1.163816321034204,
      "grad_norm": 0.8205456060038785,
      "learning_rate": 3.728052844026386e-06,
      "loss": 0.28881402015686036,
      "num_input_tokens_seen": 283067504,
      "step": 17285,
      "train_runtime": 33282.666,
      "train_tokens_per_second": 8504.953
    },
    {
      "epoch": 1.1641529760301643,
      "grad_norm": 1.0233117287398885,
      "learning_rate": 3.7254959252099745e-06,
      "loss": 0.277022385597229,
      "num_input_tokens_seen": 283149424,
      "step": 17290,
      "train_runtime": 33292.1698,
      "train_tokens_per_second": 8504.986
    },
    {
      "epoch": 1.1644896310261243,
      "grad_norm": 1.2957930125246957,
      "learning_rate": 3.7229393628045617e-06,
      "loss": 0.2824831485748291,
      "num_input_tokens_seen": 283231344,
      "step": 17295,
      "train_runtime": 33301.7102,
      "train_tokens_per_second": 8505.009
    },
    {
      "epoch": 1.1648262860220846,
      "grad_norm": 1.143690398701633,
      "learning_rate": 3.720383157525087e-06,
      "loss": 0.28919620513916017,
      "num_input_tokens_seen": 283313264,
      "step": 17300,
      "train_runtime": 33311.4275,
      "train_tokens_per_second": 8504.987
    },
    {
      "epoch": 1.1651629410180446,
      "grad_norm": 1.1803675232950934,
      "learning_rate": 3.717827310086379e-06,
      "loss": 0.2570946216583252,
      "num_input_tokens_seen": 283395184,
      "step": 17305,
      "train_runtime": 33321.0949,
      "train_tokens_per_second": 8504.978
    },
    {
      "epoch": 1.1654995960140049,
      "grad_norm": 1.3946993164657742,
      "learning_rate": 3.7152718212031778e-06,
      "loss": 0.28161938190460206,
      "num_input_tokens_seen": 283477104,
      "step": 17310,
      "train_runtime": 33330.7453,
      "train_tokens_per_second": 8504.973
    },
    {
      "epoch": 1.165836251009965,
      "grad_norm": 1.3191326247325046,
      "learning_rate": 3.7127166915901123e-06,
      "loss": 0.31797597408294676,
      "num_input_tokens_seen": 283559024,
      "step": 17315,
      "train_runtime": 33340.4116,
      "train_tokens_per_second": 8504.965
    },
    {
      "epoch": 1.1661729060059252,
      "grad_norm": 1.184248048084017,
      "learning_rate": 3.710161921961721e-06,
      "loss": 0.297136116027832,
      "num_input_tokens_seen": 283640392,
      "step": 17320,
      "train_runtime": 33350.3473,
      "train_tokens_per_second": 8504.871
    },
    {
      "epoch": 1.1665095610018852,
      "grad_norm": 1.1398345729249197,
      "learning_rate": 3.707607513032434e-06,
      "loss": 0.31151347160339354,
      "num_input_tokens_seen": 283722312,
      "step": 17325,
      "train_runtime": 33360.0412,
      "train_tokens_per_second": 8504.855
    },
    {
      "epoch": 1.1668462159978454,
      "grad_norm": 1.1626995660909532,
      "learning_rate": 3.705053465516587e-06,
      "loss": 0.29609112739562987,
      "num_input_tokens_seen": 283804232,
      "step": 17330,
      "train_runtime": 33369.7994,
      "train_tokens_per_second": 8504.823
    },
    {
      "epoch": 1.1671828709938055,
      "grad_norm": 1.1144813127419595,
      "learning_rate": 3.702499780128407e-06,
      "loss": 0.2620687961578369,
      "num_input_tokens_seen": 283886152,
      "step": 17335,
      "train_runtime": 33379.4772,
      "train_tokens_per_second": 8504.811
    },
    {
      "epoch": 1.1675195259897657,
      "grad_norm": 1.139991082738153,
      "learning_rate": 3.6999464575820266e-06,
      "loss": 0.31541297435760496,
      "num_input_tokens_seen": 283968072,
      "step": 17340,
      "train_runtime": 33389.0156,
      "train_tokens_per_second": 8504.835
    },
    {
      "epoch": 1.1678561809857257,
      "grad_norm": 1.201531895223142,
      "learning_rate": 3.697393498591473e-06,
      "loss": 0.30834431648254396,
      "num_input_tokens_seen": 284049992,
      "step": 17345,
      "train_runtime": 33398.8537,
      "train_tokens_per_second": 8504.783
    },
    {
      "epoch": 1.168192835981686,
      "grad_norm": 0.9860826949416679,
      "learning_rate": 3.694840903870676e-06,
      "loss": 0.33098931312561036,
      "num_input_tokens_seen": 284131912,
      "step": 17350,
      "train_runtime": 33408.3599,
      "train_tokens_per_second": 8504.815
    },
    {
      "epoch": 1.168529490977646,
      "grad_norm": 1.1615866271702588,
      "learning_rate": 3.6922886741334564e-06,
      "loss": 0.2776726007461548,
      "num_input_tokens_seen": 284213832,
      "step": 17355,
      "train_runtime": 33417.8709,
      "train_tokens_per_second": 8504.846
    },
    {
      "epoch": 1.1688661459736063,
      "grad_norm": 1.0705101842854226,
      "learning_rate": 3.689736810093541e-06,
      "loss": 0.26390528678894043,
      "num_input_tokens_seen": 284295752,
      "step": 17360,
      "train_runtime": 33427.3825,
      "train_tokens_per_second": 8504.876
    },
    {
      "epoch": 1.1692028009695663,
      "grad_norm": 1.2118095652244467,
      "learning_rate": 3.6871853124645464e-06,
      "loss": 0.29779088497161865,
      "num_input_tokens_seen": 284377672,
      "step": 17365,
      "train_runtime": 33436.9807,
      "train_tokens_per_second": 8504.885
    },
    {
      "epoch": 1.1695394559655266,
      "grad_norm": 1.0677149912047275,
      "learning_rate": 3.684634181959994e-06,
      "loss": 0.3009209156036377,
      "num_input_tokens_seen": 284459592,
      "step": 17370,
      "train_runtime": 33446.4785,
      "train_tokens_per_second": 8504.919
    },
    {
      "epoch": 1.1698761109614866,
      "grad_norm": 1.09442138280953,
      "learning_rate": 3.6820834192932974e-06,
      "loss": 0.3021111249923706,
      "num_input_tokens_seen": 284541032,
      "step": 17375,
      "train_runtime": 33456.3282,
      "train_tokens_per_second": 8504.849
    },
    {
      "epoch": 1.1702127659574468,
      "grad_norm": 1.1837987973376551,
      "learning_rate": 3.679533025177772e-06,
      "loss": 0.3373254776000977,
      "num_input_tokens_seen": 284622952,
      "step": 17380,
      "train_runtime": 33465.9368,
      "train_tokens_per_second": 8504.855
    },
    {
      "epoch": 1.1705494209534069,
      "grad_norm": 1.1329637514408957,
      "learning_rate": 3.6769830003266234e-06,
      "loss": 0.30596342086791994,
      "num_input_tokens_seen": 284704872,
      "step": 17385,
      "train_runtime": 33475.5644,
      "train_tokens_per_second": 8504.857
    },
    {
      "epoch": 1.1708860759493671,
      "grad_norm": 1.0795122540989752,
      "learning_rate": 3.6744333454529607e-06,
      "loss": 0.2955460071563721,
      "num_input_tokens_seen": 284786792,
      "step": 17390,
      "train_runtime": 33485.0647,
      "train_tokens_per_second": 8504.89
    },
    {
      "epoch": 1.1712227309453271,
      "grad_norm": 0.9706971607460226,
      "learning_rate": 3.6718840612697847e-06,
      "loss": 0.2942418098449707,
      "num_input_tokens_seen": 284868712,
      "step": 17395,
      "train_runtime": 33494.6918,
      "train_tokens_per_second": 8504.891
    },
    {
      "epoch": 1.1715593859412874,
      "grad_norm": 1.0322031917402776,
      "learning_rate": 3.6693351484899986e-06,
      "loss": 0.28673219680786133,
      "num_input_tokens_seen": 284950632,
      "step": 17400,
      "train_runtime": 33504.3076,
      "train_tokens_per_second": 8504.895
    },
    {
      "epoch": 1.1718960409372474,
      "grad_norm": 1.1756141724682905,
      "learning_rate": 3.6667866078263937e-06,
      "loss": 0.27104711532592773,
      "num_input_tokens_seen": 285032552,
      "step": 17405,
      "train_runtime": 33513.9166,
      "train_tokens_per_second": 8504.901
    },
    {
      "epoch": 1.1722326959332077,
      "grad_norm": 0.9406332433644942,
      "learning_rate": 3.664238439991663e-06,
      "loss": 0.317548131942749,
      "num_input_tokens_seen": 285114472,
      "step": 17410,
      "train_runtime": 33523.423,
      "train_tokens_per_second": 8504.933
    },
    {
      "epoch": 1.1725693509291677,
      "grad_norm": 0.9459135437134258,
      "learning_rate": 3.6616906456983925e-06,
      "loss": 0.31169734001159666,
      "num_input_tokens_seen": 285196392,
      "step": 17415,
      "train_runtime": 33533.0392,
      "train_tokens_per_second": 8504.937
    },
    {
      "epoch": 1.172906005925128,
      "grad_norm": 1.1780479137846955,
      "learning_rate": 3.659143225659065e-06,
      "loss": 0.30247719287872316,
      "num_input_tokens_seen": 285278312,
      "step": 17420,
      "train_runtime": 33542.6588,
      "train_tokens_per_second": 8504.94
    },
    {
      "epoch": 1.173242660921088,
      "grad_norm": 1.1947746869508262,
      "learning_rate": 3.6565961805860595e-06,
      "loss": 0.3178609848022461,
      "num_input_tokens_seen": 285360232,
      "step": 17425,
      "train_runtime": 33552.2274,
      "train_tokens_per_second": 8504.956
    },
    {
      "epoch": 1.1735793159170482,
      "grad_norm": 1.0550232353133615,
      "learning_rate": 3.654049511191649e-06,
      "loss": 0.27347507476806643,
      "num_input_tokens_seen": 285442152,
      "step": 17430,
      "train_runtime": 33561.727,
      "train_tokens_per_second": 8504.99
    },
    {
      "epoch": 1.1739159709130083,
      "grad_norm": 1.2737981881822487,
      "learning_rate": 3.6515032181879995e-06,
      "loss": 0.29825031757354736,
      "num_input_tokens_seen": 285524072,
      "step": 17435,
      "train_runtime": 33571.3296,
      "train_tokens_per_second": 8504.997
    },
    {
      "epoch": 1.1742526259089685,
      "grad_norm": 0.9876751107031868,
      "learning_rate": 3.6489573022871755e-06,
      "loss": 0.27126684188842776,
      "num_input_tokens_seen": 285605992,
      "step": 17440,
      "train_runtime": 33580.8326,
      "train_tokens_per_second": 8505.03
    },
    {
      "epoch": 1.1745892809049285,
      "grad_norm": 1.180383122229331,
      "learning_rate": 3.6464117642011333e-06,
      "loss": 0.30537829399108884,
      "num_input_tokens_seen": 285687912,
      "step": 17445,
      "train_runtime": 33590.461,
      "train_tokens_per_second": 8505.031
    },
    {
      "epoch": 1.1749259359008888,
      "grad_norm": 1.1843727939884041,
      "learning_rate": 3.643866604641726e-06,
      "loss": 0.3104104518890381,
      "num_input_tokens_seen": 285769832,
      "step": 17450,
      "train_runtime": 33600.052,
      "train_tokens_per_second": 8505.041
    },
    {
      "epoch": 1.1752625908968488,
      "grad_norm": 1.1693508520073845,
      "learning_rate": 3.641321824320697e-06,
      "loss": 0.28719310760498046,
      "num_input_tokens_seen": 285851752,
      "step": 17455,
      "train_runtime": 33609.662,
      "train_tokens_per_second": 8505.047
    },
    {
      "epoch": 1.175599245892809,
      "grad_norm": 1.1261848603792213,
      "learning_rate": 3.6387774239496893e-06,
      "loss": 0.30427374839782717,
      "num_input_tokens_seen": 285933672,
      "step": 17460,
      "train_runtime": 33619.1623,
      "train_tokens_per_second": 8505.08
    },
    {
      "epoch": 1.175935900888769,
      "grad_norm": 1.0828600149092422,
      "learning_rate": 3.6362334042402326e-06,
      "loss": 0.28088109493255614,
      "num_input_tokens_seen": 286015592,
      "step": 17465,
      "train_runtime": 33628.6621,
      "train_tokens_per_second": 8505.114
    },
    {
      "epoch": 1.1762725558847293,
      "grad_norm": 0.9970356787822127,
      "learning_rate": 3.6336897659037563e-06,
      "loss": 0.2558363676071167,
      "num_input_tokens_seen": 286097008,
      "step": 17470,
      "train_runtime": 33638.4886,
      "train_tokens_per_second": 8505.049
    },
    {
      "epoch": 1.1766092108806894,
      "grad_norm": 1.2116986013291635,
      "learning_rate": 3.6311465096515787e-06,
      "loss": 0.26922357082366943,
      "num_input_tokens_seen": 286178928,
      "step": 17475,
      "train_runtime": 33648.0936,
      "train_tokens_per_second": 8505.056
    },
    {
      "epoch": 1.1769458658766496,
      "grad_norm": 0.9911847505573459,
      "learning_rate": 3.6286036361949175e-06,
      "loss": 0.25720791816711425,
      "num_input_tokens_seen": 286260848,
      "step": 17480,
      "train_runtime": 33657.7418,
      "train_tokens_per_second": 8505.052
    },
    {
      "epoch": 1.1772825208726099,
      "grad_norm": 1.0311715205733456,
      "learning_rate": 3.6260611462448736e-06,
      "loss": 0.2549643278121948,
      "num_input_tokens_seen": 286342768,
      "step": 17485,
      "train_runtime": 33667.5481,
      "train_tokens_per_second": 8505.008
    },
    {
      "epoch": 1.17761917586857,
      "grad_norm": 1.4918696949076204,
      "learning_rate": 3.623519040512451e-06,
      "loss": 0.32391350269317626,
      "num_input_tokens_seen": 286424688,
      "step": 17490,
      "train_runtime": 33677.056,
      "train_tokens_per_second": 8505.039
    },
    {
      "epoch": 1.17795583086453,
      "grad_norm": 1.1062613590774595,
      "learning_rate": 3.6209773197085373e-06,
      "loss": 0.26546692848205566,
      "num_input_tokens_seen": 286506608,
      "step": 17495,
      "train_runtime": 33686.71,
      "train_tokens_per_second": 8505.034
    },
    {
      "epoch": 1.1782924858604902,
      "grad_norm": 1.209752522718659,
      "learning_rate": 3.6184359845439214e-06,
      "loss": 0.3284714460372925,
      "num_input_tokens_seen": 286588528,
      "step": 17500,
      "train_runtime": 33696.3325,
      "train_tokens_per_second": 8505.036
    },
    {
      "epoch": 1.1786291408564504,
      "grad_norm": 1.037059042240738,
      "learning_rate": 3.615895035729275e-06,
      "loss": 0.28177971839904786,
      "num_input_tokens_seen": 286670448,
      "step": 17505,
      "train_runtime": 33705.865,
      "train_tokens_per_second": 8505.061
    },
    {
      "epoch": 1.1789657958524105,
      "grad_norm": 0.9374280778956007,
      "learning_rate": 3.6133544739751703e-06,
      "loss": 0.2817511558532715,
      "num_input_tokens_seen": 286752368,
      "step": 17510,
      "train_runtime": 33715.4665,
      "train_tokens_per_second": 8505.069
    },
    {
      "epoch": 1.1793024508483705,
      "grad_norm": 0.9140556205217384,
      "learning_rate": 3.6108142999920633e-06,
      "loss": 0.2786051034927368,
      "num_input_tokens_seen": 286834288,
      "step": 17515,
      "train_runtime": 33725.13,
      "train_tokens_per_second": 8505.061
    },
    {
      "epoch": 1.1796391058443307,
      "grad_norm": 0.9635060271685484,
      "learning_rate": 3.608274514490308e-06,
      "loss": 0.2888925075531006,
      "num_input_tokens_seen": 286916208,
      "step": 17520,
      "train_runtime": 33734.8639,
      "train_tokens_per_second": 8505.035
    },
    {
      "epoch": 1.179975760840291,
      "grad_norm": 1.1215457043264816,
      "learning_rate": 3.605735118180146e-06,
      "loss": 0.2813228368759155,
      "num_input_tokens_seen": 286998128,
      "step": 17525,
      "train_runtime": 33744.53,
      "train_tokens_per_second": 8505.027
    },
    {
      "epoch": 1.180312415836251,
      "grad_norm": 1.004139437359447,
      "learning_rate": 3.6031961117717144e-06,
      "loss": 0.27404792308807374,
      "num_input_tokens_seen": 287080048,
      "step": 17530,
      "train_runtime": 33754.0459,
      "train_tokens_per_second": 8505.056
    },
    {
      "epoch": 1.180649070832211,
      "grad_norm": 1.1927203466619243,
      "learning_rate": 3.600657495975034e-06,
      "loss": 0.30934462547302244,
      "num_input_tokens_seen": 287161968,
      "step": 17535,
      "train_runtime": 33763.6319,
      "train_tokens_per_second": 8505.067
    },
    {
      "epoch": 1.1809857258281713,
      "grad_norm": 1.1549916236774782,
      "learning_rate": 3.598119271500024e-06,
      "loss": 0.29888556003570554,
      "num_input_tokens_seen": 287243888,
      "step": 17540,
      "train_runtime": 33773.2662,
      "train_tokens_per_second": 8505.067
    },
    {
      "epoch": 1.1813223808241315,
      "grad_norm": 1.1251596546318121,
      "learning_rate": 3.5955814390564882e-06,
      "loss": 0.2820159435272217,
      "num_input_tokens_seen": 287325808,
      "step": 17545,
      "train_runtime": 33782.8525,
      "train_tokens_per_second": 8505.078
    },
    {
      "epoch": 1.1816590358200916,
      "grad_norm": 1.2185102012683486,
      "learning_rate": 3.5930439993541265e-06,
      "loss": 0.31166701316833495,
      "num_input_tokens_seen": 287407728,
      "step": 17550,
      "train_runtime": 33792.4791,
      "train_tokens_per_second": 8505.08
    },
    {
      "epoch": 1.1819956908160516,
      "grad_norm": 1.1644284733578563,
      "learning_rate": 3.590506953102522e-06,
      "loss": 0.3300513029098511,
      "num_input_tokens_seen": 287489648,
      "step": 17555,
      "train_runtime": 33802.0312,
      "train_tokens_per_second": 8505.1
    },
    {
      "epoch": 1.1823323458120119,
      "grad_norm": 0.9125025631995084,
      "learning_rate": 3.5879703010111554e-06,
      "loss": 0.25591325759887695,
      "num_input_tokens_seen": 287571568,
      "step": 17560,
      "train_runtime": 33811.5788,
      "train_tokens_per_second": 8505.121
    },
    {
      "epoch": 1.182669000807972,
      "grad_norm": 1.1838655323496643,
      "learning_rate": 3.5854340437893898e-06,
      "loss": 0.27482926845550537,
      "num_input_tokens_seen": 287653488,
      "step": 17565,
      "train_runtime": 33821.1757,
      "train_tokens_per_second": 8505.13
    },
    {
      "epoch": 1.1830056558039321,
      "grad_norm": 1.2559542943470214,
      "learning_rate": 3.5828981821464838e-06,
      "loss": 0.29930500984191893,
      "num_input_tokens_seen": 287735408,
      "step": 17570,
      "train_runtime": 33830.9078,
      "train_tokens_per_second": 8505.105
    },
    {
      "epoch": 1.1833423107998922,
      "grad_norm": 1.369338816411758,
      "learning_rate": 3.580362716791582e-06,
      "loss": 0.33022730350494384,
      "num_input_tokens_seen": 287817328,
      "step": 17575,
      "train_runtime": 33840.4248,
      "train_tokens_per_second": 8505.133
    },
    {
      "epoch": 1.1836789657958524,
      "grad_norm": 1.0397068819796622,
      "learning_rate": 3.5778276484337205e-06,
      "loss": 0.31552653312683104,
      "num_input_tokens_seen": 287899248,
      "step": 17580,
      "train_runtime": 33849.9319,
      "train_tokens_per_second": 8505.165
    },
    {
      "epoch": 1.1840156207918127,
      "grad_norm": 0.9021325258310822,
      "learning_rate": 3.575292977781821e-06,
      "loss": 0.2608304500579834,
      "num_input_tokens_seen": 287981168,
      "step": 17585,
      "train_runtime": 33859.569,
      "train_tokens_per_second": 8505.163
    },
    {
      "epoch": 1.1843522757877727,
      "grad_norm": 1.2546907814411772,
      "learning_rate": 3.5727587055446976e-06,
      "loss": 0.3144099235534668,
      "num_input_tokens_seen": 288063088,
      "step": 17590,
      "train_runtime": 33869.1977,
      "train_tokens_per_second": 8505.164
    },
    {
      "epoch": 1.1846889307837327,
      "grad_norm": 1.0984781546469522,
      "learning_rate": 3.570224832431051e-06,
      "loss": 0.31807329654693606,
      "num_input_tokens_seen": 288145008,
      "step": 17595,
      "train_runtime": 33878.8328,
      "train_tokens_per_second": 8505.163
    },
    {
      "epoch": 1.185025585779693,
      "grad_norm": 1.0691083333278826,
      "learning_rate": 3.5676913591494715e-06,
      "loss": 0.27736620903015136,
      "num_input_tokens_seen": 288226928,
      "step": 17600,
      "train_runtime": 33888.3446,
      "train_tokens_per_second": 8505.193
    },
    {
      "epoch": 1.1853622407756532,
      "grad_norm": 1.0915779615554582,
      "learning_rate": 3.5651582864084345e-06,
      "loss": 0.32399716377258303,
      "num_input_tokens_seen": 288307760,
      "step": 17605,
      "train_runtime": 33897.9986,
      "train_tokens_per_second": 8505.156
    },
    {
      "epoch": 1.1856988957716132,
      "grad_norm": 1.6399199693356725,
      "learning_rate": 3.5626256149163096e-06,
      "loss": 0.2952075719833374,
      "num_input_tokens_seen": 288389680,
      "step": 17610,
      "train_runtime": 33907.6295,
      "train_tokens_per_second": 8505.156
    },
    {
      "epoch": 1.1860355507675733,
      "grad_norm": 1.0959892134661011,
      "learning_rate": 3.5600933453813464e-06,
      "loss": 0.2897643566131592,
      "num_input_tokens_seen": 288471600,
      "step": 17615,
      "train_runtime": 33917.1917,
      "train_tokens_per_second": 8505.173
    },
    {
      "epoch": 1.1863722057635335,
      "grad_norm": 0.9849070845402209,
      "learning_rate": 3.5575614785116886e-06,
      "loss": 0.2678213357925415,
      "num_input_tokens_seen": 288553520,
      "step": 17620,
      "train_runtime": 33926.7003,
      "train_tokens_per_second": 8505.204
    },
    {
      "epoch": 1.1867088607594938,
      "grad_norm": 1.1334066128153224,
      "learning_rate": 3.5550300150153618e-06,
      "loss": 0.3053257942199707,
      "num_input_tokens_seen": 288635440,
      "step": 17625,
      "train_runtime": 33936.2046,
      "train_tokens_per_second": 8505.236
    },
    {
      "epoch": 1.1870455157554538,
      "grad_norm": 1.1655212663438548,
      "learning_rate": 3.5524989556002863e-06,
      "loss": 0.2896293640136719,
      "num_input_tokens_seen": 288717360,
      "step": 17630,
      "train_runtime": 33945.8615,
      "train_tokens_per_second": 8505.23
    },
    {
      "epoch": 1.1873821707514138,
      "grad_norm": 0.9573709570383852,
      "learning_rate": 3.5499683009742604e-06,
      "loss": 0.28679351806640624,
      "num_input_tokens_seen": 288799280,
      "step": 17635,
      "train_runtime": 33955.3589,
      "train_tokens_per_second": 8505.264
    },
    {
      "epoch": 1.187718825747374,
      "grad_norm": 1.1273002054298267,
      "learning_rate": 3.5474380518449757e-06,
      "loss": 0.26717243194580076,
      "num_input_tokens_seen": 288881200,
      "step": 17640,
      "train_runtime": 33965.0589,
      "train_tokens_per_second": 8505.247
    },
    {
      "epoch": 1.1880554807433343,
      "grad_norm": 1.2456876750813095,
      "learning_rate": 3.5449082089200072e-06,
      "loss": 0.2742879390716553,
      "num_input_tokens_seen": 288963120,
      "step": 17645,
      "train_runtime": 33974.567,
      "train_tokens_per_second": 8505.277
    },
    {
      "epoch": 1.1883921357392944,
      "grad_norm": 1.0005982779464473,
      "learning_rate": 3.5423787729068206e-06,
      "loss": 0.33940463066101073,
      "num_input_tokens_seen": 289045040,
      "step": 17650,
      "train_runtime": 33984.1763,
      "train_tokens_per_second": 8505.283
    },
    {
      "epoch": 1.1887287907352544,
      "grad_norm": 0.9692730429828226,
      "learning_rate": 3.539849744512759e-06,
      "loss": 0.3061560392379761,
      "num_input_tokens_seen": 289126960,
      "step": 17655,
      "train_runtime": 33993.7901,
      "train_tokens_per_second": 8505.288
    },
    {
      "epoch": 1.1890654457312146,
      "grad_norm": 1.068765054122536,
      "learning_rate": 3.5373211244450633e-06,
      "loss": 0.27522454261779783,
      "num_input_tokens_seen": 289208880,
      "step": 17660,
      "train_runtime": 34003.299,
      "train_tokens_per_second": 8505.318
    },
    {
      "epoch": 1.189402100727175,
      "grad_norm": 1.132816601942514,
      "learning_rate": 3.5347929134108483e-06,
      "loss": 0.3100718975067139,
      "num_input_tokens_seen": 289290800,
      "step": 17665,
      "train_runtime": 34012.8156,
      "train_tokens_per_second": 8505.347
    },
    {
      "epoch": 1.189738755723135,
      "grad_norm": 1.110172398052259,
      "learning_rate": 3.532265112117125e-06,
      "loss": 0.29237229824066163,
      "num_input_tokens_seen": 289372720,
      "step": 17670,
      "train_runtime": 34022.4436,
      "train_tokens_per_second": 8505.348
    },
    {
      "epoch": 1.190075410719095,
      "grad_norm": 1.0587044363327982,
      "learning_rate": 3.5297377212707805e-06,
      "loss": 0.26635098457336426,
      "num_input_tokens_seen": 289454640,
      "step": 17675,
      "train_runtime": 34032.025,
      "train_tokens_per_second": 8505.36
    },
    {
      "epoch": 1.1904120657150552,
      "grad_norm": 1.1682947065181724,
      "learning_rate": 3.527210741578596e-06,
      "loss": 0.31251220703125,
      "num_input_tokens_seen": 289536560,
      "step": 17680,
      "train_runtime": 34041.6361,
      "train_tokens_per_second": 8505.366
    },
    {
      "epoch": 1.1907487207110155,
      "grad_norm": 1.134221865618371,
      "learning_rate": 3.5246841737472304e-06,
      "loss": 0.30175960063934326,
      "num_input_tokens_seen": 289617528,
      "step": 17685,
      "train_runtime": 34051.3147,
      "train_tokens_per_second": 8505.326
    },
    {
      "epoch": 1.1910853757069755,
      "grad_norm": 1.0993585234780632,
      "learning_rate": 3.522158018483231e-06,
      "loss": 0.26383771896362307,
      "num_input_tokens_seen": 289699448,
      "step": 17690,
      "train_runtime": 34060.9367,
      "train_tokens_per_second": 8505.328
    },
    {
      "epoch": 1.1914220307029355,
      "grad_norm": 1.0077887084130885,
      "learning_rate": 3.519632276493029e-06,
      "loss": 0.29793705940246584,
      "num_input_tokens_seen": 289781368,
      "step": 17695,
      "train_runtime": 34070.5559,
      "train_tokens_per_second": 8505.331
    },
    {
      "epoch": 1.1917586856988958,
      "grad_norm": 1.1399847153216178,
      "learning_rate": 3.5171069484829417e-06,
      "loss": 0.31617653369903564,
      "num_input_tokens_seen": 289863288,
      "step": 17700,
      "train_runtime": 34080.1184,
      "train_tokens_per_second": 8505.349
    },
    {
      "epoch": 1.192095340694856,
      "grad_norm": 1.2168717852127386,
      "learning_rate": 3.5145820351591663e-06,
      "loss": 0.26170222759246825,
      "num_input_tokens_seen": 289945208,
      "step": 17705,
      "train_runtime": 34089.6367,
      "train_tokens_per_second": 8505.377
    },
    {
      "epoch": 1.192431995690816,
      "grad_norm": 1.2317598950441897,
      "learning_rate": 3.51205753722779e-06,
      "loss": 0.32098956108093263,
      "num_input_tokens_seen": 290027128,
      "step": 17710,
      "train_runtime": 34099.2607,
      "train_tokens_per_second": 8505.379
    },
    {
      "epoch": 1.1927686506867763,
      "grad_norm": 1.1852140203090789,
      "learning_rate": 3.5095334553947757e-06,
      "loss": 0.31770806312561034,
      "num_input_tokens_seen": 290109048,
      "step": 17715,
      "train_runtime": 34108.8929,
      "train_tokens_per_second": 8505.379
    },
    {
      "epoch": 1.1931053056827363,
      "grad_norm": 1.005971437100752,
      "learning_rate": 3.50700979036598e-06,
      "loss": 0.27314419746398927,
      "num_input_tokens_seen": 290190968,
      "step": 17720,
      "train_runtime": 34118.555,
      "train_tokens_per_second": 8505.371
    },
    {
      "epoch": 1.1934419606786966,
      "grad_norm": 1.3162770237326626,
      "learning_rate": 3.5044865428471343e-06,
      "loss": 0.31949942111968993,
      "num_input_tokens_seen": 290272888,
      "step": 17725,
      "train_runtime": 34128.2033,
      "train_tokens_per_second": 8505.367
    },
    {
      "epoch": 1.1937786156746566,
      "grad_norm": 1.0168148129250345,
      "learning_rate": 3.50196371354386e-06,
      "loss": 0.31839985847473146,
      "num_input_tokens_seen": 290354808,
      "step": 17730,
      "train_runtime": 34137.7375,
      "train_tokens_per_second": 8505.391
    },
    {
      "epoch": 1.1941152706706168,
      "grad_norm": 1.4338877060509116,
      "learning_rate": 3.499441303161655e-06,
      "loss": 0.32309536933898925,
      "num_input_tokens_seen": 290436728,
      "step": 17735,
      "train_runtime": 34147.3988,
      "train_tokens_per_second": 8505.384
    },
    {
      "epoch": 1.1944519256665769,
      "grad_norm": 0.9764687773101428,
      "learning_rate": 3.4969193124059042e-06,
      "loss": 0.27397737503051756,
      "num_input_tokens_seen": 290518648,
      "step": 17740,
      "train_runtime": 34156.9997,
      "train_tokens_per_second": 8505.391
    },
    {
      "epoch": 1.1947885806625371,
      "grad_norm": 1.0882538469529244,
      "learning_rate": 3.4943977419818748e-06,
      "loss": 0.2787665367126465,
      "num_input_tokens_seen": 290600568,
      "step": 17745,
      "train_runtime": 34166.712,
      "train_tokens_per_second": 8505.371
    },
    {
      "epoch": 1.1951252356584972,
      "grad_norm": 1.0842032824528123,
      "learning_rate": 3.4918765925947173e-06,
      "loss": 0.25905015468597414,
      "num_input_tokens_seen": 290682488,
      "step": 17750,
      "train_runtime": 34176.3361,
      "train_tokens_per_second": 8505.373
    },
    {
      "epoch": 1.1954618906544574,
      "grad_norm": 1.1094815215947411,
      "learning_rate": 3.4893558649494595e-06,
      "loss": 0.29352993965148927,
      "num_input_tokens_seen": 290764408,
      "step": 17755,
      "train_runtime": 34185.9066,
      "train_tokens_per_second": 8505.388
    },
    {
      "epoch": 1.1957985456504174,
      "grad_norm": 1.117839687872536,
      "learning_rate": 3.4868355597510194e-06,
      "loss": 0.2792452335357666,
      "num_input_tokens_seen": 290846328,
      "step": 17760,
      "train_runtime": 34195.491,
      "train_tokens_per_second": 8505.4
    },
    {
      "epoch": 1.1961352006463777,
      "grad_norm": 1.2649309376056161,
      "learning_rate": 3.4843156777041887e-06,
      "loss": 0.2823976993560791,
      "num_input_tokens_seen": 290928248,
      "step": 17765,
      "train_runtime": 34205.1546,
      "train_tokens_per_second": 8505.392
    },
    {
      "epoch": 1.1964718556423377,
      "grad_norm": 1.0847610564687589,
      "learning_rate": 3.4817962195136457e-06,
      "loss": 0.2926262378692627,
      "num_input_tokens_seen": 291010168,
      "step": 17770,
      "train_runtime": 34214.6605,
      "train_tokens_per_second": 8505.423
    },
    {
      "epoch": 1.196808510638298,
      "grad_norm": 1.2784521302607432,
      "learning_rate": 3.479277185883947e-06,
      "loss": 0.2991974115371704,
      "num_input_tokens_seen": 291092088,
      "step": 17775,
      "train_runtime": 34224.276,
      "train_tokens_per_second": 8505.427
    },
    {
      "epoch": 1.197145165634258,
      "grad_norm": 0.9769782696922231,
      "learning_rate": 3.476758577519537e-06,
      "loss": 0.2417271375656128,
      "num_input_tokens_seen": 291174008,
      "step": 17780,
      "train_runtime": 34233.8465,
      "train_tokens_per_second": 8505.442
    },
    {
      "epoch": 1.1974818206302182,
      "grad_norm": 1.3146296906153518,
      "learning_rate": 3.4742403951247318e-06,
      "loss": 0.282707953453064,
      "num_input_tokens_seen": 291255792,
      "step": 17785,
      "train_runtime": 34243.4323,
      "train_tokens_per_second": 8505.45
    },
    {
      "epoch": 1.1978184756261783,
      "grad_norm": 1.0744820198724294,
      "learning_rate": 3.471722639403735e-06,
      "loss": 0.28229196071624757,
      "num_input_tokens_seen": 291337712,
      "step": 17790,
      "train_runtime": 34253.076,
      "train_tokens_per_second": 8505.447
    },
    {
      "epoch": 1.1981551306221385,
      "grad_norm": 1.1740978448826775,
      "learning_rate": 3.4692053110606277e-06,
      "loss": 0.29292755126953124,
      "num_input_tokens_seen": 291419632,
      "step": 17795,
      "train_runtime": 34262.6683,
      "train_tokens_per_second": 8505.456
    },
    {
      "epoch": 1.1984917856180985,
      "grad_norm": 1.1453858362683251,
      "learning_rate": 3.4666884107993766e-06,
      "loss": 0.2882412910461426,
      "num_input_tokens_seen": 291501552,
      "step": 17800,
      "train_runtime": 34272.17,
      "train_tokens_per_second": 8505.489
    },
    {
      "epoch": 1.1988284406140588,
      "grad_norm": 1.0372771558998057,
      "learning_rate": 3.464171939323819e-06,
      "loss": 0.25309321880340574,
      "num_input_tokens_seen": 291583472,
      "step": 17805,
      "train_runtime": 34281.6794,
      "train_tokens_per_second": 8505.519
    },
    {
      "epoch": 1.1991650956100188,
      "grad_norm": 1.1580025254317563,
      "learning_rate": 3.4616558973376848e-06,
      "loss": 0.2919482707977295,
      "num_input_tokens_seen": 291665392,
      "step": 17810,
      "train_runtime": 34291.6926,
      "train_tokens_per_second": 8505.424
    },
    {
      "epoch": 1.199501750605979,
      "grad_norm": 1.1451661965419266,
      "learning_rate": 3.4591402855445697e-06,
      "loss": 0.2955304145812988,
      "num_input_tokens_seen": 291747312,
      "step": 17815,
      "train_runtime": 34301.1975,
      "train_tokens_per_second": 8505.456
    },
    {
      "epoch": 1.199838405601939,
      "grad_norm": 0.976510799182,
      "learning_rate": 3.4566251046479625e-06,
      "loss": 0.2938955783843994,
      "num_input_tokens_seen": 291829232,
      "step": 17820,
      "train_runtime": 34310.7128,
      "train_tokens_per_second": 8505.484
    },
    {
      "epoch": 1.2001750605978994,
      "grad_norm": 1.1282504039178756,
      "learning_rate": 3.454110355351221e-06,
      "loss": 0.30265035629272463,
      "num_input_tokens_seen": 291911152,
      "step": 17825,
      "train_runtime": 34320.2284,
      "train_tokens_per_second": 8505.513
    },
    {
      "epoch": 1.2005117155938594,
      "grad_norm": 1.019180181757055,
      "learning_rate": 3.451596038357593e-06,
      "loss": 0.2859029769897461,
      "num_input_tokens_seen": 291993072,
      "step": 17830,
      "train_runtime": 34329.861,
      "train_tokens_per_second": 8505.513
    },
    {
      "epoch": 1.2008483705898196,
      "grad_norm": 1.0198221898334234,
      "learning_rate": 3.4490821543701924e-06,
      "loss": 0.317336368560791,
      "num_input_tokens_seen": 292074992,
      "step": 17835,
      "train_runtime": 34339.5116,
      "train_tokens_per_second": 8505.508
    },
    {
      "epoch": 1.2011850255857797,
      "grad_norm": 1.1492475034420877,
      "learning_rate": 3.4465687040920227e-06,
      "loss": 0.2818615198135376,
      "num_input_tokens_seen": 292156912,
      "step": 17840,
      "train_runtime": 34349.1569,
      "train_tokens_per_second": 8505.505
    },
    {
      "epoch": 1.20152168058174,
      "grad_norm": 1.1171446713359712,
      "learning_rate": 3.44405568822596e-06,
      "loss": 0.2627789735794067,
      "num_input_tokens_seen": 292238832,
      "step": 17845,
      "train_runtime": 34358.8045,
      "train_tokens_per_second": 8505.501
    },
    {
      "epoch": 1.2018583355777,
      "grad_norm": 1.2993033783292052,
      "learning_rate": 3.441543107474765e-06,
      "loss": 0.308770751953125,
      "num_input_tokens_seen": 292320752,
      "step": 17850,
      "train_runtime": 34368.3501,
      "train_tokens_per_second": 8505.522
    },
    {
      "epoch": 1.2021949905736602,
      "grad_norm": 1.2980830660833602,
      "learning_rate": 3.439030962541069e-06,
      "loss": 0.28975481986999513,
      "num_input_tokens_seen": 292402672,
      "step": 17855,
      "train_runtime": 34377.8535,
      "train_tokens_per_second": 8505.553
    },
    {
      "epoch": 1.2025316455696202,
      "grad_norm": 1.183143533090801,
      "learning_rate": 3.4365192541273885e-06,
      "loss": 0.27566769123077395,
      "num_input_tokens_seen": 292484592,
      "step": 17860,
      "train_runtime": 34387.5137,
      "train_tokens_per_second": 8505.546
    },
    {
      "epoch": 1.2028683005655805,
      "grad_norm": 1.0563563317744344,
      "learning_rate": 3.4340079829361108e-06,
      "loss": 0.281511926651001,
      "num_input_tokens_seen": 292565616,
      "step": 17865,
      "train_runtime": 34397.4716,
      "train_tokens_per_second": 8505.44
    },
    {
      "epoch": 1.2032049555615405,
      "grad_norm": 1.092489886706704,
      "learning_rate": 3.4314971496695078e-06,
      "loss": 0.3164204120635986,
      "num_input_tokens_seen": 292647536,
      "step": 17870,
      "train_runtime": 34406.9739,
      "train_tokens_per_second": 8505.471
    },
    {
      "epoch": 1.2035416105575008,
      "grad_norm": 0.9606799375020725,
      "learning_rate": 3.428986755029724e-06,
      "loss": 0.2953948497772217,
      "num_input_tokens_seen": 292729456,
      "step": 17875,
      "train_runtime": 34416.4816,
      "train_tokens_per_second": 8505.502
    },
    {
      "epoch": 1.2038782655534608,
      "grad_norm": 1.061933328497973,
      "learning_rate": 3.4264767997187874e-06,
      "loss": 0.2840822696685791,
      "num_input_tokens_seen": 292811376,
      "step": 17880,
      "train_runtime": 34426.0305,
      "train_tokens_per_second": 8505.522
    },
    {
      "epoch": 1.204214920549421,
      "grad_norm": 1.1451255961086373,
      "learning_rate": 3.423967284438594e-06,
      "loss": 0.31098475456237795,
      "num_input_tokens_seen": 292892440,
      "step": 17885,
      "train_runtime": 34435.7023,
      "train_tokens_per_second": 8505.488
    },
    {
      "epoch": 1.204551575545381,
      "grad_norm": 0.9603527801791862,
      "learning_rate": 3.421458209890925e-06,
      "loss": 0.27209174633026123,
      "num_input_tokens_seen": 292974360,
      "step": 17890,
      "train_runtime": 34445.3893,
      "train_tokens_per_second": 8505.474
    },
    {
      "epoch": 1.2048882305413413,
      "grad_norm": 0.966921681076144,
      "learning_rate": 3.418949576777433e-06,
      "loss": 0.28141489028930666,
      "num_input_tokens_seen": 293056280,
      "step": 17895,
      "train_runtime": 34454.8942,
      "train_tokens_per_second": 8505.505
    },
    {
      "epoch": 1.2052248855373013,
      "grad_norm": 1.1458849664997963,
      "learning_rate": 3.4164413857996513e-06,
      "loss": 0.2961673021316528,
      "num_input_tokens_seen": 293138200,
      "step": 17900,
      "train_runtime": 34464.4002,
      "train_tokens_per_second": 8505.536
    },
    {
      "epoch": 1.2055615405332616,
      "grad_norm": 1.1567062946665347,
      "learning_rate": 3.4139336376589853e-06,
      "loss": 0.2764486074447632,
      "num_input_tokens_seen": 293220120,
      "step": 17905,
      "train_runtime": 34473.9038,
      "train_tokens_per_second": 8505.568
    },
    {
      "epoch": 1.2058981955292216,
      "grad_norm": 1.1302302536096265,
      "learning_rate": 3.411426333056722e-06,
      "loss": 0.2990281581878662,
      "num_input_tokens_seen": 293301680,
      "step": 17910,
      "train_runtime": 34483.9817,
      "train_tokens_per_second": 8505.447
    },
    {
      "epoch": 1.2062348505251819,
      "grad_norm": 1.4491938263964903,
      "learning_rate": 3.408919472694017e-06,
      "loss": 0.29686527252197265,
      "num_input_tokens_seen": 293383600,
      "step": 17915,
      "train_runtime": 34493.7978,
      "train_tokens_per_second": 8505.402
    },
    {
      "epoch": 1.206571505521142,
      "grad_norm": 1.1803139952724002,
      "learning_rate": 3.406413057271909e-06,
      "loss": 0.2973795413970947,
      "num_input_tokens_seen": 293465520,
      "step": 17920,
      "train_runtime": 34503.3091,
      "train_tokens_per_second": 8505.431
    },
    {
      "epoch": 1.2069081605171021,
      "grad_norm": 0.954307071273247,
      "learning_rate": 3.403907087491306e-06,
      "loss": 0.2631570816040039,
      "num_input_tokens_seen": 293547440,
      "step": 17925,
      "train_runtime": 34512.8098,
      "train_tokens_per_second": 8505.463
    },
    {
      "epoch": 1.2072448155130622,
      "grad_norm": 1.0572194639776362,
      "learning_rate": 3.4014015640529996e-06,
      "loss": 0.2788405895233154,
      "num_input_tokens_seen": 293629192,
      "step": 17930,
      "train_runtime": 34522.8066,
      "train_tokens_per_second": 8505.369
    },
    {
      "epoch": 1.2075814705090224,
      "grad_norm": 1.0787687255795728,
      "learning_rate": 3.398896487657646e-06,
      "loss": 0.3129566669464111,
      "num_input_tokens_seen": 293711112,
      "step": 17935,
      "train_runtime": 34532.3125,
      "train_tokens_per_second": 8505.399
    },
    {
      "epoch": 1.2079181255049825,
      "grad_norm": 1.0809377115319498,
      "learning_rate": 3.396391859005785e-06,
      "loss": 0.28866019248962405,
      "num_input_tokens_seen": 293793032,
      "step": 17940,
      "train_runtime": 34541.8167,
      "train_tokens_per_second": 8505.431
    },
    {
      "epoch": 1.2082547805009427,
      "grad_norm": 0.8848466002782008,
      "learning_rate": 3.393887678797826e-06,
      "loss": 0.272682785987854,
      "num_input_tokens_seen": 293874952,
      "step": 17945,
      "train_runtime": 34551.3301,
      "train_tokens_per_second": 8505.46
    },
    {
      "epoch": 1.2085914354969027,
      "grad_norm": 1.018748842444151,
      "learning_rate": 3.3913839477340575e-06,
      "loss": 0.29358787536621095,
      "num_input_tokens_seen": 293956872,
      "step": 17950,
      "train_runtime": 34560.8326,
      "train_tokens_per_second": 8505.492
    },
    {
      "epoch": 1.208928090492863,
      "grad_norm": 1.063890317212777,
      "learning_rate": 3.388880666514637e-06,
      "loss": 0.2653756856918335,
      "num_input_tokens_seen": 294038792,
      "step": 17955,
      "train_runtime": 34570.3313,
      "train_tokens_per_second": 8505.524
    },
    {
      "epoch": 1.209264745488823,
      "grad_norm": 1.1563719843028006,
      "learning_rate": 3.3863778358396033e-06,
      "loss": 0.2993776798248291,
      "num_input_tokens_seen": 294120712,
      "step": 17960,
      "train_runtime": 34579.847,
      "train_tokens_per_second": 8505.553
    },
    {
      "epoch": 1.2096014004847833,
      "grad_norm": 0.9521905645592822,
      "learning_rate": 3.3838754564088592e-06,
      "loss": 0.2633402347564697,
      "num_input_tokens_seen": 294202632,
      "step": 17965,
      "train_runtime": 34589.3564,
      "train_tokens_per_second": 8505.583
    },
    {
      "epoch": 1.2099380554807433,
      "grad_norm": 0.901431649786334,
      "learning_rate": 3.3813735289221923e-06,
      "loss": 0.2865684986114502,
      "num_input_tokens_seen": 294284552,
      "step": 17970,
      "train_runtime": 34598.8557,
      "train_tokens_per_second": 8505.615
    },
    {
      "epoch": 1.2102747104767035,
      "grad_norm": 0.8958073268537815,
      "learning_rate": 3.378872054079255e-06,
      "loss": 0.2568507194519043,
      "num_input_tokens_seen": 294364728,
      "step": 17975,
      "train_runtime": 34608.4133,
      "train_tokens_per_second": 8505.583
    },
    {
      "epoch": 1.2106113654726636,
      "grad_norm": 1.1985374126953485,
      "learning_rate": 3.3763710325795784e-06,
      "loss": 0.3038368225097656,
      "num_input_tokens_seen": 294446648,
      "step": 17980,
      "train_runtime": 34617.9185,
      "train_tokens_per_second": 8505.614
    },
    {
      "epoch": 1.2109480204686238,
      "grad_norm": 1.0294813912066094,
      "learning_rate": 3.3738704651225673e-06,
      "loss": 0.27220821380615234,
      "num_input_tokens_seen": 294528568,
      "step": 17985,
      "train_runtime": 34627.526,
      "train_tokens_per_second": 8505.62
    },
    {
      "epoch": 1.2112846754645838,
      "grad_norm": 1.0174895828357768,
      "learning_rate": 3.3713703524074936e-06,
      "loss": 0.27738146781921386,
      "num_input_tokens_seen": 294610488,
      "step": 17990,
      "train_runtime": 34637.1357,
      "train_tokens_per_second": 8505.625
    },
    {
      "epoch": 1.211621330460544,
      "grad_norm": 1.0698163895339883,
      "learning_rate": 3.3688706951335082e-06,
      "loss": 0.28127174377441405,
      "num_input_tokens_seen": 294691608,
      "step": 17995,
      "train_runtime": 34646.8244,
      "train_tokens_per_second": 8505.588
    },
    {
      "epoch": 1.2119579854565041,
      "grad_norm": 1.2364213494363587,
      "learning_rate": 3.3663714939996305e-06,
      "loss": 0.2932534456253052,
      "num_input_tokens_seen": 294773528,
      "step": 18000,
      "train_runtime": 34656.3429,
      "train_tokens_per_second": 8505.616
    },
    {
      "epoch": 1.2122946404524644,
      "grad_norm": 1.2996395588963034,
      "learning_rate": 3.3638727497047575e-06,
      "loss": 0.28820137977600097,
      "num_input_tokens_seen": 294855448,
      "step": 18005,
      "train_runtime": 34666.0467,
      "train_tokens_per_second": 8505.598
    },
    {
      "epoch": 1.2126312954484244,
      "grad_norm": 1.143014071496677,
      "learning_rate": 3.3613744629476507e-06,
      "loss": 0.27965168952941893,
      "num_input_tokens_seen": 294937368,
      "step": 18010,
      "train_runtime": 34675.6448,
      "train_tokens_per_second": 8505.606
    },
    {
      "epoch": 1.2129679504443847,
      "grad_norm": 1.0631852940425581,
      "learning_rate": 3.358876634426952e-06,
      "loss": 0.26172442436218263,
      "num_input_tokens_seen": 295019288,
      "step": 18015,
      "train_runtime": 34685.2578,
      "train_tokens_per_second": 8505.61
    },
    {
      "epoch": 1.2133046054403447,
      "grad_norm": 1.0727755970877813,
      "learning_rate": 3.3563792648411676e-06,
      "loss": 0.3022951364517212,
      "num_input_tokens_seen": 295101208,
      "step": 18020,
      "train_runtime": 34694.7631,
      "train_tokens_per_second": 8505.641
    },
    {
      "epoch": 1.213641260436305,
      "grad_norm": 1.1214007113445432,
      "learning_rate": 3.353882354888681e-06,
      "loss": 0.2699776887893677,
      "num_input_tokens_seen": 295183128,
      "step": 18025,
      "train_runtime": 34704.3879,
      "train_tokens_per_second": 8505.643
    },
    {
      "epoch": 1.213977915432265,
      "grad_norm": 1.0807624125055828,
      "learning_rate": 3.3513859052677444e-06,
      "loss": 0.26034936904907224,
      "num_input_tokens_seen": 295265048,
      "step": 18030,
      "train_runtime": 34714.0271,
      "train_tokens_per_second": 8505.641
    },
    {
      "epoch": 1.2143145704282252,
      "grad_norm": 1.1426286910594805,
      "learning_rate": 3.3488899166764842e-06,
      "loss": 0.31810870170593264,
      "num_input_tokens_seen": 295346968,
      "step": 18035,
      "train_runtime": 34723.5284,
      "train_tokens_per_second": 8505.673
    },
    {
      "epoch": 1.2146512254241852,
      "grad_norm": 1.1468489807764568,
      "learning_rate": 3.346394389812892e-06,
      "loss": 0.28781416416168215,
      "num_input_tokens_seen": 295428480,
      "step": 18040,
      "train_runtime": 34733.4277,
      "train_tokens_per_second": 8505.595
    },
    {
      "epoch": 1.2149878804201455,
      "grad_norm": 1.0429254553863396,
      "learning_rate": 3.3438993253748365e-06,
      "loss": 0.29687206745147704,
      "num_input_tokens_seen": 295510400,
      "step": 18045,
      "train_runtime": 34743.0014,
      "train_tokens_per_second": 8505.609
    },
    {
      "epoch": 1.2153245354161055,
      "grad_norm": 1.1052520644457942,
      "learning_rate": 3.3414047240600526e-06,
      "loss": 0.29827728271484377,
      "num_input_tokens_seen": 295592320,
      "step": 18050,
      "train_runtime": 34752.602,
      "train_tokens_per_second": 8505.617
    },
    {
      "epoch": 1.2156611904120658,
      "grad_norm": 1.0801486636440956,
      "learning_rate": 3.338910586566151e-06,
      "loss": 0.3007194995880127,
      "num_input_tokens_seen": 295674240,
      "step": 18055,
      "train_runtime": 34762.1041,
      "train_tokens_per_second": 8505.649
    },
    {
      "epoch": 1.2159978454080258,
      "grad_norm": 1.2592228463143873,
      "learning_rate": 3.336416913590605e-06,
      "loss": 0.31157104969024657,
      "num_input_tokens_seen": 295756160,
      "step": 18060,
      "train_runtime": 34771.7201,
      "train_tokens_per_second": 8505.652
    },
    {
      "epoch": 1.216334500403986,
      "grad_norm": 1.1658176208405528,
      "learning_rate": 3.333923705830766e-06,
      "loss": 0.26777989864349366,
      "num_input_tokens_seen": 295838080,
      "step": 18065,
      "train_runtime": 34781.4071,
      "train_tokens_per_second": 8505.639
    },
    {
      "epoch": 1.216671155399946,
      "grad_norm": 1.1030631871452674,
      "learning_rate": 3.3314309639838484e-06,
      "loss": 0.3226517677307129,
      "num_input_tokens_seen": 295920000,
      "step": 18070,
      "train_runtime": 34791.0423,
      "train_tokens_per_second": 8505.638
    },
    {
      "epoch": 1.2170078103959063,
      "grad_norm": 1.0667927777744945,
      "learning_rate": 3.328938688746942e-06,
      "loss": 0.28281216621398925,
      "num_input_tokens_seen": 296001920,
      "step": 18075,
      "train_runtime": 34800.6911,
      "train_tokens_per_second": 8505.633
    },
    {
      "epoch": 1.2173444653918664,
      "grad_norm": 1.19727741619116,
      "learning_rate": 3.3264468808170012e-06,
      "loss": 0.3138002634048462,
      "num_input_tokens_seen": 296083840,
      "step": 18080,
      "train_runtime": 34810.2422,
      "train_tokens_per_second": 8505.653
    },
    {
      "epoch": 1.2176811203878266,
      "grad_norm": 1.1499596213789554,
      "learning_rate": 3.3239555408908562e-06,
      "loss": 0.29733221530914306,
      "num_input_tokens_seen": 296165760,
      "step": 18085,
      "train_runtime": 34819.8404,
      "train_tokens_per_second": 8505.661
    },
    {
      "epoch": 1.2180177753837866,
      "grad_norm": 1.1065424796557564,
      "learning_rate": 3.3214646696651974e-06,
      "loss": 0.28372979164123535,
      "num_input_tokens_seen": 296247160,
      "step": 18090,
      "train_runtime": 34829.7986,
      "train_tokens_per_second": 8505.566
    },
    {
      "epoch": 1.2183544303797469,
      "grad_norm": 1.1910757126262141,
      "learning_rate": 3.3189742678365923e-06,
      "loss": 0.2972620725631714,
      "num_input_tokens_seen": 296329080,
      "step": 18095,
      "train_runtime": 34839.5675,
      "train_tokens_per_second": 8505.533
    },
    {
      "epoch": 1.218691085375707,
      "grad_norm": 1.0866946611936055,
      "learning_rate": 3.316484336101472e-06,
      "loss": 0.2689074993133545,
      "num_input_tokens_seen": 296411000,
      "step": 18100,
      "train_runtime": 34849.2533,
      "train_tokens_per_second": 8505.519
    },
    {
      "epoch": 1.2190277403716672,
      "grad_norm": 1.0323539218267253,
      "learning_rate": 3.313994875156141e-06,
      "loss": 0.3023322105407715,
      "num_input_tokens_seen": 296492920,
      "step": 18105,
      "train_runtime": 34858.759,
      "train_tokens_per_second": 8505.55
    },
    {
      "epoch": 1.2193643953676272,
      "grad_norm": 1.0713662396099493,
      "learning_rate": 3.3115058856967643e-06,
      "loss": 0.296788763999939,
      "num_input_tokens_seen": 296574840,
      "step": 18110,
      "train_runtime": 34868.4329,
      "train_tokens_per_second": 8505.54
    },
    {
      "epoch": 1.2197010503635874,
      "grad_norm": 1.2268598125814212,
      "learning_rate": 3.3090173684193845e-06,
      "loss": 0.2716069221496582,
      "num_input_tokens_seen": 296656760,
      "step": 18115,
      "train_runtime": 34877.9725,
      "train_tokens_per_second": 8505.562
    },
    {
      "epoch": 1.2200377053595475,
      "grad_norm": 1.0660914372348276,
      "learning_rate": 3.3065293240199037e-06,
      "loss": 0.2910308361053467,
      "num_input_tokens_seen": 296738680,
      "step": 18120,
      "train_runtime": 34887.5971,
      "train_tokens_per_second": 8505.564
    },
    {
      "epoch": 1.2203743603555077,
      "grad_norm": 1.0987600152340742,
      "learning_rate": 3.3040417531940982e-06,
      "loss": 0.3300910949707031,
      "num_input_tokens_seen": 296820600,
      "step": 18125,
      "train_runtime": 34897.1211,
      "train_tokens_per_second": 8505.59
    },
    {
      "epoch": 1.2207110153514678,
      "grad_norm": 1.1220716820908838,
      "learning_rate": 3.301554656637607e-06,
      "loss": 0.28183808326721194,
      "num_input_tokens_seen": 296902520,
      "step": 18130,
      "train_runtime": 34906.7642,
      "train_tokens_per_second": 8505.587
    },
    {
      "epoch": 1.221047670347428,
      "grad_norm": 1.0306166098368055,
      "learning_rate": 3.2990680350459427e-06,
      "loss": 0.28535192012786864,
      "num_input_tokens_seen": 296984064,
      "step": 18135,
      "train_runtime": 34916.796,
      "train_tokens_per_second": 8505.479
    },
    {
      "epoch": 1.221384325343388,
      "grad_norm": 1.0650992607184362,
      "learning_rate": 3.2965818891144763e-06,
      "loss": 0.29346590042114257,
      "num_input_tokens_seen": 297065984,
      "step": 18140,
      "train_runtime": 34926.4343,
      "train_tokens_per_second": 8505.477
    },
    {
      "epoch": 1.2217209803393483,
      "grad_norm": 1.070567458302702,
      "learning_rate": 3.294096219538454e-06,
      "loss": 0.30739216804504393,
      "num_input_tokens_seen": 297147904,
      "step": 18145,
      "train_runtime": 34935.9578,
      "train_tokens_per_second": 8505.503
    },
    {
      "epoch": 1.2220576353353083,
      "grad_norm": 0.923713055426218,
      "learning_rate": 3.291611027012983e-06,
      "loss": 0.2761296987533569,
      "num_input_tokens_seen": 297229824,
      "step": 18150,
      "train_runtime": 34945.4696,
      "train_tokens_per_second": 8505.532
    },
    {
      "epoch": 1.2223942903312686,
      "grad_norm": 1.3294844771097323,
      "learning_rate": 3.289126312233043e-06,
      "loss": 0.3072872877120972,
      "num_input_tokens_seen": 297311744,
      "step": 18155,
      "train_runtime": 34955.136,
      "train_tokens_per_second": 8505.524
    },
    {
      "epoch": 1.2227309453272286,
      "grad_norm": 1.0312281857504295,
      "learning_rate": 3.2866420758934714e-06,
      "loss": 0.2903941869735718,
      "num_input_tokens_seen": 297393576,
      "step": 18160,
      "train_runtime": 34965.2963,
      "train_tokens_per_second": 8505.393
    },
    {
      "epoch": 1.2230676003231888,
      "grad_norm": 1.1604253530910396,
      "learning_rate": 3.2841583186889826e-06,
      "loss": 0.27792205810546877,
      "num_input_tokens_seen": 297475496,
      "step": 18165,
      "train_runtime": 34974.8404,
      "train_tokens_per_second": 8505.414
    },
    {
      "epoch": 1.2234042553191489,
      "grad_norm": 1.0557033980545925,
      "learning_rate": 3.281675041314146e-06,
      "loss": 0.2737730026245117,
      "num_input_tokens_seen": 297557416,
      "step": 18170,
      "train_runtime": 34984.3443,
      "train_tokens_per_second": 8505.445
    },
    {
      "epoch": 1.2237409103151091,
      "grad_norm": 1.071880413115055,
      "learning_rate": 3.279192244463406e-06,
      "loss": 0.3208924293518066,
      "num_input_tokens_seen": 297639336,
      "step": 18175,
      "train_runtime": 34993.9695,
      "train_tokens_per_second": 8505.447
    },
    {
      "epoch": 1.2240775653110691,
      "grad_norm": 1.16593981216457,
      "learning_rate": 3.2767099288310667e-06,
      "loss": 0.2806739330291748,
      "num_input_tokens_seen": 297721256,
      "step": 18180,
      "train_runtime": 35003.5596,
      "train_tokens_per_second": 8505.457
    },
    {
      "epoch": 1.2244142203070294,
      "grad_norm": 1.1758135452811038,
      "learning_rate": 3.2742280951113025e-06,
      "loss": 0.2911004781723022,
      "num_input_tokens_seen": 297803176,
      "step": 18185,
      "train_runtime": 35013.0648,
      "train_tokens_per_second": 8505.487
    },
    {
      "epoch": 1.2247508753029894,
      "grad_norm": 1.0119974767593334,
      "learning_rate": 3.2717467439981464e-06,
      "loss": 0.2801973342895508,
      "num_input_tokens_seen": 297885096,
      "step": 18190,
      "train_runtime": 35022.5719,
      "train_tokens_per_second": 8505.517
    },
    {
      "epoch": 1.2250875302989497,
      "grad_norm": 1.225911996570755,
      "learning_rate": 3.2692658761855033e-06,
      "loss": 0.2701519012451172,
      "num_input_tokens_seen": 297967016,
      "step": 18195,
      "train_runtime": 35032.0752,
      "train_tokens_per_second": 8505.549
    },
    {
      "epoch": 1.2254241852949097,
      "grad_norm": 1.1112481212170782,
      "learning_rate": 3.266785492367138e-06,
      "loss": 0.29944658279418945,
      "num_input_tokens_seen": 298048936,
      "step": 18200,
      "train_runtime": 35041.6063,
      "train_tokens_per_second": 8505.573
    },
    {
      "epoch": 1.22576084029087,
      "grad_norm": 1.2595672057192993,
      "learning_rate": 3.2643055932366852e-06,
      "loss": 0.2746922492980957,
      "num_input_tokens_seen": 298129888,
      "step": 18205,
      "train_runtime": 35051.64,
      "train_tokens_per_second": 8505.448
    },
    {
      "epoch": 1.22609749528683,
      "grad_norm": 0.9958832941960424,
      "learning_rate": 3.2618261794876365e-06,
      "loss": 0.27780392169952395,
      "num_input_tokens_seen": 298211808,
      "step": 18210,
      "train_runtime": 35061.1591,
      "train_tokens_per_second": 8505.475
    },
    {
      "epoch": 1.2264341502827902,
      "grad_norm": 0.9560215904563768,
      "learning_rate": 3.259347251813356e-06,
      "loss": 0.26649842262268064,
      "num_input_tokens_seen": 298293728,
      "step": 18215,
      "train_runtime": 35070.665,
      "train_tokens_per_second": 8505.505
    },
    {
      "epoch": 1.2267708052787503,
      "grad_norm": 1.2288042659036533,
      "learning_rate": 3.256868810907064e-06,
      "loss": 0.30932011604309084,
      "num_input_tokens_seen": 298375648,
      "step": 18220,
      "train_runtime": 35080.2874,
      "train_tokens_per_second": 8505.508
    },
    {
      "epoch": 1.2271074602747105,
      "grad_norm": 1.1068947677909795,
      "learning_rate": 3.2543908574618514e-06,
      "loss": 0.2795151710510254,
      "num_input_tokens_seen": 298457568,
      "step": 18225,
      "train_runtime": 35089.7972,
      "train_tokens_per_second": 8505.537
    },
    {
      "epoch": 1.2274441152706705,
      "grad_norm": 1.0145656710845248,
      "learning_rate": 3.251913392170668e-06,
      "loss": 0.28255209922790525,
      "num_input_tokens_seen": 298539488,
      "step": 18230,
      "train_runtime": 35099.432,
      "train_tokens_per_second": 8505.536
    },
    {
      "epoch": 1.2277807702666308,
      "grad_norm": 1.127254453195675,
      "learning_rate": 3.249436415726333e-06,
      "loss": 0.2967740535736084,
      "num_input_tokens_seen": 298621408,
      "step": 18235,
      "train_runtime": 35109.0561,
      "train_tokens_per_second": 8505.538
    },
    {
      "epoch": 1.2281174252625908,
      "grad_norm": 1.2622422378712332,
      "learning_rate": 3.24695992882152e-06,
      "loss": 0.29569268226623535,
      "num_input_tokens_seen": 298702984,
      "step": 18240,
      "train_runtime": 35119.0832,
      "train_tokens_per_second": 8505.432
    },
    {
      "epoch": 1.228454080258551,
      "grad_norm": 1.155873525872413,
      "learning_rate": 3.244483932148773e-06,
      "loss": 0.2625589370727539,
      "num_input_tokens_seen": 298784904,
      "step": 18245,
      "train_runtime": 35128.7208,
      "train_tokens_per_second": 8505.431
    },
    {
      "epoch": 1.228790735254511,
      "grad_norm": 1.034996668054105,
      "learning_rate": 3.2420084264004966e-06,
      "loss": 0.2650510549545288,
      "num_input_tokens_seen": 298866824,
      "step": 18250,
      "train_runtime": 35138.4007,
      "train_tokens_per_second": 8505.419
    },
    {
      "epoch": 1.2291273902504714,
      "grad_norm": 1.1066790004777742,
      "learning_rate": 3.2395334122689594e-06,
      "loss": 0.28680148124694826,
      "num_input_tokens_seen": 298948744,
      "step": 18255,
      "train_runtime": 35147.9132,
      "train_tokens_per_second": 8505.448
    },
    {
      "epoch": 1.2294640452464314,
      "grad_norm": 1.1301170293025753,
      "learning_rate": 3.2370588904462873e-06,
      "loss": 0.29141418933868407,
      "num_input_tokens_seen": 299030664,
      "step": 18260,
      "train_runtime": 35157.4533,
      "train_tokens_per_second": 8505.47
    },
    {
      "epoch": 1.2298007002423916,
      "grad_norm": 1.2396491773252156,
      "learning_rate": 3.2345848616244775e-06,
      "loss": 0.29670867919921873,
      "num_input_tokens_seen": 299112584,
      "step": 18265,
      "train_runtime": 35167.1378,
      "train_tokens_per_second": 8505.457
    },
    {
      "epoch": 1.2301373552383517,
      "grad_norm": 1.1288923926097973,
      "learning_rate": 3.2321113264953797e-06,
      "loss": 0.2755516290664673,
      "num_input_tokens_seen": 299194504,
      "step": 18270,
      "train_runtime": 35176.7701,
      "train_tokens_per_second": 8505.457
    },
    {
      "epoch": 1.230474010234312,
      "grad_norm": 0.8942065603673365,
      "learning_rate": 3.2296382857507124e-06,
      "loss": 0.26826484203338624,
      "num_input_tokens_seen": 299276424,
      "step": 18275,
      "train_runtime": 35186.4721,
      "train_tokens_per_second": 8505.44
    },
    {
      "epoch": 1.230810665230272,
      "grad_norm": 0.9710464097609378,
      "learning_rate": 3.227165740082053e-06,
      "loss": 0.311414098739624,
      "num_input_tokens_seen": 299358344,
      "step": 18280,
      "train_runtime": 35196.2114,
      "train_tokens_per_second": 8505.414
    },
    {
      "epoch": 1.2311473202262322,
      "grad_norm": 0.9821649549333856,
      "learning_rate": 3.2246936901808424e-06,
      "loss": 0.2845176696777344,
      "num_input_tokens_seen": 299440264,
      "step": 18285,
      "train_runtime": 35205.8832,
      "train_tokens_per_second": 8505.404
    },
    {
      "epoch": 1.2314839752221922,
      "grad_norm": 1.3131864716025143,
      "learning_rate": 3.2222221367383786e-06,
      "loss": 0.300424861907959,
      "num_input_tokens_seen": 299522184,
      "step": 18290,
      "train_runtime": 35215.3984,
      "train_tokens_per_second": 8505.432
    },
    {
      "epoch": 1.2318206302181525,
      "grad_norm": 1.0083697834034535,
      "learning_rate": 3.2197510804458265e-06,
      "loss": 0.24087836742401122,
      "num_input_tokens_seen": 299604104,
      "step": 18295,
      "train_runtime": 35224.9155,
      "train_tokens_per_second": 8505.46
    },
    {
      "epoch": 1.2321572852141125,
      "grad_norm": 1.0532881483636574,
      "learning_rate": 3.217280521994207e-06,
      "loss": 0.28246798515319826,
      "num_input_tokens_seen": 299686024,
      "step": 18300,
      "train_runtime": 35234.4298,
      "train_tokens_per_second": 8505.488
    },
    {
      "epoch": 1.2324939402100727,
      "grad_norm": 1.1684560191887314,
      "learning_rate": 3.214810462074407e-06,
      "loss": 0.3057229995727539,
      "num_input_tokens_seen": 299767784,
      "step": 18305,
      "train_runtime": 35244.1018,
      "train_tokens_per_second": 8505.474
    },
    {
      "epoch": 1.2328305952060328,
      "grad_norm": 1.195125542990422,
      "learning_rate": 3.2123409013771666e-06,
      "loss": 0.29502077102661134,
      "num_input_tokens_seen": 299849704,
      "step": 18310,
      "train_runtime": 35253.7575,
      "train_tokens_per_second": 8505.468
    },
    {
      "epoch": 1.233167250201993,
      "grad_norm": 0.964290647261302,
      "learning_rate": 3.209871840593095e-06,
      "loss": 0.291918420791626,
      "num_input_tokens_seen": 299931624,
      "step": 18315,
      "train_runtime": 35263.2651,
      "train_tokens_per_second": 8505.498
    },
    {
      "epoch": 1.233503905197953,
      "grad_norm": 1.0292151520395896,
      "learning_rate": 3.207403280412652e-06,
      "loss": 0.2865922451019287,
      "num_input_tokens_seen": 300013544,
      "step": 18320,
      "train_runtime": 35272.7841,
      "train_tokens_per_second": 8505.525
    },
    {
      "epoch": 1.2338405601939133,
      "grad_norm": 0.95186247893033,
      "learning_rate": 3.2049352215261677e-06,
      "loss": 0.26020660400390627,
      "num_input_tokens_seen": 300095464,
      "step": 18325,
      "train_runtime": 35282.451,
      "train_tokens_per_second": 8505.516
    },
    {
      "epoch": 1.2341772151898733,
      "grad_norm": 1.1169369773059639,
      "learning_rate": 3.2024676646238222e-06,
      "loss": 0.28460566997528075,
      "num_input_tokens_seen": 300177384,
      "step": 18330,
      "train_runtime": 35292.0787,
      "train_tokens_per_second": 8505.517
    },
    {
      "epoch": 1.2345138701858336,
      "grad_norm": 1.023637818956653,
      "learning_rate": 3.2000006103956653e-06,
      "loss": 0.2969768524169922,
      "num_input_tokens_seen": 300259304,
      "step": 18335,
      "train_runtime": 35301.6999,
      "train_tokens_per_second": 8505.52
    },
    {
      "epoch": 1.2348505251817936,
      "grad_norm": 1.1899915361303974,
      "learning_rate": 3.1975340595315956e-06,
      "loss": 0.2731428623199463,
      "num_input_tokens_seen": 300341224,
      "step": 18340,
      "train_runtime": 35311.3856,
      "train_tokens_per_second": 8505.507
    },
    {
      "epoch": 1.2351871801777539,
      "grad_norm": 1.2879459183913458,
      "learning_rate": 3.1950680127213796e-06,
      "loss": 0.28988473415374755,
      "num_input_tokens_seen": 300423144,
      "step": 18345,
      "train_runtime": 35321.0306,
      "train_tokens_per_second": 8505.503
    },
    {
      "epoch": 1.235523835173714,
      "grad_norm": 1.0146693818137524,
      "learning_rate": 3.192602470654638e-06,
      "loss": 0.29022912979125975,
      "num_input_tokens_seen": 300505064,
      "step": 18350,
      "train_runtime": 35330.5374,
      "train_tokens_per_second": 8505.533
    },
    {
      "epoch": 1.2358604901696741,
      "grad_norm": 0.8941711954399651,
      "learning_rate": 3.190137434020853e-06,
      "loss": 0.29668655395507815,
      "num_input_tokens_seen": 300586984,
      "step": 18355,
      "train_runtime": 35340.0422,
      "train_tokens_per_second": 8505.564
    },
    {
      "epoch": 1.2361971451656342,
      "grad_norm": 1.1414424183405412,
      "learning_rate": 3.1876729035093616e-06,
      "loss": 0.27527427673339844,
      "num_input_tokens_seen": 300668904,
      "step": 18360,
      "train_runtime": 35349.5974,
      "train_tokens_per_second": 8505.582
    },
    {
      "epoch": 1.2365338001615944,
      "grad_norm": 1.1577396811825802,
      "learning_rate": 3.185208879809366e-06,
      "loss": 0.27464444637298585,
      "num_input_tokens_seen": 300750824,
      "step": 18365,
      "train_runtime": 35359.2247,
      "train_tokens_per_second": 8505.583
    },
    {
      "epoch": 1.2368704551575544,
      "grad_norm": 1.1321794042771194,
      "learning_rate": 3.1827453636099183e-06,
      "loss": 0.29783799648284914,
      "num_input_tokens_seen": 300832744,
      "step": 18370,
      "train_runtime": 35368.8182,
      "train_tokens_per_second": 8505.592
    },
    {
      "epoch": 1.2372071101535147,
      "grad_norm": 1.2609563845810063,
      "learning_rate": 3.1802823555999356e-06,
      "loss": 0.27928619384765624,
      "num_input_tokens_seen": 300914664,
      "step": 18375,
      "train_runtime": 35378.3247,
      "train_tokens_per_second": 8505.622
    },
    {
      "epoch": 1.2375437651494747,
      "grad_norm": 1.3639379652714914,
      "learning_rate": 3.1778198564681885e-06,
      "loss": 0.3152610301971436,
      "num_input_tokens_seen": 300996584,
      "step": 18380,
      "train_runtime": 35387.8463,
      "train_tokens_per_second": 8505.649
    },
    {
      "epoch": 1.237880420145435,
      "grad_norm": 1.3736919081561971,
      "learning_rate": 3.17535786690331e-06,
      "loss": 0.3151847362518311,
      "num_input_tokens_seen": 301078504,
      "step": 18385,
      "train_runtime": 35397.3514,
      "train_tokens_per_second": 8505.679
    },
    {
      "epoch": 1.238217075141395,
      "grad_norm": 1.2689980025963792,
      "learning_rate": 3.172896387593784e-06,
      "loss": 0.3442038059234619,
      "num_input_tokens_seen": 301160424,
      "step": 18390,
      "train_runtime": 35406.8605,
      "train_tokens_per_second": 8505.708
    },
    {
      "epoch": 1.2385537301373553,
      "grad_norm": 1.1314737424452588,
      "learning_rate": 3.1704354192279573e-06,
      "loss": 0.2921198606491089,
      "num_input_tokens_seen": 301242344,
      "step": 18395,
      "train_runtime": 35416.4823,
      "train_tokens_per_second": 8505.71
    },
    {
      "epoch": 1.2388903851333153,
      "grad_norm": 1.2333123824243106,
      "learning_rate": 3.1679749624940316e-06,
      "loss": 0.29438908100128175,
      "num_input_tokens_seen": 301324264,
      "step": 18400,
      "train_runtime": 35426.1049,
      "train_tokens_per_second": 8505.713
    },
    {
      "epoch": 1.2392270401292755,
      "grad_norm": 1.2116530621874386,
      "learning_rate": 3.165515018080067e-06,
      "loss": 0.30624613761901853,
      "num_input_tokens_seen": 301406184,
      "step": 18405,
      "train_runtime": 35435.6216,
      "train_tokens_per_second": 8505.74
    },
    {
      "epoch": 1.2395636951252356,
      "grad_norm": 1.1485798925774762,
      "learning_rate": 3.163055586673976e-06,
      "loss": 0.2660425901412964,
      "num_input_tokens_seen": 301486928,
      "step": 18410,
      "train_runtime": 35445.3907,
      "train_tokens_per_second": 8505.674
    },
    {
      "epoch": 1.2399003501211958,
      "grad_norm": 1.3156069393026322,
      "learning_rate": 3.1605966689635348e-06,
      "loss": 0.2939130783081055,
      "num_input_tokens_seen": 301568848,
      "step": 18415,
      "train_runtime": 35455.0681,
      "train_tokens_per_second": 8505.663
    },
    {
      "epoch": 1.2402370051171558,
      "grad_norm": 1.1042682357679487,
      "learning_rate": 3.1581382656363666e-06,
      "loss": 0.2943774700164795,
      "num_input_tokens_seen": 301650560,
      "step": 18420,
      "train_runtime": 35465.064,
      "train_tokens_per_second": 8505.569
    },
    {
      "epoch": 1.240573660113116,
      "grad_norm": 1.0924449530392728,
      "learning_rate": 3.1556803773799616e-06,
      "loss": 0.30319387912750245,
      "num_input_tokens_seen": 301732480,
      "step": 18425,
      "train_runtime": 35474.6672,
      "train_tokens_per_second": 8505.576
    },
    {
      "epoch": 1.2409103151090761,
      "grad_norm": 1.0122235107705495,
      "learning_rate": 3.1532230048816564e-06,
      "loss": 0.2763458967208862,
      "num_input_tokens_seen": 301814400,
      "step": 18430,
      "train_runtime": 35484.2776,
      "train_tokens_per_second": 8505.581
    },
    {
      "epoch": 1.2412469701050364,
      "grad_norm": 1.018580790416065,
      "learning_rate": 3.150766148828651e-06,
      "loss": 0.2770243167877197,
      "num_input_tokens_seen": 301896320,
      "step": 18435,
      "train_runtime": 35494.1106,
      "train_tokens_per_second": 8505.533
    },
    {
      "epoch": 1.2415836251009964,
      "grad_norm": 1.0517717937602005,
      "learning_rate": 3.1483098099079934e-06,
      "loss": 0.30229403972625735,
      "num_input_tokens_seen": 301978240,
      "step": 18440,
      "train_runtime": 35503.7503,
      "train_tokens_per_second": 8505.531
    },
    {
      "epoch": 1.2419202800969567,
      "grad_norm": 1.1842076122857659,
      "learning_rate": 3.1458539888065937e-06,
      "loss": 0.2681442737579346,
      "num_input_tokens_seen": 302060160,
      "step": 18445,
      "train_runtime": 35513.3877,
      "train_tokens_per_second": 8505.529
    },
    {
      "epoch": 1.2422569350929167,
      "grad_norm": 1.079051858634197,
      "learning_rate": 3.143398686211212e-06,
      "loss": 0.2829803466796875,
      "num_input_tokens_seen": 302142080,
      "step": 18450,
      "train_runtime": 35522.8957,
      "train_tokens_per_second": 8505.559
    },
    {
      "epoch": 1.242593590088877,
      "grad_norm": 1.2878454392648564,
      "learning_rate": 3.1409439028084706e-06,
      "loss": 0.30536880493164065,
      "num_input_tokens_seen": 302224000,
      "step": 18455,
      "train_runtime": 35532.4023,
      "train_tokens_per_second": 8505.589
    },
    {
      "epoch": 1.2429302450848372,
      "grad_norm": 1.0483339034033852,
      "learning_rate": 3.1384896392848363e-06,
      "loss": 0.28885018825531006,
      "num_input_tokens_seen": 302305920,
      "step": 18460,
      "train_runtime": 35541.9086,
      "train_tokens_per_second": 8505.619
    },
    {
      "epoch": 1.2432669000807972,
      "grad_norm": 1.2225522098583408,
      "learning_rate": 3.1360358963266404e-06,
      "loss": 0.3377589225769043,
      "num_input_tokens_seen": 302387840,
      "step": 18465,
      "train_runtime": 35551.4182,
      "train_tokens_per_second": 8505.648
    },
    {
      "epoch": 1.2436035550767572,
      "grad_norm": 1.109788489817371,
      "learning_rate": 3.1335826746200594e-06,
      "loss": 0.29207477569580076,
      "num_input_tokens_seen": 302469760,
      "step": 18470,
      "train_runtime": 35560.9266,
      "train_tokens_per_second": 8505.677
    },
    {
      "epoch": 1.2439402100727175,
      "grad_norm": 0.9802418879511295,
      "learning_rate": 3.1311299748511335e-06,
      "loss": 0.3102996587753296,
      "num_input_tokens_seen": 302551680,
      "step": 18475,
      "train_runtime": 35570.4337,
      "train_tokens_per_second": 8505.707
    },
    {
      "epoch": 1.2442768650686777,
      "grad_norm": 1.0236188419629333,
      "learning_rate": 3.1286777977057504e-06,
      "loss": 0.28977384567260744,
      "num_input_tokens_seen": 302633600,
      "step": 18480,
      "train_runtime": 35579.9407,
      "train_tokens_per_second": 8505.736
    },
    {
      "epoch": 1.2446135200646378,
      "grad_norm": 1.022789601297971,
      "learning_rate": 3.126226143869655e-06,
      "loss": 0.300982666015625,
      "num_input_tokens_seen": 302715096,
      "step": 18485,
      "train_runtime": 35589.8113,
      "train_tokens_per_second": 8505.667
    },
    {
      "epoch": 1.2449501750605978,
      "grad_norm": 1.1006874312606845,
      "learning_rate": 3.1237750140284424e-06,
      "loss": 0.27502658367156985,
      "num_input_tokens_seen": 302797016,
      "step": 18490,
      "train_runtime": 35599.4922,
      "train_tokens_per_second": 8505.655
    },
    {
      "epoch": 1.245286830056558,
      "grad_norm": 0.9352875043319326,
      "learning_rate": 3.1213244088675647e-06,
      "loss": 0.31262285709381105,
      "num_input_tokens_seen": 302878936,
      "step": 18495,
      "train_runtime": 35609.0069,
      "train_tokens_per_second": 8505.683
    },
    {
      "epoch": 1.2456234850525183,
      "grad_norm": 1.1937509499015742,
      "learning_rate": 3.1188743290723246e-06,
      "loss": 0.28815526962280275,
      "num_input_tokens_seen": 302960856,
      "step": 18500,
      "train_runtime": 35618.5302,
      "train_tokens_per_second": 8505.709
    },
    {
      "epoch": 1.2459601400484783,
      "grad_norm": 1.0138993766910533,
      "learning_rate": 3.116424775327882e-06,
      "loss": 0.2984570026397705,
      "num_input_tokens_seen": 303042776,
      "step": 18505,
      "train_runtime": 35628.2294,
      "train_tokens_per_second": 8505.693
    },
    {
      "epoch": 1.2462967950444384,
      "grad_norm": 0.9737806302326928,
      "learning_rate": 3.1139757483192423e-06,
      "loss": 0.25830845832824706,
      "num_input_tokens_seen": 303124696,
      "step": 18510,
      "train_runtime": 35637.7338,
      "train_tokens_per_second": 8505.723
    },
    {
      "epoch": 1.2466334500403986,
      "grad_norm": 0.9983580217328789,
      "learning_rate": 3.1115272487312727e-06,
      "loss": 0.26416616439819335,
      "num_input_tokens_seen": 303206616,
      "step": 18515,
      "train_runtime": 35647.2363,
      "train_tokens_per_second": 8505.754
    },
    {
      "epoch": 1.2469701050363589,
      "grad_norm": 1.1315089858037428,
      "learning_rate": 3.109079277248684e-06,
      "loss": 0.26569571495056155,
      "num_input_tokens_seen": 303288536,
      "step": 18520,
      "train_runtime": 35656.7449,
      "train_tokens_per_second": 8505.783
    },
    {
      "epoch": 1.2473067600323189,
      "grad_norm": 1.2171950776746876,
      "learning_rate": 3.1066318345560463e-06,
      "loss": 0.328033971786499,
      "num_input_tokens_seen": 303369160,
      "step": 18525,
      "train_runtime": 35666.867,
      "train_tokens_per_second": 8505.63
    },
    {
      "epoch": 1.247643415028279,
      "grad_norm": 1.1313563189744147,
      "learning_rate": 3.1041849213377782e-06,
      "loss": 0.3405263900756836,
      "num_input_tokens_seen": 303451080,
      "step": 18530,
      "train_runtime": 35676.3757,
      "train_tokens_per_second": 8505.659
    },
    {
      "epoch": 1.2479800700242392,
      "grad_norm": 1.2266419158360016,
      "learning_rate": 3.101738538278154e-06,
      "loss": 0.29714260101318357,
      "num_input_tokens_seen": 303533000,
      "step": 18535,
      "train_runtime": 35685.8793,
      "train_tokens_per_second": 8505.689
    },
    {
      "epoch": 1.2483167250201994,
      "grad_norm": 1.0164339966546836,
      "learning_rate": 3.0992926860612914e-06,
      "loss": 0.2788541793823242,
      "num_input_tokens_seen": 303614920,
      "step": 18540,
      "train_runtime": 35695.3964,
      "train_tokens_per_second": 8505.716
    },
    {
      "epoch": 1.2486533800161594,
      "grad_norm": 1.2829453669885635,
      "learning_rate": 3.0968473653711693e-06,
      "loss": 0.31631221771240237,
      "num_input_tokens_seen": 303696840,
      "step": 18545,
      "train_runtime": 35704.9059,
      "train_tokens_per_second": 8505.745
    },
    {
      "epoch": 1.2489900350121195,
      "grad_norm": 0.994307958173931,
      "learning_rate": 3.094402576891613e-06,
      "loss": 0.29046926498413084,
      "num_input_tokens_seen": 303778760,
      "step": 18550,
      "train_runtime": 35714.4111,
      "train_tokens_per_second": 8505.775
    },
    {
      "epoch": 1.2493266900080797,
      "grad_norm": 1.1463060980715771,
      "learning_rate": 3.0919583213063004e-06,
      "loss": 0.30491325855255125,
      "num_input_tokens_seen": 303859832,
      "step": 18555,
      "train_runtime": 35724.226,
      "train_tokens_per_second": 8505.708
    },
    {
      "epoch": 1.24966334500404,
      "grad_norm": 1.1121529944616018,
      "learning_rate": 3.0895145992987574e-06,
      "loss": 0.29411802291870115,
      "num_input_tokens_seen": 303941752,
      "step": 18560,
      "train_runtime": 35733.728,
      "train_tokens_per_second": 8505.739
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.9142782083484386,
      "learning_rate": 3.087071411552366e-06,
      "loss": 0.2769670009613037,
      "num_input_tokens_seen": 304023672,
      "step": 18565,
      "train_runtime": 35743.2423,
      "train_tokens_per_second": 8505.766
    },
    {
      "epoch": 1.25033665499596,
      "grad_norm": 1.1109458084812396,
      "learning_rate": 3.084628758750353e-06,
      "loss": 0.3088361740112305,
      "num_input_tokens_seen": 304105592,
      "step": 18570,
      "train_runtime": 35753.0248,
      "train_tokens_per_second": 8505.73
    },
    {
      "epoch": 1.2506733099919203,
      "grad_norm": 1.2097499523544142,
      "learning_rate": 3.082186641575801e-06,
      "loss": 0.31412630081176757,
      "num_input_tokens_seen": 304187512,
      "step": 18575,
      "train_runtime": 35762.529,
      "train_tokens_per_second": 8505.761
    },
    {
      "epoch": 1.2510099649878805,
      "grad_norm": 0.9377230797424457,
      "learning_rate": 3.0797450607116384e-06,
      "loss": 0.2729046106338501,
      "num_input_tokens_seen": 304269432,
      "step": 18580,
      "train_runtime": 35772.0341,
      "train_tokens_per_second": 8505.791
    },
    {
      "epoch": 1.2513466199838406,
      "grad_norm": 1.838916613424329,
      "learning_rate": 3.0773040168406475e-06,
      "loss": 0.30835561752319335,
      "num_input_tokens_seen": 304351352,
      "step": 18585,
      "train_runtime": 35781.5477,
      "train_tokens_per_second": 8505.819
    },
    {
      "epoch": 1.2516832749798006,
      "grad_norm": 1.2780608181977815,
      "learning_rate": 3.074863510645456e-06,
      "loss": 0.3083754301071167,
      "num_input_tokens_seen": 304433272,
      "step": 18590,
      "train_runtime": 35791.0619,
      "train_tokens_per_second": 8505.846
    },
    {
      "epoch": 1.2520199299757608,
      "grad_norm": 0.9510385833597766,
      "learning_rate": 3.072423542808546e-06,
      "loss": 0.2954937696456909,
      "num_input_tokens_seen": 304515192,
      "step": 18595,
      "train_runtime": 35800.5781,
      "train_tokens_per_second": 8505.874
    },
    {
      "epoch": 1.252356584971721,
      "grad_norm": 1.1366615341201542,
      "learning_rate": 3.0699841140122443e-06,
      "loss": 0.28786616325378417,
      "num_input_tokens_seen": 304597112,
      "step": 18600,
      "train_runtime": 35810.1222,
      "train_tokens_per_second": 8505.894
    },
    {
      "epoch": 1.2526932399676811,
      "grad_norm": 0.9265638886734409,
      "learning_rate": 3.0675452249387327e-06,
      "loss": 0.25420689582824707,
      "num_input_tokens_seen": 304679032,
      "step": 18605,
      "train_runtime": 35819.6553,
      "train_tokens_per_second": 8505.917
    },
    {
      "epoch": 1.2530298949636411,
      "grad_norm": 1.1225123020842656,
      "learning_rate": 3.0651068762700354e-06,
      "loss": 0.32227439880371095,
      "num_input_tokens_seen": 304760952,
      "step": 18610,
      "train_runtime": 35829.4617,
      "train_tokens_per_second": 8505.876
    },
    {
      "epoch": 1.2533665499596014,
      "grad_norm": 1.1149573584224048,
      "learning_rate": 3.0626690686880313e-06,
      "loss": 0.2783778190612793,
      "num_input_tokens_seen": 304842872,
      "step": 18615,
      "train_runtime": 35838.9731,
      "train_tokens_per_second": 8505.904
    },
    {
      "epoch": 1.2537032049555616,
      "grad_norm": 1.2519110546415175,
      "learning_rate": 3.060231802874443e-06,
      "loss": 0.33163890838623045,
      "num_input_tokens_seen": 304924792,
      "step": 18620,
      "train_runtime": 35848.4908,
      "train_tokens_per_second": 8505.931
    },
    {
      "epoch": 1.2540398599515217,
      "grad_norm": 1.0236076817630135,
      "learning_rate": 3.0577950795108462e-06,
      "loss": 0.27165923118591306,
      "num_input_tokens_seen": 305006712,
      "step": 18625,
      "train_runtime": 35858.0132,
      "train_tokens_per_second": 8505.957
    },
    {
      "epoch": 1.2543765149474817,
      "grad_norm": 1.1056086122308444,
      "learning_rate": 3.055358899278662e-06,
      "loss": 0.2989553451538086,
      "num_input_tokens_seen": 305088632,
      "step": 18630,
      "train_runtime": 35867.5223,
      "train_tokens_per_second": 8505.986
    },
    {
      "epoch": 1.254713169943442,
      "grad_norm": 1.1126736831812474,
      "learning_rate": 3.052923262859162e-06,
      "loss": 0.3104963541030884,
      "num_input_tokens_seen": 305170552,
      "step": 18635,
      "train_runtime": 35877.0387,
      "train_tokens_per_second": 8506.013
    },
    {
      "epoch": 1.2550498249394022,
      "grad_norm": 0.9620283865411794,
      "learning_rate": 3.0504881709334603e-06,
      "loss": 0.2898105621337891,
      "num_input_tokens_seen": 305252472,
      "step": 18640,
      "train_runtime": 35886.5543,
      "train_tokens_per_second": 8506.04
    },
    {
      "epoch": 1.2553864799353622,
      "grad_norm": 1.0725090503456502,
      "learning_rate": 3.0480536241825263e-06,
      "loss": 0.2752610921859741,
      "num_input_tokens_seen": 305334392,
      "step": 18645,
      "train_runtime": 35896.068,
      "train_tokens_per_second": 8506.068
    },
    {
      "epoch": 1.2557231349313223,
      "grad_norm": 0.9699165675476208,
      "learning_rate": 3.0456196232871715e-06,
      "loss": 0.28852219581604005,
      "num_input_tokens_seen": 305416312,
      "step": 18650,
      "train_runtime": 35905.5711,
      "train_tokens_per_second": 8506.098
    },
    {
      "epoch": 1.2560597899272825,
      "grad_norm": 1.027312503696368,
      "learning_rate": 3.043186168928059e-06,
      "loss": 0.2867791175842285,
      "num_input_tokens_seen": 305498232,
      "step": 18655,
      "train_runtime": 35915.094,
      "train_tokens_per_second": 8506.124
    },
    {
      "epoch": 1.2563964449232428,
      "grad_norm": 1.1910280501642312,
      "learning_rate": 3.040753261785692e-06,
      "loss": 0.25563549995422363,
      "num_input_tokens_seen": 305580152,
      "step": 18660,
      "train_runtime": 35924.6154,
      "train_tokens_per_second": 8506.15
    },
    {
      "epoch": 1.2567330999192028,
      "grad_norm": 1.0893025476571072,
      "learning_rate": 3.038320902540429e-06,
      "loss": 0.29209160804748535,
      "num_input_tokens_seen": 305662072,
      "step": 18665,
      "train_runtime": 35934.1682,
      "train_tokens_per_second": 8506.168
    },
    {
      "epoch": 1.2570697549151628,
      "grad_norm": 1.134632247280001,
      "learning_rate": 3.0358890918724706e-06,
      "loss": 0.29150350093841554,
      "num_input_tokens_seen": 305743992,
      "step": 18670,
      "train_runtime": 35943.8747,
      "train_tokens_per_second": 8506.15
    },
    {
      "epoch": 1.257406409911123,
      "grad_norm": 1.0731096833274292,
      "learning_rate": 3.0334578304618646e-06,
      "loss": 0.28538222312927247,
      "num_input_tokens_seen": 305825896,
      "step": 18675,
      "train_runtime": 35953.9681,
      "train_tokens_per_second": 8506.04
    },
    {
      "epoch": 1.2577430649070833,
      "grad_norm": 1.2592030394276454,
      "learning_rate": 3.0310271189885037e-06,
      "loss": 0.32431080341339114,
      "num_input_tokens_seen": 305906072,
      "step": 18680,
      "train_runtime": 35963.5302,
      "train_tokens_per_second": 8506.008
    },
    {
      "epoch": 1.2580797199030433,
      "grad_norm": 1.1061346750046357,
      "learning_rate": 3.0285969581321328e-06,
      "loss": 0.2931052207946777,
      "num_input_tokens_seen": 305987992,
      "step": 18685,
      "train_runtime": 35973.0385,
      "train_tokens_per_second": 8506.037
    },
    {
      "epoch": 1.2584163748990034,
      "grad_norm": 1.1613899427746388,
      "learning_rate": 3.0261673485723343e-06,
      "loss": 0.3250977277755737,
      "num_input_tokens_seen": 306069912,
      "step": 18690,
      "train_runtime": 35982.5457,
      "train_tokens_per_second": 8506.066
    },
    {
      "epoch": 1.2587530298949636,
      "grad_norm": 1.0426170902667713,
      "learning_rate": 3.023738290988544e-06,
      "loss": 0.24356703758239745,
      "num_input_tokens_seen": 306151832,
      "step": 18695,
      "train_runtime": 35992.0568,
      "train_tokens_per_second": 8506.094
    },
    {
      "epoch": 1.2590896848909239,
      "grad_norm": 1.106758128029084,
      "learning_rate": 3.021309786060037e-06,
      "loss": 0.28958878517150877,
      "num_input_tokens_seen": 306233752,
      "step": 18700,
      "train_runtime": 36001.5634,
      "train_tokens_per_second": 8506.124
    },
    {
      "epoch": 1.259426339886884,
      "grad_norm": 1.1364919892999292,
      "learning_rate": 3.018881834465941e-06,
      "loss": 0.28104588985443113,
      "num_input_tokens_seen": 306315672,
      "step": 18705,
      "train_runtime": 36011.0858,
      "train_tokens_per_second": 8506.149
    },
    {
      "epoch": 1.259762994882844,
      "grad_norm": 1.123191150504549,
      "learning_rate": 3.016454436885221e-06,
      "loss": 0.2924778938293457,
      "num_input_tokens_seen": 306397592,
      "step": 18710,
      "train_runtime": 36020.6021,
      "train_tokens_per_second": 8506.176
    },
    {
      "epoch": 1.2600996498788042,
      "grad_norm": 1.1399926720195326,
      "learning_rate": 3.0140275939966934e-06,
      "loss": 0.31231257915496824,
      "num_input_tokens_seen": 306479512,
      "step": 18715,
      "train_runtime": 36030.1125,
      "train_tokens_per_second": 8506.205
    },
    {
      "epoch": 1.2604363048747644,
      "grad_norm": 1.104875562415364,
      "learning_rate": 3.0116013064790166e-06,
      "loss": 0.3195568323135376,
      "num_input_tokens_seen": 306561432,
      "step": 18720,
      "train_runtime": 36039.785,
      "train_tokens_per_second": 8506.195
    },
    {
      "epoch": 1.2607729598707245,
      "grad_norm": 1.1037184617978906,
      "learning_rate": 3.0091755750106937e-06,
      "loss": 0.3023014783859253,
      "num_input_tokens_seen": 306643352,
      "step": 18725,
      "train_runtime": 36049.4178,
      "train_tokens_per_second": 8506.194
    },
    {
      "epoch": 1.2611096148666845,
      "grad_norm": 0.9991165954794219,
      "learning_rate": 3.006750400270072e-06,
      "loss": 0.2497702121734619,
      "num_input_tokens_seen": 306725272,
      "step": 18730,
      "train_runtime": 36058.9396,
      "train_tokens_per_second": 8506.22
    },
    {
      "epoch": 1.2614462698626447,
      "grad_norm": 1.143708168215399,
      "learning_rate": 3.0043257829353466e-06,
      "loss": 0.3154584884643555,
      "num_input_tokens_seen": 306805872,
      "step": 18735,
      "train_runtime": 36068.5652,
      "train_tokens_per_second": 8506.185
    },
    {
      "epoch": 1.261782924858605,
      "grad_norm": 1.2108856849143,
      "learning_rate": 3.0019017236845504e-06,
      "loss": 0.2889856815338135,
      "num_input_tokens_seen": 306887792,
      "step": 18740,
      "train_runtime": 36078.0718,
      "train_tokens_per_second": 8506.214
    },
    {
      "epoch": 1.262119579854565,
      "grad_norm": 1.2435038225982695,
      "learning_rate": 2.9994782231955673e-06,
      "loss": 0.3114027500152588,
      "num_input_tokens_seen": 306969456,
      "step": 18745,
      "train_runtime": 36087.97,
      "train_tokens_per_second": 8506.144
    },
    {
      "epoch": 1.262456234850525,
      "grad_norm": 0.8787100365372356,
      "learning_rate": 2.9970552821461185e-06,
      "loss": 0.2691160202026367,
      "num_input_tokens_seen": 307051376,
      "step": 18750,
      "train_runtime": 36097.4766,
      "train_tokens_per_second": 8506.173
    },
    {
      "epoch": 1.2627928898464853,
      "grad_norm": 1.1755378090283002,
      "learning_rate": 2.994632901213775e-06,
      "loss": 0.2952919006347656,
      "num_input_tokens_seen": 307132800,
      "step": 18755,
      "train_runtime": 36107.3306,
      "train_tokens_per_second": 8506.107
    },
    {
      "epoch": 1.2631295448424456,
      "grad_norm": 1.1824576695033375,
      "learning_rate": 2.9922110810759443e-06,
      "loss": 0.28104071617126464,
      "num_input_tokens_seen": 307214720,
      "step": 18760,
      "train_runtime": 36116.8433,
      "train_tokens_per_second": 8506.134
    },
    {
      "epoch": 1.2634661998384056,
      "grad_norm": 1.302747455320205,
      "learning_rate": 2.989789822409883e-06,
      "loss": 0.31969451904296875,
      "num_input_tokens_seen": 307296640,
      "step": 18765,
      "train_runtime": 36126.5023,
      "train_tokens_per_second": 8506.128
    },
    {
      "epoch": 1.2638028548343656,
      "grad_norm": 1.1768114898287294,
      "learning_rate": 2.9873691258926864e-06,
      "loss": 0.313887882232666,
      "num_input_tokens_seen": 307378560,
      "step": 18770,
      "train_runtime": 36136.0126,
      "train_tokens_per_second": 8506.156
    },
    {
      "epoch": 1.2641395098303259,
      "grad_norm": 1.113907310687978,
      "learning_rate": 2.9849489922012985e-06,
      "loss": 0.289617133140564,
      "num_input_tokens_seen": 307460480,
      "step": 18775,
      "train_runtime": 36145.6722,
      "train_tokens_per_second": 8506.149
    },
    {
      "epoch": 1.264476164826286,
      "grad_norm": 1.0953947191654982,
      "learning_rate": 2.982529422012498e-06,
      "loss": 0.31287901401519774,
      "num_input_tokens_seen": 307542400,
      "step": 18780,
      "train_runtime": 36155.1802,
      "train_tokens_per_second": 8506.178
    },
    {
      "epoch": 1.2648128198222461,
      "grad_norm": 1.155144880765586,
      "learning_rate": 2.9801104160029127e-06,
      "loss": 0.25144343376159667,
      "num_input_tokens_seen": 307624320,
      "step": 18785,
      "train_runtime": 36164.6883,
      "train_tokens_per_second": 8506.207
    },
    {
      "epoch": 1.2651494748182062,
      "grad_norm": 1.228560207947634,
      "learning_rate": 2.9776919748490063e-06,
      "loss": 0.302061653137207,
      "num_input_tokens_seen": 307706240,
      "step": 18790,
      "train_runtime": 36174.1855,
      "train_tokens_per_second": 8506.238
    },
    {
      "epoch": 1.2654861298141664,
      "grad_norm": 1.0876944486765898,
      "learning_rate": 2.975274099227092e-06,
      "loss": 0.29230399131774903,
      "num_input_tokens_seen": 307788160,
      "step": 18795,
      "train_runtime": 36183.6907,
      "train_tokens_per_second": 8506.268
    },
    {
      "epoch": 1.2658227848101267,
      "grad_norm": 1.1756712333356847,
      "learning_rate": 2.9728567898133183e-06,
      "loss": 0.29855852127075194,
      "num_input_tokens_seen": 307870080,
      "step": 18800,
      "train_runtime": 36193.1935,
      "train_tokens_per_second": 8506.298
    },
    {
      "epoch": 1.2661594398060867,
      "grad_norm": 1.0325966661077035,
      "learning_rate": 2.9704400472836816e-06,
      "loss": 0.2918875694274902,
      "num_input_tokens_seen": 307952000,
      "step": 18805,
      "train_runtime": 36202.6954,
      "train_tokens_per_second": 8506.328
    },
    {
      "epoch": 1.266496094802047,
      "grad_norm": 1.1694335634339696,
      "learning_rate": 2.9680238723140108e-06,
      "loss": 0.31146678924560545,
      "num_input_tokens_seen": 308033920,
      "step": 18810,
      "train_runtime": 36212.2079,
      "train_tokens_per_second": 8506.356
    },
    {
      "epoch": 1.266832749798007,
      "grad_norm": 1.1448046345079201,
      "learning_rate": 2.9656082655799866e-06,
      "loss": 0.295251727104187,
      "num_input_tokens_seen": 308115840,
      "step": 18815,
      "train_runtime": 36221.7183,
      "train_tokens_per_second": 8506.384
    },
    {
      "epoch": 1.2671694047939672,
      "grad_norm": 1.3441283285919643,
      "learning_rate": 2.9631932277571225e-06,
      "loss": 0.30377748012542727,
      "num_input_tokens_seen": 308197760,
      "step": 18820,
      "train_runtime": 36231.3864,
      "train_tokens_per_second": 8506.375
    },
    {
      "epoch": 1.2675060597899273,
      "grad_norm": 1.1474067162823098,
      "learning_rate": 2.9607787595207794e-06,
      "loss": 0.2778609752655029,
      "num_input_tokens_seen": 308279680,
      "step": 18825,
      "train_runtime": 36240.8916,
      "train_tokens_per_second": 8506.404
    },
    {
      "epoch": 1.2678427147858875,
      "grad_norm": 1.172526946862639,
      "learning_rate": 2.9583648615461514e-06,
      "loss": 0.27657880783081057,
      "num_input_tokens_seen": 308361600,
      "step": 18830,
      "train_runtime": 36250.4083,
      "train_tokens_per_second": 8506.431
    },
    {
      "epoch": 1.2681793697818475,
      "grad_norm": 1.1791757698773127,
      "learning_rate": 2.955951534508281e-06,
      "loss": 0.31696572303771975,
      "num_input_tokens_seen": 308443520,
      "step": 18835,
      "train_runtime": 36259.9225,
      "train_tokens_per_second": 8506.458
    },
    {
      "epoch": 1.2685160247778078,
      "grad_norm": 1.0008235765386049,
      "learning_rate": 2.9535387790820452e-06,
      "loss": 0.2801079750061035,
      "num_input_tokens_seen": 308525440,
      "step": 18840,
      "train_runtime": 36269.4317,
      "train_tokens_per_second": 8506.487
    },
    {
      "epoch": 1.2688526797737678,
      "grad_norm": 1.0117971608783696,
      "learning_rate": 2.951126595942165e-06,
      "loss": 0.29158458709716795,
      "num_input_tokens_seen": 308607360,
      "step": 18845,
      "train_runtime": 36278.9468,
      "train_tokens_per_second": 8506.514
    },
    {
      "epoch": 1.269189334769728,
      "grad_norm": 1.2034975206504452,
      "learning_rate": 2.948714985763197e-06,
      "loss": 0.2915506839752197,
      "num_input_tokens_seen": 308689280,
      "step": 18850,
      "train_runtime": 36288.4573,
      "train_tokens_per_second": 8506.542
    },
    {
      "epoch": 1.269525989765688,
      "grad_norm": 1.2735184955645928,
      "learning_rate": 2.9463039492195445e-06,
      "loss": 0.3226282835006714,
      "num_input_tokens_seen": 308771200,
      "step": 18855,
      "train_runtime": 36297.9722,
      "train_tokens_per_second": 8506.569
    },
    {
      "epoch": 1.2698626447616483,
      "grad_norm": 1.181558724352968,
      "learning_rate": 2.9438934869854417e-06,
      "loss": 0.2903972864151001,
      "num_input_tokens_seen": 308853120,
      "step": 18860,
      "train_runtime": 36307.4789,
      "train_tokens_per_second": 8506.598
    },
    {
      "epoch": 1.2701992997576084,
      "grad_norm": 1.2757536570496235,
      "learning_rate": 2.9414835997349705e-06,
      "loss": 0.2893181324005127,
      "num_input_tokens_seen": 308935040,
      "step": 18865,
      "train_runtime": 36316.9944,
      "train_tokens_per_second": 8506.625
    },
    {
      "epoch": 1.2705359547535686,
      "grad_norm": 1.0873664655479678,
      "learning_rate": 2.939074288142045e-06,
      "loss": 0.30501327514648435,
      "num_input_tokens_seen": 309016960,
      "step": 18870,
      "train_runtime": 36326.5038,
      "train_tokens_per_second": 8506.653
    },
    {
      "epoch": 1.2708726097495286,
      "grad_norm": 1.1605626898217212,
      "learning_rate": 2.9366655528804257e-06,
      "loss": 0.2798912525177002,
      "num_input_tokens_seen": 309098880,
      "step": 18875,
      "train_runtime": 36336.0122,
      "train_tokens_per_second": 8506.681
    },
    {
      "epoch": 1.271209264745489,
      "grad_norm": 1.07208983931629,
      "learning_rate": 2.934257394623702e-06,
      "loss": 0.27517173290252683,
      "num_input_tokens_seen": 309180800,
      "step": 18880,
      "train_runtime": 36345.6722,
      "train_tokens_per_second": 8506.674
    },
    {
      "epoch": 1.271545919741449,
      "grad_norm": 0.9245092675466197,
      "learning_rate": 2.9318498140453134e-06,
      "loss": 0.29869709014892576,
      "num_input_tokens_seen": 309262720,
      "step": 18885,
      "train_runtime": 36355.1858,
      "train_tokens_per_second": 8506.702
    },
    {
      "epoch": 1.2718825747374092,
      "grad_norm": 1.1571991734155311,
      "learning_rate": 2.929442811818527e-06,
      "loss": 0.2845757007598877,
      "num_input_tokens_seen": 309344640,
      "step": 18890,
      "train_runtime": 36364.7074,
      "train_tokens_per_second": 8506.727
    },
    {
      "epoch": 1.2722192297333692,
      "grad_norm": 1.1414012174846777,
      "learning_rate": 2.927036388616457e-06,
      "loss": 0.2847564697265625,
      "num_input_tokens_seen": 309426560,
      "step": 18895,
      "train_runtime": 36374.4453,
      "train_tokens_per_second": 8506.702
    },
    {
      "epoch": 1.2725558847293295,
      "grad_norm": 1.1936847869210818,
      "learning_rate": 2.9246305451120494e-06,
      "loss": 0.3074820518493652,
      "num_input_tokens_seen": 309508480,
      "step": 18900,
      "train_runtime": 36384.081,
      "train_tokens_per_second": 8506.7
    },
    {
      "epoch": 1.2728925397252895,
      "grad_norm": 1.0213015570185955,
      "learning_rate": 2.922225281978095e-06,
      "loss": 0.2870392560958862,
      "num_input_tokens_seen": 309590400,
      "step": 18905,
      "train_runtime": 36393.5992,
      "train_tokens_per_second": 8506.727
    },
    {
      "epoch": 1.2732291947212497,
      "grad_norm": 1.1396064916623991,
      "learning_rate": 2.919820599887212e-06,
      "loss": 0.28125159740447997,
      "num_input_tokens_seen": 309672320,
      "step": 18910,
      "train_runtime": 36403.1057,
      "train_tokens_per_second": 8506.756
    },
    {
      "epoch": 1.2735658497172098,
      "grad_norm": 1.2176595246010096,
      "learning_rate": 2.917416499511866e-06,
      "loss": 0.2900550365447998,
      "num_input_tokens_seen": 309754240,
      "step": 18915,
      "train_runtime": 36412.6248,
      "train_tokens_per_second": 8506.781
    },
    {
      "epoch": 1.27390250471317,
      "grad_norm": 1.008931625550489,
      "learning_rate": 2.9150129815243553e-06,
      "loss": 0.29366426467895507,
      "num_input_tokens_seen": 309836160,
      "step": 18920,
      "train_runtime": 36422.1249,
      "train_tokens_per_second": 8506.812
    },
    {
      "epoch": 1.27423915970913,
      "grad_norm": 1.1844075980943667,
      "learning_rate": 2.912610046596816e-06,
      "loss": 0.2912489652633667,
      "num_input_tokens_seen": 309918080,
      "step": 18925,
      "train_runtime": 36431.6389,
      "train_tokens_per_second": 8506.839
    },
    {
      "epoch": 1.2745758147050903,
      "grad_norm": 1.0142702145577054,
      "learning_rate": 2.91020769540122e-06,
      "loss": 0.2830839157104492,
      "num_input_tokens_seen": 310000000,
      "step": 18930,
      "train_runtime": 36441.1757,
      "train_tokens_per_second": 8506.861
    },
    {
      "epoch": 1.2749124697010503,
      "grad_norm": 0.9044088726158949,
      "learning_rate": 2.9078059286093795e-06,
      "loss": 0.30228638648986816,
      "num_input_tokens_seen": 310081920,
      "step": 18935,
      "train_runtime": 36450.8794,
      "train_tokens_per_second": 8506.843
    },
    {
      "epoch": 1.2752491246970106,
      "grad_norm": 1.1674068252132745,
      "learning_rate": 2.905404746892937e-06,
      "loss": 0.2947513580322266,
      "num_input_tokens_seen": 310163840,
      "step": 18940,
      "train_runtime": 36460.3824,
      "train_tokens_per_second": 8506.873
    },
    {
      "epoch": 1.2755857796929706,
      "grad_norm": 1.1818984191368798,
      "learning_rate": 2.903004150923382e-06,
      "loss": 0.3212242126464844,
      "num_input_tokens_seen": 310245760,
      "step": 18945,
      "train_runtime": 36470.0326,
      "train_tokens_per_second": 8506.868
    },
    {
      "epoch": 1.2759224346889309,
      "grad_norm": 0.9029979554216733,
      "learning_rate": 2.9006041413720252e-06,
      "loss": 0.2987320899963379,
      "num_input_tokens_seen": 310327680,
      "step": 18950,
      "train_runtime": 36479.5482,
      "train_tokens_per_second": 8506.895
    },
    {
      "epoch": 1.2762590896848909,
      "grad_norm": 1.1672455637414259,
      "learning_rate": 2.8982047189100276e-06,
      "loss": 0.31108062267303466,
      "num_input_tokens_seen": 310409600,
      "step": 18955,
      "train_runtime": 36489.0473,
      "train_tokens_per_second": 8506.925
    },
    {
      "epoch": 1.2765957446808511,
      "grad_norm": 1.2341805725123858,
      "learning_rate": 2.895805884208378e-06,
      "loss": 0.3089598178863525,
      "num_input_tokens_seen": 310491376,
      "step": 18960,
      "train_runtime": 36499.0461,
      "train_tokens_per_second": 8506.835
    },
    {
      "epoch": 1.2769323996768112,
      "grad_norm": 1.0470197806575339,
      "learning_rate": 2.8934076379379016e-06,
      "loss": 0.30025768280029297,
      "num_input_tokens_seen": 310573296,
      "step": 18965,
      "train_runtime": 36508.5599,
      "train_tokens_per_second": 8506.862
    },
    {
      "epoch": 1.2772690546727714,
      "grad_norm": 0.9973947936571621,
      "learning_rate": 2.891009980769266e-06,
      "loss": 0.27584807872772216,
      "num_input_tokens_seen": 310655216,
      "step": 18970,
      "train_runtime": 36518.064,
      "train_tokens_per_second": 8506.892
    },
    {
      "epoch": 1.2776057096687314,
      "grad_norm": 1.1632745887017226,
      "learning_rate": 2.88861291337296e-06,
      "loss": 0.28819003105163576,
      "num_input_tokens_seen": 310737136,
      "step": 18975,
      "train_runtime": 36527.5806,
      "train_tokens_per_second": 8506.918
    },
    {
      "epoch": 1.2779423646646917,
      "grad_norm": 1.239887822802284,
      "learning_rate": 2.8862164364193236e-06,
      "loss": 0.27062673568725587,
      "num_input_tokens_seen": 310819056,
      "step": 18980,
      "train_runtime": 36537.3744,
      "train_tokens_per_second": 8506.88
    },
    {
      "epoch": 1.2782790196606517,
      "grad_norm": 1.1194123800415516,
      "learning_rate": 2.8838205505785168e-06,
      "loss": 0.27347393035888673,
      "num_input_tokens_seen": 310900976,
      "step": 18985,
      "train_runtime": 36546.929,
      "train_tokens_per_second": 8506.897
    },
    {
      "epoch": 1.278615674656612,
      "grad_norm": 1.038014258904953,
      "learning_rate": 2.881425256520547e-06,
      "loss": 0.28636989593505857,
      "num_input_tokens_seen": 310982896,
      "step": 18990,
      "train_runtime": 36556.435,
      "train_tokens_per_second": 8506.926
    },
    {
      "epoch": 1.278952329652572,
      "grad_norm": 1.0661084610358247,
      "learning_rate": 2.8790305549152487e-06,
      "loss": 0.2536468505859375,
      "num_input_tokens_seen": 311064816,
      "step": 18995,
      "train_runtime": 36565.9364,
      "train_tokens_per_second": 8506.956
    },
    {
      "epoch": 1.2792889846485322,
      "grad_norm": 1.130077362657665,
      "learning_rate": 2.876636446432292e-06,
      "loss": 0.3113153696060181,
      "num_input_tokens_seen": 311146736,
      "step": 19000,
      "train_runtime": 36575.4469,
      "train_tokens_per_second": 8506.984
    },
    {
      "epoch": 1.2796256396444923,
      "grad_norm": 1.069035612247097,
      "learning_rate": 2.8742429317411826e-06,
      "loss": 0.2974132299423218,
      "num_input_tokens_seen": 311228656,
      "step": 19005,
      "train_runtime": 36584.9709,
      "train_tokens_per_second": 8507.008
    },
    {
      "epoch": 1.2799622946404525,
      "grad_norm": 1.1675234963564691,
      "learning_rate": 2.8718500115112585e-06,
      "loss": 0.2980727910995483,
      "num_input_tokens_seen": 311310576,
      "step": 19010,
      "train_runtime": 36594.4734,
      "train_tokens_per_second": 8507.038
    },
    {
      "epoch": 1.2802989496364126,
      "grad_norm": 1.141841941740759,
      "learning_rate": 2.869457686411691e-06,
      "loss": 0.32432682514190675,
      "num_input_tokens_seen": 311392496,
      "step": 19015,
      "train_runtime": 36603.984,
      "train_tokens_per_second": 8507.066
    },
    {
      "epoch": 1.2806356046323728,
      "grad_norm": 1.0301204368954795,
      "learning_rate": 2.8670659571114912e-06,
      "loss": 0.26265456676483157,
      "num_input_tokens_seen": 311474416,
      "step": 19020,
      "train_runtime": 36613.4924,
      "train_tokens_per_second": 8507.094
    },
    {
      "epoch": 1.2809722596283328,
      "grad_norm": 1.1121365218432955,
      "learning_rate": 2.864674824279491e-06,
      "loss": 0.29705014228820803,
      "num_input_tokens_seen": 311556336,
      "step": 19025,
      "train_runtime": 36623.0638,
      "train_tokens_per_second": 8507.107
    },
    {
      "epoch": 1.281308914624293,
      "grad_norm": 1.1437985155925694,
      "learning_rate": 2.862284288584371e-06,
      "loss": 0.28898468017578127,
      "num_input_tokens_seen": 311638256,
      "step": 19030,
      "train_runtime": 36632.5781,
      "train_tokens_per_second": 8507.134
    },
    {
      "epoch": 1.2816455696202531,
      "grad_norm": 1.1761427918204173,
      "learning_rate": 2.8598943506946297e-06,
      "loss": 0.31907033920288086,
      "num_input_tokens_seen": 311720176,
      "step": 19035,
      "train_runtime": 36642.0943,
      "train_tokens_per_second": 8507.16
    },
    {
      "epoch": 1.2819822246162134,
      "grad_norm": 1.0406773080073362,
      "learning_rate": 2.85750501127861e-06,
      "loss": 0.267558217048645,
      "num_input_tokens_seen": 311802096,
      "step": 19040,
      "train_runtime": 36651.6109,
      "train_tokens_per_second": 8507.187
    },
    {
      "epoch": 1.2823188796121734,
      "grad_norm": 0.8832702621289804,
      "learning_rate": 2.8551162710044827e-06,
      "loss": 0.29225926399230956,
      "num_input_tokens_seen": 311884016,
      "step": 19045,
      "train_runtime": 36661.1149,
      "train_tokens_per_second": 8507.216
    },
    {
      "epoch": 1.2826555346081336,
      "grad_norm": 1.1186351912066457,
      "learning_rate": 2.8527281305402503e-06,
      "loss": 0.3152174711227417,
      "num_input_tokens_seen": 311965936,
      "step": 19050,
      "train_runtime": 36670.6179,
      "train_tokens_per_second": 8507.245
    },
    {
      "epoch": 1.2829921896040937,
      "grad_norm": 1.2207843615489702,
      "learning_rate": 2.8503405905537496e-06,
      "loss": 0.2773649454116821,
      "num_input_tokens_seen": 312047856,
      "step": 19055,
      "train_runtime": 36680.1279,
      "train_tokens_per_second": 8507.273
    },
    {
      "epoch": 1.283328844600054,
      "grad_norm": 1.2035741634836117,
      "learning_rate": 2.847953651712647e-06,
      "loss": 0.30078859329223634,
      "num_input_tokens_seen": 312129776,
      "step": 19060,
      "train_runtime": 36689.9493,
      "train_tokens_per_second": 8507.228
    },
    {
      "epoch": 1.283665499596014,
      "grad_norm": 1.0625320072166797,
      "learning_rate": 2.8455673146844426e-06,
      "loss": 0.28293333053588865,
      "num_input_tokens_seen": 312211696,
      "step": 19065,
      "train_runtime": 36699.4557,
      "train_tokens_per_second": 8507.257
    },
    {
      "epoch": 1.2840021545919742,
      "grad_norm": 1.0026330568866908,
      "learning_rate": 2.8431815801364717e-06,
      "loss": 0.29729547500610354,
      "num_input_tokens_seen": 312293616,
      "step": 19070,
      "train_runtime": 36708.962,
      "train_tokens_per_second": 8507.285
    },
    {
      "epoch": 1.2843388095879342,
      "grad_norm": 1.334897862783168,
      "learning_rate": 2.8407964487358906e-06,
      "loss": 0.3180110216140747,
      "num_input_tokens_seen": 312375536,
      "step": 19075,
      "train_runtime": 36718.4693,
      "train_tokens_per_second": 8507.314
    },
    {
      "epoch": 1.2846754645838945,
      "grad_norm": 1.1476785660275364,
      "learning_rate": 2.838411921149702e-06,
      "loss": 0.27646303176879883,
      "num_input_tokens_seen": 312457456,
      "step": 19080,
      "train_runtime": 36727.9876,
      "train_tokens_per_second": 8507.339
    },
    {
      "epoch": 1.2850121195798545,
      "grad_norm": 0.9585570400957085,
      "learning_rate": 2.836027998044722e-06,
      "loss": 0.27193446159362794,
      "num_input_tokens_seen": 312538824,
      "step": 19085,
      "train_runtime": 36737.7888,
      "train_tokens_per_second": 8507.285
    },
    {
      "epoch": 1.2853487745758148,
      "grad_norm": 1.2242018095482614,
      "learning_rate": 2.8336446800876143e-06,
      "loss": 0.28291568756103513,
      "num_input_tokens_seen": 312620744,
      "step": 19090,
      "train_runtime": 36747.2955,
      "train_tokens_per_second": 8507.313
    },
    {
      "epoch": 1.2856854295717748,
      "grad_norm": 1.1487873457441866,
      "learning_rate": 2.8312619679448637e-06,
      "loss": 0.29815614223480225,
      "num_input_tokens_seen": 312702664,
      "step": 19095,
      "train_runtime": 36756.8049,
      "train_tokens_per_second": 8507.341
    },
    {
      "epoch": 1.286022084567735,
      "grad_norm": 1.061745929420883,
      "learning_rate": 2.8288798622827884e-06,
      "loss": 0.2957751274108887,
      "num_input_tokens_seen": 312784584,
      "step": 19100,
      "train_runtime": 36766.4145,
      "train_tokens_per_second": 8507.345
    },
    {
      "epoch": 1.286358739563695,
      "grad_norm": 1.1840841769614494,
      "learning_rate": 2.826498363767536e-06,
      "loss": 0.3011630058288574,
      "num_input_tokens_seen": 312866504,
      "step": 19105,
      "train_runtime": 36775.9205,
      "train_tokens_per_second": 8507.374
    },
    {
      "epoch": 1.2866953945596553,
      "grad_norm": 1.0186736348585337,
      "learning_rate": 2.8241174730650855e-06,
      "loss": 0.29763073921203614,
      "num_input_tokens_seen": 312948424,
      "step": 19110,
      "train_runtime": 36785.4324,
      "train_tokens_per_second": 8507.401
    },
    {
      "epoch": 1.2870320495556153,
      "grad_norm": 1.0094823159401736,
      "learning_rate": 2.821737190841244e-06,
      "loss": 0.2782706022262573,
      "num_input_tokens_seen": 313030344,
      "step": 19115,
      "train_runtime": 36794.9363,
      "train_tokens_per_second": 8507.43
    },
    {
      "epoch": 1.2873687045515756,
      "grad_norm": 1.2709081211620508,
      "learning_rate": 2.8193575177616537e-06,
      "loss": 0.30514423847198485,
      "num_input_tokens_seen": 313112264,
      "step": 19120,
      "train_runtime": 36804.5117,
      "train_tokens_per_second": 8507.442
    },
    {
      "epoch": 1.2877053595475356,
      "grad_norm": 0.9585025473057937,
      "learning_rate": 2.816978454491777e-06,
      "loss": 0.2636375427246094,
      "num_input_tokens_seen": 313194184,
      "step": 19125,
      "train_runtime": 36814.0426,
      "train_tokens_per_second": 8507.465
    },
    {
      "epoch": 1.2880420145434959,
      "grad_norm": 1.1333530367789266,
      "learning_rate": 2.814600001696919e-06,
      "loss": 0.2942920446395874,
      "num_input_tokens_seen": 313276104,
      "step": 19130,
      "train_runtime": 36823.6225,
      "train_tokens_per_second": 8507.477
    },
    {
      "epoch": 1.288378669539456,
      "grad_norm": 1.0737203321121769,
      "learning_rate": 2.812222160042196e-06,
      "loss": 0.29947853088378906,
      "num_input_tokens_seen": 313358024,
      "step": 19135,
      "train_runtime": 36833.1249,
      "train_tokens_per_second": 8507.506
    },
    {
      "epoch": 1.2887153245354162,
      "grad_norm": 0.9970737041484858,
      "learning_rate": 2.809844930192573e-06,
      "loss": 0.2745089292526245,
      "num_input_tokens_seen": 313439944,
      "step": 19140,
      "train_runtime": 36842.6408,
      "train_tokens_per_second": 8507.532
    },
    {
      "epoch": 1.2890519795313762,
      "grad_norm": 1.1637134804451874,
      "learning_rate": 2.807468312812831e-06,
      "loss": 0.3033177852630615,
      "num_input_tokens_seen": 313521864,
      "step": 19145,
      "train_runtime": 36852.2237,
      "train_tokens_per_second": 8507.543
    },
    {
      "epoch": 1.2893886345273364,
      "grad_norm": 1.1305651582092031,
      "learning_rate": 2.805092308567584e-06,
      "loss": 0.2706321716308594,
      "num_input_tokens_seen": 313603784,
      "step": 19150,
      "train_runtime": 36861.7991,
      "train_tokens_per_second": 8507.555
    },
    {
      "epoch": 1.2897252895232965,
      "grad_norm": 1.0910251237468707,
      "learning_rate": 2.802716918121273e-06,
      "loss": 0.2898510456085205,
      "num_input_tokens_seen": 313685704,
      "step": 19155,
      "train_runtime": 36871.5711,
      "train_tokens_per_second": 8507.522
    },
    {
      "epoch": 1.2900619445192567,
      "grad_norm": 1.3972170984624273,
      "learning_rate": 2.8003421421381704e-06,
      "loss": 0.3070762872695923,
      "num_input_tokens_seen": 313767624,
      "step": 19160,
      "train_runtime": 36881.0776,
      "train_tokens_per_second": 8507.55
    },
    {
      "epoch": 1.2903985995152167,
      "grad_norm": 1.110801447660655,
      "learning_rate": 2.797967981282371e-06,
      "loss": 0.29026877880096436,
      "num_input_tokens_seen": 313849544,
      "step": 19165,
      "train_runtime": 36890.5811,
      "train_tokens_per_second": 8507.579
    },
    {
      "epoch": 1.290735254511177,
      "grad_norm": 1.1861939392211953,
      "learning_rate": 2.795594436217808e-06,
      "loss": 0.28043394088745116,
      "num_input_tokens_seen": 313931464,
      "step": 19170,
      "train_runtime": 36900.0871,
      "train_tokens_per_second": 8507.608
    },
    {
      "epoch": 1.291071909507137,
      "grad_norm": 1.0054479777999938,
      "learning_rate": 2.7932215076082277e-06,
      "loss": 0.2721073865890503,
      "num_input_tokens_seen": 314013384,
      "step": 19175,
      "train_runtime": 36909.8727,
      "train_tokens_per_second": 8507.572
    },
    {
      "epoch": 1.2914085645030973,
      "grad_norm": 1.1525763520945744,
      "learning_rate": 2.790849196117219e-06,
      "loss": 0.2870074033737183,
      "num_input_tokens_seen": 314095304,
      "step": 19180,
      "train_runtime": 36919.3764,
      "train_tokens_per_second": 8507.6
    },
    {
      "epoch": 1.2917452194990573,
      "grad_norm": 1.059784405338845,
      "learning_rate": 2.7884775024081856e-06,
      "loss": 0.2953347682952881,
      "num_input_tokens_seen": 314176832,
      "step": 19185,
      "train_runtime": 36929.2504,
      "train_tokens_per_second": 8507.533
    },
    {
      "epoch": 1.2920818744950175,
      "grad_norm": 0.9785977520508237,
      "learning_rate": 2.786106427144367e-06,
      "loss": 0.2885937213897705,
      "num_input_tokens_seen": 314258752,
      "step": 19190,
      "train_runtime": 36938.7502,
      "train_tokens_per_second": 8507.563
    },
    {
      "epoch": 1.2924185294909776,
      "grad_norm": 1.0309396473954322,
      "learning_rate": 2.783735970988828e-06,
      "loss": 0.25931658744812014,
      "num_input_tokens_seen": 314340672,
      "step": 19195,
      "train_runtime": 36948.2649,
      "train_tokens_per_second": 8507.59
    },
    {
      "epoch": 1.2927551844869378,
      "grad_norm": 1.2222185034990127,
      "learning_rate": 2.7813661346044566e-06,
      "loss": 0.31450700759887695,
      "num_input_tokens_seen": 314422592,
      "step": 19200,
      "train_runtime": 36957.7681,
      "train_tokens_per_second": 8507.619
    },
    {
      "epoch": 1.2930918394828979,
      "grad_norm": 1.2155083841905896,
      "learning_rate": 2.7789969186539713e-06,
      "loss": 0.3074111223220825,
      "num_input_tokens_seen": 314504512,
      "step": 19205,
      "train_runtime": 36967.2718,
      "train_tokens_per_second": 8507.647
    },
    {
      "epoch": 1.293428494478858,
      "grad_norm": 1.0560371058239175,
      "learning_rate": 2.776628323799915e-06,
      "loss": 0.2712681770324707,
      "num_input_tokens_seen": 314586432,
      "step": 19210,
      "train_runtime": 36976.7751,
      "train_tokens_per_second": 8507.676
    },
    {
      "epoch": 1.2937651494748181,
      "grad_norm": 0.8901642300912217,
      "learning_rate": 2.774260350704657e-06,
      "loss": 0.2697617769241333,
      "num_input_tokens_seen": 314668352,
      "step": 19215,
      "train_runtime": 36986.27,
      "train_tokens_per_second": 8507.707
    },
    {
      "epoch": 1.2941018044707784,
      "grad_norm": 1.322692107456363,
      "learning_rate": 2.771893000030398e-06,
      "loss": 0.29811973571777345,
      "num_input_tokens_seen": 314750272,
      "step": 19220,
      "train_runtime": 36996.0594,
      "train_tokens_per_second": 8507.67
    },
    {
      "epoch": 1.2944384594667384,
      "grad_norm": 1.1798321094086024,
      "learning_rate": 2.7695262724391526e-06,
      "loss": 0.30265002250671386,
      "num_input_tokens_seen": 314832192,
      "step": 19225,
      "train_runtime": 37005.5712,
      "train_tokens_per_second": 8507.697
    },
    {
      "epoch": 1.2947751144626987,
      "grad_norm": 1.227367700152121,
      "learning_rate": 2.767160168592777e-06,
      "loss": 0.3166023254394531,
      "num_input_tokens_seen": 314914112,
      "step": 19230,
      "train_runtime": 37015.0846,
      "train_tokens_per_second": 8507.724
    },
    {
      "epoch": 1.2951117694586587,
      "grad_norm": 1.1345608482257978,
      "learning_rate": 2.7647946891529355e-06,
      "loss": 0.2972844123840332,
      "num_input_tokens_seen": 314996032,
      "step": 19235,
      "train_runtime": 37024.5864,
      "train_tokens_per_second": 8507.753
    },
    {
      "epoch": 1.295448424454619,
      "grad_norm": 1.0510836422635301,
      "learning_rate": 2.7624298347811342e-06,
      "loss": 0.293131947517395,
      "num_input_tokens_seen": 315077952,
      "step": 19240,
      "train_runtime": 37034.1535,
      "train_tokens_per_second": 8507.767
    },
    {
      "epoch": 1.295785079450579,
      "grad_norm": 1.273547075485315,
      "learning_rate": 2.7600656061386945e-06,
      "loss": 0.299485969543457,
      "num_input_tokens_seen": 315159872,
      "step": 19245,
      "train_runtime": 37043.6671,
      "train_tokens_per_second": 8507.794
    },
    {
      "epoch": 1.2961217344465392,
      "grad_norm": 1.093376096337639,
      "learning_rate": 2.757702003886765e-06,
      "loss": 0.3041447639465332,
      "num_input_tokens_seen": 315241792,
      "step": 19250,
      "train_runtime": 37053.1746,
      "train_tokens_per_second": 8507.821
    },
    {
      "epoch": 1.2964583894424992,
      "grad_norm": 1.1884541030380125,
      "learning_rate": 2.7553390286863205e-06,
      "loss": 0.2923215389251709,
      "num_input_tokens_seen": 315323712,
      "step": 19255,
      "train_runtime": 37062.6873,
      "train_tokens_per_second": 8507.848
    },
    {
      "epoch": 1.2967950444384595,
      "grad_norm": 1.1659601853505477,
      "learning_rate": 2.752976681198159e-06,
      "loss": 0.2870516777038574,
      "num_input_tokens_seen": 315405632,
      "step": 19260,
      "train_runtime": 37072.1958,
      "train_tokens_per_second": 8507.876
    },
    {
      "epoch": 1.2971316994344195,
      "grad_norm": 1.1540632290994035,
      "learning_rate": 2.750614962082901e-06,
      "loss": 0.27757012844085693,
      "num_input_tokens_seen": 315487552,
      "step": 19265,
      "train_runtime": 37081.7156,
      "train_tokens_per_second": 8507.901
    },
    {
      "epoch": 1.2974683544303798,
      "grad_norm": 0.9943551323078834,
      "learning_rate": 2.748253872001e-06,
      "loss": 0.2903022766113281,
      "num_input_tokens_seen": 315569472,
      "step": 19270,
      "train_runtime": 37091.217,
      "train_tokens_per_second": 8507.93
    },
    {
      "epoch": 1.2978050094263398,
      "grad_norm": 1.240961603199762,
      "learning_rate": 2.74589341161272e-06,
      "loss": 0.2996648073196411,
      "num_input_tokens_seen": 315651392,
      "step": 19275,
      "train_runtime": 37100.717,
      "train_tokens_per_second": 8507.959
    },
    {
      "epoch": 1.2981416644223,
      "grad_norm": 0.9812113650411415,
      "learning_rate": 2.743533581578163e-06,
      "loss": 0.2800112724304199,
      "num_input_tokens_seen": 315733312,
      "step": 19280,
      "train_runtime": 37110.2173,
      "train_tokens_per_second": 8507.989
    },
    {
      "epoch": 1.29847831941826,
      "grad_norm": 1.4214975610630303,
      "learning_rate": 2.7411743825572407e-06,
      "loss": 0.34023303985595704,
      "num_input_tokens_seen": 315815232,
      "step": 19285,
      "train_runtime": 37119.7265,
      "train_tokens_per_second": 8508.016
    },
    {
      "epoch": 1.2988149744142203,
      "grad_norm": 1.2502458105319068,
      "learning_rate": 2.7388158152097015e-06,
      "loss": 0.3102290391921997,
      "num_input_tokens_seen": 315897152,
      "step": 19290,
      "train_runtime": 37129.2447,
      "train_tokens_per_second": 8508.041
    },
    {
      "epoch": 1.2991516294101804,
      "grad_norm": 1.0985337024839787,
      "learning_rate": 2.736457880195109e-06,
      "loss": 0.29321513175964353,
      "num_input_tokens_seen": 315979072,
      "step": 19295,
      "train_runtime": 37138.8355,
      "train_tokens_per_second": 8508.05
    },
    {
      "epoch": 1.2994882844061406,
      "grad_norm": 1.086164124810761,
      "learning_rate": 2.7341005781728523e-06,
      "loss": 0.2850809574127197,
      "num_input_tokens_seen": 316060992,
      "step": 19300,
      "train_runtime": 37148.3993,
      "train_tokens_per_second": 8508.065
    },
    {
      "epoch": 1.2998249394021006,
      "grad_norm": 1.2017514906909024,
      "learning_rate": 2.7317439098021444e-06,
      "loss": 0.2865572214126587,
      "num_input_tokens_seen": 316142912,
      "step": 19305,
      "train_runtime": 37157.9487,
      "train_tokens_per_second": 8508.083
    },
    {
      "epoch": 1.300161594398061,
      "grad_norm": 1.194959429978033,
      "learning_rate": 2.7293878757420185e-06,
      "loss": 0.2953918695449829,
      "num_input_tokens_seen": 316224832,
      "step": 19310,
      "train_runtime": 37167.4716,
      "train_tokens_per_second": 8508.107
    },
    {
      "epoch": 1.3004982493940211,
      "grad_norm": 1.2270019559694574,
      "learning_rate": 2.7270324766513306e-06,
      "loss": 0.3101369380950928,
      "num_input_tokens_seen": 316306752,
      "step": 19315,
      "train_runtime": 37176.9923,
      "train_tokens_per_second": 8508.132
    },
    {
      "epoch": 1.3008349043899812,
      "grad_norm": 1.2169267120731466,
      "learning_rate": 2.724677713188767e-06,
      "loss": 0.2844986438751221,
      "num_input_tokens_seen": 316388672,
      "step": 19320,
      "train_runtime": 37186.5177,
      "train_tokens_per_second": 8508.155
    },
    {
      "epoch": 1.3011715593859412,
      "grad_norm": 1.2335038127346225,
      "learning_rate": 2.722323586012821e-06,
      "loss": 0.2814591884613037,
      "num_input_tokens_seen": 316470592,
      "step": 19325,
      "train_runtime": 37196.0332,
      "train_tokens_per_second": 8508.181
    },
    {
      "epoch": 1.3015082143819015,
      "grad_norm": 1.2360966776478304,
      "learning_rate": 2.7199700957818252e-06,
      "loss": 0.2806647062301636,
      "num_input_tokens_seen": 316552512,
      "step": 19330,
      "train_runtime": 37205.5434,
      "train_tokens_per_second": 8508.208
    },
    {
      "epoch": 1.3018448693778617,
      "grad_norm": 1.1253383476584546,
      "learning_rate": 2.717617243153918e-06,
      "loss": 0.30061044692993166,
      "num_input_tokens_seen": 316634192,
      "step": 19335,
      "train_runtime": 37215.5561,
      "train_tokens_per_second": 8508.114
    },
    {
      "epoch": 1.3021815243738217,
      "grad_norm": 1.0399195484672183,
      "learning_rate": 2.715265028787073e-06,
      "loss": 0.2819680213928223,
      "num_input_tokens_seen": 316716112,
      "step": 19340,
      "train_runtime": 37225.0802,
      "train_tokens_per_second": 8508.138
    },
    {
      "epoch": 1.3025181793697818,
      "grad_norm": 1.493505578847941,
      "learning_rate": 2.7129134533390765e-06,
      "loss": 0.3211517810821533,
      "num_input_tokens_seen": 316798032,
      "step": 19345,
      "train_runtime": 37234.5832,
      "train_tokens_per_second": 8508.166
    },
    {
      "epoch": 1.302854834365742,
      "grad_norm": 1.4041540369781595,
      "learning_rate": 2.7105625174675404e-06,
      "loss": 0.28890109062194824,
      "num_input_tokens_seen": 316879952,
      "step": 19350,
      "train_runtime": 37244.0926,
      "train_tokens_per_second": 8508.194
    },
    {
      "epoch": 1.3031914893617023,
      "grad_norm": 1.0821203215080257,
      "learning_rate": 2.7082122218298957e-06,
      "loss": 0.27535090446472166,
      "num_input_tokens_seen": 316961872,
      "step": 19355,
      "train_runtime": 37253.6099,
      "train_tokens_per_second": 8508.219
    },
    {
      "epoch": 1.3035281443576623,
      "grad_norm": 1.155491281260749,
      "learning_rate": 2.7058625670833947e-06,
      "loss": 0.2837100028991699,
      "num_input_tokens_seen": 317043696,
      "step": 19360,
      "train_runtime": 37263.683,
      "train_tokens_per_second": 8508.115
    },
    {
      "epoch": 1.3038647993536223,
      "grad_norm": 1.109426177321046,
      "learning_rate": 2.70351355388511e-06,
      "loss": 0.3202249050140381,
      "num_input_tokens_seen": 317125616,
      "step": 19365,
      "train_runtime": 37273.1877,
      "train_tokens_per_second": 8508.143
    },
    {
      "epoch": 1.3042014543495826,
      "grad_norm": 1.0878781841666858,
      "learning_rate": 2.7011651828919407e-06,
      "loss": 0.3119924783706665,
      "num_input_tokens_seen": 317207536,
      "step": 19370,
      "train_runtime": 37282.6949,
      "train_tokens_per_second": 8508.171
    },
    {
      "epoch": 1.3045381093455428,
      "grad_norm": 1.19686445897332,
      "learning_rate": 2.698817454760593e-06,
      "loss": 0.3077390670776367,
      "num_input_tokens_seen": 317289456,
      "step": 19375,
      "train_runtime": 37292.202,
      "train_tokens_per_second": 8508.198
    },
    {
      "epoch": 1.3048747643415028,
      "grad_norm": 1.0051509797138742,
      "learning_rate": 2.696470370147611e-06,
      "loss": 0.28298900127410886,
      "num_input_tokens_seen": 317371376,
      "step": 19380,
      "train_runtime": 37301.7088,
      "train_tokens_per_second": 8508.226
    },
    {
      "epoch": 1.3052114193374629,
      "grad_norm": 1.0701268956189711,
      "learning_rate": 2.69412392970934e-06,
      "loss": 0.29651033878326416,
      "num_input_tokens_seen": 317453296,
      "step": 19385,
      "train_runtime": 37311.2321,
      "train_tokens_per_second": 8508.25
    },
    {
      "epoch": 1.3055480743334231,
      "grad_norm": 0.9632455999994367,
      "learning_rate": 2.6917781341019615e-06,
      "loss": 0.2986374616622925,
      "num_input_tokens_seen": 317535216,
      "step": 19390,
      "train_runtime": 37320.7483,
      "train_tokens_per_second": 8508.276
    },
    {
      "epoch": 1.3058847293293834,
      "grad_norm": 0.9947331557562157,
      "learning_rate": 2.689432983981467e-06,
      "loss": 0.2551482439041138,
      "num_input_tokens_seen": 317617136,
      "step": 19395,
      "train_runtime": 37330.2452,
      "train_tokens_per_second": 8508.306
    },
    {
      "epoch": 1.3062213843253434,
      "grad_norm": 1.0480121126751516,
      "learning_rate": 2.6870884800036713e-06,
      "loss": 0.3035200357437134,
      "num_input_tokens_seen": 317699056,
      "step": 19400,
      "train_runtime": 37339.746,
      "train_tokens_per_second": 8508.335
    },
    {
      "epoch": 1.3065580393213034,
      "grad_norm": 1.077789486822648,
      "learning_rate": 2.684744622824207e-06,
      "loss": 0.3006195783615112,
      "num_input_tokens_seen": 317780976,
      "step": 19405,
      "train_runtime": 37349.3255,
      "train_tokens_per_second": 8508.346
    },
    {
      "epoch": 1.3068946943172637,
      "grad_norm": 1.0754753119702645,
      "learning_rate": 2.6824014130985255e-06,
      "loss": 0.26563115119934083,
      "num_input_tokens_seen": 317862760,
      "step": 19410,
      "train_runtime": 37359.3203,
      "train_tokens_per_second": 8508.259
    },
    {
      "epoch": 1.307231349313224,
      "grad_norm": 1.0723138872328102,
      "learning_rate": 2.680058851481896e-06,
      "loss": 0.2885449409484863,
      "num_input_tokens_seen": 317944136,
      "step": 19415,
      "train_runtime": 37369.1358,
      "train_tokens_per_second": 8508.201
    },
    {
      "epoch": 1.307568004309184,
      "grad_norm": 1.3019773350583521,
      "learning_rate": 2.6777169386294154e-06,
      "loss": 0.30610480308532717,
      "num_input_tokens_seen": 318026056,
      "step": 19420,
      "train_runtime": 37378.6669,
      "train_tokens_per_second": 8508.224
    },
    {
      "epoch": 1.307904659305144,
      "grad_norm": 1.2412257941706129,
      "learning_rate": 2.6753756751959846e-06,
      "loss": 0.27954413890838625,
      "num_input_tokens_seen": 318107976,
      "step": 19425,
      "train_runtime": 37388.1761,
      "train_tokens_per_second": 8508.251
    },
    {
      "epoch": 1.3082413143011042,
      "grad_norm": 1.0828617126930002,
      "learning_rate": 2.6730350618363367e-06,
      "loss": 0.2956111907958984,
      "num_input_tokens_seen": 318189896,
      "step": 19430,
      "train_runtime": 37397.6866,
      "train_tokens_per_second": 8508.278
    },
    {
      "epoch": 1.3085779692970645,
      "grad_norm": 1.1661451467674773,
      "learning_rate": 2.6706950992050097e-06,
      "loss": 0.29066901206970214,
      "num_input_tokens_seen": 318271816,
      "step": 19435,
      "train_runtime": 37407.19,
      "train_tokens_per_second": 8508.306
    },
    {
      "epoch": 1.3089146242930245,
      "grad_norm": 1.2863310902812963,
      "learning_rate": 2.6683557879563726e-06,
      "loss": 0.3180710315704346,
      "num_input_tokens_seen": 318353736,
      "step": 19440,
      "train_runtime": 37416.6968,
      "train_tokens_per_second": 8508.334
    },
    {
      "epoch": 1.3092512792889845,
      "grad_norm": 1.087581117825262,
      "learning_rate": 2.6660171287446045e-06,
      "loss": 0.3092444658279419,
      "num_input_tokens_seen": 318435656,
      "step": 19445,
      "train_runtime": 37426.2057,
      "train_tokens_per_second": 8508.361
    },
    {
      "epoch": 1.3095879342849448,
      "grad_norm": 1.312656938112002,
      "learning_rate": 2.6636791222237034e-06,
      "loss": 0.3131858825683594,
      "num_input_tokens_seen": 318517576,
      "step": 19450,
      "train_runtime": 37435.7067,
      "train_tokens_per_second": 8508.39
    },
    {
      "epoch": 1.309924589280905,
      "grad_norm": 1.1673587201862774,
      "learning_rate": 2.6613417690474864e-06,
      "loss": 0.30432126522064207,
      "num_input_tokens_seen": 318599496,
      "step": 19455,
      "train_runtime": 37445.2698,
      "train_tokens_per_second": 8508.404
    },
    {
      "epoch": 1.310261244276865,
      "grad_norm": 1.013361761212174,
      "learning_rate": 2.6590050698695856e-06,
      "loss": 0.2666019916534424,
      "num_input_tokens_seen": 318681416,
      "step": 19460,
      "train_runtime": 37454.8056,
      "train_tokens_per_second": 8508.425
    },
    {
      "epoch": 1.310597899272825,
      "grad_norm": 1.002297524965031,
      "learning_rate": 2.6566690253434508e-06,
      "loss": 0.2967123508453369,
      "num_input_tokens_seen": 318763336,
      "step": 19465,
      "train_runtime": 37464.3159,
      "train_tokens_per_second": 8508.452
    },
    {
      "epoch": 1.3109345542687854,
      "grad_norm": 1.1430159701779106,
      "learning_rate": 2.654333636122354e-06,
      "loss": 0.28548493385314944,
      "num_input_tokens_seen": 318845256,
      "step": 19470,
      "train_runtime": 37473.8235,
      "train_tokens_per_second": 8508.479
    },
    {
      "epoch": 1.3112712092647456,
      "grad_norm": 1.2765379307150595,
      "learning_rate": 2.6519989028593728e-06,
      "loss": 0.32734503746032717,
      "num_input_tokens_seen": 318927176,
      "step": 19475,
      "train_runtime": 37483.3938,
      "train_tokens_per_second": 8508.493
    },
    {
      "epoch": 1.3116078642607056,
      "grad_norm": 1.068530292103892,
      "learning_rate": 2.649664826207414e-06,
      "loss": 0.3053431510925293,
      "num_input_tokens_seen": 319009096,
      "step": 19480,
      "train_runtime": 37492.8971,
      "train_tokens_per_second": 8508.521
    },
    {
      "epoch": 1.3119445192566657,
      "grad_norm": 1.135244841810317,
      "learning_rate": 2.647331406819189e-06,
      "loss": 0.30963711738586425,
      "num_input_tokens_seen": 319091016,
      "step": 19485,
      "train_runtime": 37502.4019,
      "train_tokens_per_second": 8508.549
    },
    {
      "epoch": 1.312281174252626,
      "grad_norm": 0.8586258828572727,
      "learning_rate": 2.644998645347236e-06,
      "loss": 0.2760464668273926,
      "num_input_tokens_seen": 319172936,
      "step": 19490,
      "train_runtime": 37511.901,
      "train_tokens_per_second": 8508.578
    },
    {
      "epoch": 1.3126178292485862,
      "grad_norm": 1.195281309831034,
      "learning_rate": 2.6426665424439025e-06,
      "loss": 0.3007453441619873,
      "num_input_tokens_seen": 319254856,
      "step": 19495,
      "train_runtime": 37521.4054,
      "train_tokens_per_second": 8508.606
    },
    {
      "epoch": 1.3129544842445462,
      "grad_norm": 1.2253157256644012,
      "learning_rate": 2.6403350987613526e-06,
      "loss": 0.2593554973602295,
      "num_input_tokens_seen": 319336776,
      "step": 19500,
      "train_runtime": 37530.9246,
      "train_tokens_per_second": 8508.631
    },
    {
      "epoch": 1.3132911392405062,
      "grad_norm": 1.2826259027394593,
      "learning_rate": 2.6380043149515676e-06,
      "loss": 0.30573737621307373,
      "num_input_tokens_seen": 319418696,
      "step": 19505,
      "train_runtime": 37540.5162,
      "train_tokens_per_second": 8508.639
    },
    {
      "epoch": 1.3136277942364665,
      "grad_norm": 1.2132877966561992,
      "learning_rate": 2.6356741916663453e-06,
      "loss": 0.2955209016799927,
      "num_input_tokens_seen": 319500616,
      "step": 19510,
      "train_runtime": 37550.0252,
      "train_tokens_per_second": 8508.666
    },
    {
      "epoch": 1.3139644492324267,
      "grad_norm": 1.0048461428926374,
      "learning_rate": 2.6333447295572933e-06,
      "loss": 0.27246334552764895,
      "num_input_tokens_seen": 319582536,
      "step": 19515,
      "train_runtime": 37559.5289,
      "train_tokens_per_second": 8508.694
    },
    {
      "epoch": 1.3143011042283868,
      "grad_norm": 1.2048713228042898,
      "learning_rate": 2.631015929275845e-06,
      "loss": 0.2901583671569824,
      "num_input_tokens_seen": 319664288,
      "step": 19520,
      "train_runtime": 37569.5485,
      "train_tokens_per_second": 8508.601
    },
    {
      "epoch": 1.3146377592243468,
      "grad_norm": 1.4522257606456332,
      "learning_rate": 2.628687791473234e-06,
      "loss": 0.29902920722961424,
      "num_input_tokens_seen": 319746208,
      "step": 19525,
      "train_runtime": 37579.0559,
      "train_tokens_per_second": 8508.628
    },
    {
      "epoch": 1.314974414220307,
      "grad_norm": 1.1580277967640913,
      "learning_rate": 2.6263603168005224e-06,
      "loss": 0.2747054100036621,
      "num_input_tokens_seen": 319828128,
      "step": 19530,
      "train_runtime": 37588.562,
      "train_tokens_per_second": 8508.656
    },
    {
      "epoch": 1.3153110692162673,
      "grad_norm": 1.1109521682561716,
      "learning_rate": 2.6240335059085797e-06,
      "loss": 0.28760039806365967,
      "num_input_tokens_seen": 319910048,
      "step": 19535,
      "train_runtime": 37598.0767,
      "train_tokens_per_second": 8508.681
    },
    {
      "epoch": 1.3156477242122273,
      "grad_norm": 1.272826927139275,
      "learning_rate": 2.62170735944809e-06,
      "loss": 0.29998106956481935,
      "num_input_tokens_seen": 319991024,
      "step": 19540,
      "train_runtime": 37607.755,
      "train_tokens_per_second": 8508.645
    },
    {
      "epoch": 1.3159843792081873,
      "grad_norm": 0.9797301928535989,
      "learning_rate": 2.6193818780695537e-06,
      "loss": 0.2613550662994385,
      "num_input_tokens_seen": 320072944,
      "step": 19545,
      "train_runtime": 37617.2635,
      "train_tokens_per_second": 8508.672
    },
    {
      "epoch": 1.3163210342041476,
      "grad_norm": 1.2617064254423964,
      "learning_rate": 2.6170570624232837e-06,
      "loss": 0.3036994695663452,
      "num_input_tokens_seen": 320154864,
      "step": 19550,
      "train_runtime": 37626.8534,
      "train_tokens_per_second": 8508.68
    },
    {
      "epoch": 1.3166576892001078,
      "grad_norm": 1.1653561992949502,
      "learning_rate": 2.6147329131594078e-06,
      "loss": 0.28125059604644775,
      "num_input_tokens_seen": 320236784,
      "step": 19555,
      "train_runtime": 37636.43,
      "train_tokens_per_second": 8508.692
    },
    {
      "epoch": 1.3169943441960679,
      "grad_norm": 1.077898324943614,
      "learning_rate": 2.612409430927867e-06,
      "loss": 0.27687869071960447,
      "num_input_tokens_seen": 320318704,
      "step": 19560,
      "train_runtime": 37645.9373,
      "train_tokens_per_second": 8508.719
    },
    {
      "epoch": 1.317330999192028,
      "grad_norm": 1.1260819428456192,
      "learning_rate": 2.610086616378413e-06,
      "loss": 0.26977882385253904,
      "num_input_tokens_seen": 320400624,
      "step": 19565,
      "train_runtime": 37655.44,
      "train_tokens_per_second": 8508.747
    },
    {
      "epoch": 1.3176676541879881,
      "grad_norm": 1.0603545520395017,
      "learning_rate": 2.60776447016062e-06,
      "loss": 0.2930527925491333,
      "num_input_tokens_seen": 320482544,
      "step": 19570,
      "train_runtime": 37664.9459,
      "train_tokens_per_second": 8508.775
    },
    {
      "epoch": 1.3180043091839484,
      "grad_norm": 1.167399647947802,
      "learning_rate": 2.6054429929238607e-06,
      "loss": 0.28945603370666506,
      "num_input_tokens_seen": 320563400,
      "step": 19575,
      "train_runtime": 37674.6137,
      "train_tokens_per_second": 8508.738
    },
    {
      "epoch": 1.3183409641799084,
      "grad_norm": 0.8974917689743485,
      "learning_rate": 2.603122185317335e-06,
      "loss": 0.2763512134552002,
      "num_input_tokens_seen": 320645320,
      "step": 19580,
      "train_runtime": 37684.1401,
      "train_tokens_per_second": 8508.76
    },
    {
      "epoch": 1.3186776191758685,
      "grad_norm": 1.0668696527803543,
      "learning_rate": 2.6008020479900465e-06,
      "loss": 0.2618814706802368,
      "num_input_tokens_seen": 320727240,
      "step": 19585,
      "train_runtime": 37693.7415,
      "train_tokens_per_second": 8508.766
    },
    {
      "epoch": 1.3190142741718287,
      "grad_norm": 1.0396793788084635,
      "learning_rate": 2.598482581590815e-06,
      "loss": 0.2890535831451416,
      "num_input_tokens_seen": 320809160,
      "step": 19590,
      "train_runtime": 37703.3911,
      "train_tokens_per_second": 8508.761
    },
    {
      "epoch": 1.319350929167789,
      "grad_norm": 1.012975640788668,
      "learning_rate": 2.5961637867682715e-06,
      "loss": 0.25083279609680176,
      "num_input_tokens_seen": 320891080,
      "step": 19595,
      "train_runtime": 37712.9886,
      "train_tokens_per_second": 8508.768
    },
    {
      "epoch": 1.319687584163749,
      "grad_norm": 0.950177088794363,
      "learning_rate": 2.5938456641708598e-06,
      "loss": 0.2853292465209961,
      "num_input_tokens_seen": 320972936,
      "step": 19600,
      "train_runtime": 37723.1246,
      "train_tokens_per_second": 8508.652
    },
    {
      "epoch": 1.320024239159709,
      "grad_norm": 1.2206325363422263,
      "learning_rate": 2.591528214446836e-06,
      "loss": 0.2692802906036377,
      "num_input_tokens_seen": 321054856,
      "step": 19605,
      "train_runtime": 37732.7026,
      "train_tokens_per_second": 8508.663
    },
    {
      "epoch": 1.3203608941556693,
      "grad_norm": 1.2427699913915413,
      "learning_rate": 2.5892114382442664e-06,
      "loss": 0.2969292163848877,
      "num_input_tokens_seen": 321136776,
      "step": 19610,
      "train_runtime": 37742.2646,
      "train_tokens_per_second": 8508.678
    },
    {
      "epoch": 1.3206975491516295,
      "grad_norm": 0.945823101904977,
      "learning_rate": 2.5868953362110287e-06,
      "loss": 0.2970273494720459,
      "num_input_tokens_seen": 321218696,
      "step": 19615,
      "train_runtime": 37751.8929,
      "train_tokens_per_second": 8508.678
    },
    {
      "epoch": 1.3210342041475895,
      "grad_norm": 0.9908704118044106,
      "learning_rate": 2.5845799089948196e-06,
      "loss": 0.27908124923706057,
      "num_input_tokens_seen": 321300096,
      "step": 19620,
      "train_runtime": 37761.8086,
      "train_tokens_per_second": 8508.599
    },
    {
      "epoch": 1.3213708591435496,
      "grad_norm": 1.2700808758516404,
      "learning_rate": 2.5822651572431323e-06,
      "loss": 0.29024577140808105,
      "num_input_tokens_seen": 321381592,
      "step": 19625,
      "train_runtime": 37771.7485,
      "train_tokens_per_second": 8508.518
    },
    {
      "epoch": 1.3217075141395098,
      "grad_norm": 1.2423806865454754,
      "learning_rate": 2.5799510816032867e-06,
      "loss": 0.30250000953674316,
      "num_input_tokens_seen": 321463512,
      "step": 19630,
      "train_runtime": 37781.3338,
      "train_tokens_per_second": 8508.527
    },
    {
      "epoch": 1.32204416913547,
      "grad_norm": 1.190513018783329,
      "learning_rate": 2.577637682722404e-06,
      "loss": 0.28919153213500975,
      "num_input_tokens_seen": 321545432,
      "step": 19635,
      "train_runtime": 37790.8773,
      "train_tokens_per_second": 8508.546
    },
    {
      "epoch": 1.32238082413143,
      "grad_norm": 1.241143477105986,
      "learning_rate": 2.5753249612474195e-06,
      "loss": 0.29706802368164065,
      "num_input_tokens_seen": 321627352,
      "step": 19640,
      "train_runtime": 37800.4944,
      "train_tokens_per_second": 8508.549
    },
    {
      "epoch": 1.3227174791273901,
      "grad_norm": 1.0362029376953807,
      "learning_rate": 2.5730129178250774e-06,
      "loss": 0.2958677291870117,
      "num_input_tokens_seen": 321709272,
      "step": 19645,
      "train_runtime": 37810.1024,
      "train_tokens_per_second": 8508.553
    },
    {
      "epoch": 1.3230541341233504,
      "grad_norm": 1.033949431665192,
      "learning_rate": 2.570701553101933e-06,
      "loss": 0.2929606199264526,
      "num_input_tokens_seen": 321791192,
      "step": 19650,
      "train_runtime": 37819.705,
      "train_tokens_per_second": 8508.559
    },
    {
      "epoch": 1.3233907891193106,
      "grad_norm": 1.2629917198407952,
      "learning_rate": 2.568390867724354e-06,
      "loss": 0.29915785789489746,
      "num_input_tokens_seen": 321873112,
      "step": 19655,
      "train_runtime": 37829.2716,
      "train_tokens_per_second": 8508.573
    },
    {
      "epoch": 1.3237274441152707,
      "grad_norm": 1.2147764895659776,
      "learning_rate": 2.5660808623385143e-06,
      "loss": 0.3067183017730713,
      "num_input_tokens_seen": 321955032,
      "step": 19660,
      "train_runtime": 37838.8498,
      "train_tokens_per_second": 8508.584
    },
    {
      "epoch": 1.3240640991112307,
      "grad_norm": 0.9686756808906278,
      "learning_rate": 2.5637715375903984e-06,
      "loss": 0.2627187490463257,
      "num_input_tokens_seen": 322036952,
      "step": 19665,
      "train_runtime": 37848.427,
      "train_tokens_per_second": 8508.595
    },
    {
      "epoch": 1.324400754107191,
      "grad_norm": 1.0754471506874657,
      "learning_rate": 2.5614628941258075e-06,
      "loss": 0.2934550762176514,
      "num_input_tokens_seen": 322118872,
      "step": 19670,
      "train_runtime": 37857.999,
      "train_tokens_per_second": 8508.608
    },
    {
      "epoch": 1.3247374091031512,
      "grad_norm": 1.2540711152308515,
      "learning_rate": 2.5591549325903388e-06,
      "loss": 0.3260610342025757,
      "num_input_tokens_seen": 322200792,
      "step": 19675,
      "train_runtime": 37867.5759,
      "train_tokens_per_second": 8508.619
    },
    {
      "epoch": 1.3250740640991112,
      "grad_norm": 1.1114306427934446,
      "learning_rate": 2.5568476536294105e-06,
      "loss": 0.2802501916885376,
      "num_input_tokens_seen": 322282712,
      "step": 19680,
      "train_runtime": 37877.2158,
      "train_tokens_per_second": 8508.617
    },
    {
      "epoch": 1.3254107190950712,
      "grad_norm": 1.1714704276575412,
      "learning_rate": 2.5545410578882466e-06,
      "loss": 0.2788105010986328,
      "num_input_tokens_seen": 322364632,
      "step": 19685,
      "train_runtime": 37886.8389,
      "train_tokens_per_second": 8508.618
    },
    {
      "epoch": 1.3257473740910315,
      "grad_norm": 1.067941697958865,
      "learning_rate": 2.5522351460118776e-06,
      "loss": 0.26045229434967043,
      "num_input_tokens_seen": 322446416,
      "step": 19690,
      "train_runtime": 37896.9734,
      "train_tokens_per_second": 8508.5
    },
    {
      "epoch": 1.3260840290869917,
      "grad_norm": 0.9800294232868729,
      "learning_rate": 2.549929918645144e-06,
      "loss": 0.30523760318756105,
      "num_input_tokens_seen": 322528336,
      "step": 19695,
      "train_runtime": 37906.5954,
      "train_tokens_per_second": 8508.502
    },
    {
      "epoch": 1.3264206840829518,
      "grad_norm": 1.0721766437905549,
      "learning_rate": 2.5476253764326963e-06,
      "loss": 0.28883678913116456,
      "num_input_tokens_seen": 322610256,
      "step": 19700,
      "train_runtime": 37916.2093,
      "train_tokens_per_second": 8508.505
    },
    {
      "epoch": 1.3267573390789118,
      "grad_norm": 1.0920290522631853,
      "learning_rate": 2.5453215200189927e-06,
      "loss": 0.2717578887939453,
      "num_input_tokens_seen": 322692176,
      "step": 19705,
      "train_runtime": 37925.7876,
      "train_tokens_per_second": 8508.516
    },
    {
      "epoch": 1.327093994074872,
      "grad_norm": 1.1985451612239704,
      "learning_rate": 2.543018350048298e-06,
      "loss": 0.30323660373687744,
      "num_input_tokens_seen": 322774096,
      "step": 19710,
      "train_runtime": 37935.3942,
      "train_tokens_per_second": 8508.521
    },
    {
      "epoch": 1.3274306490708323,
      "grad_norm": 1.0600427700741792,
      "learning_rate": 2.540715867164685e-06,
      "loss": 0.310972785949707,
      "num_input_tokens_seen": 322856016,
      "step": 19715,
      "train_runtime": 37944.9716,
      "train_tokens_per_second": 8508.532
    },
    {
      "epoch": 1.3277673040667923,
      "grad_norm": 0.9946377472913907,
      "learning_rate": 2.538414072012042e-06,
      "loss": 0.29803056716918946,
      "num_input_tokens_seen": 322937936,
      "step": 19720,
      "train_runtime": 37954.5384,
      "train_tokens_per_second": 8508.546
    },
    {
      "epoch": 1.3281039590627524,
      "grad_norm": 1.2035923818068384,
      "learning_rate": 2.53611296523405e-06,
      "loss": 0.31126227378845217,
      "num_input_tokens_seen": 323019856,
      "step": 19725,
      "train_runtime": 37964.1369,
      "train_tokens_per_second": 8508.553
    },
    {
      "epoch": 1.3284406140587126,
      "grad_norm": 1.2179306479301002,
      "learning_rate": 2.5338125474742127e-06,
      "loss": 0.298579740524292,
      "num_input_tokens_seen": 323101776,
      "step": 19730,
      "train_runtime": 37973.7415,
      "train_tokens_per_second": 8508.558
    },
    {
      "epoch": 1.3287772690546729,
      "grad_norm": 1.0850607333543756,
      "learning_rate": 2.531512819375832e-06,
      "loss": 0.2869025468826294,
      "num_input_tokens_seen": 323183696,
      "step": 19735,
      "train_runtime": 37983.3356,
      "train_tokens_per_second": 8508.565
    },
    {
      "epoch": 1.3291139240506329,
      "grad_norm": 0.959952527596008,
      "learning_rate": 2.52921378158202e-06,
      "loss": 0.25472521781921387,
      "num_input_tokens_seen": 323265232,
      "step": 19740,
      "train_runtime": 37992.9207,
      "train_tokens_per_second": 8508.565
    },
    {
      "epoch": 1.329450579046593,
      "grad_norm": 1.33202584428533,
      "learning_rate": 2.5269154347356957e-06,
      "loss": 0.28905401229858396,
      "num_input_tokens_seen": 323347152,
      "step": 19745,
      "train_runtime": 38002.5032,
      "train_tokens_per_second": 8508.575
    },
    {
      "epoch": 1.3297872340425532,
      "grad_norm": 1.457146248202956,
      "learning_rate": 2.5246177794795834e-06,
      "loss": 0.29413661956787107,
      "num_input_tokens_seen": 323429072,
      "step": 19750,
      "train_runtime": 38012.0918,
      "train_tokens_per_second": 8508.584
    },
    {
      "epoch": 1.3301238890385134,
      "grad_norm": 1.0607381711098176,
      "learning_rate": 2.5223208164562163e-06,
      "loss": 0.3201037645339966,
      "num_input_tokens_seen": 323510344,
      "step": 19755,
      "train_runtime": 38021.8616,
      "train_tokens_per_second": 8508.535
    },
    {
      "epoch": 1.3304605440344734,
      "grad_norm": 1.114434252671544,
      "learning_rate": 2.5200245463079315e-06,
      "loss": 0.31701269149780276,
      "num_input_tokens_seen": 323592264,
      "step": 19760,
      "train_runtime": 38031.4397,
      "train_tokens_per_second": 8508.546
    },
    {
      "epoch": 1.3307971990304335,
      "grad_norm": 1.0287311849303544,
      "learning_rate": 2.5177289696768725e-06,
      "loss": 0.2961724281311035,
      "num_input_tokens_seen": 323674184,
      "step": 19765,
      "train_runtime": 38041.0276,
      "train_tokens_per_second": 8508.555
    },
    {
      "epoch": 1.3311338540263937,
      "grad_norm": 1.0747950671809734,
      "learning_rate": 2.515434087204996e-06,
      "loss": 0.2812824010848999,
      "num_input_tokens_seen": 323756104,
      "step": 19770,
      "train_runtime": 38050.6444,
      "train_tokens_per_second": 8508.558
    },
    {
      "epoch": 1.331470509022354,
      "grad_norm": 1.0867473118172157,
      "learning_rate": 2.51313989953405e-06,
      "loss": 0.3200028419494629,
      "num_input_tokens_seen": 323838024,
      "step": 19775,
      "train_runtime": 38060.2691,
      "train_tokens_per_second": 8508.558
    },
    {
      "epoch": 1.331807164018314,
      "grad_norm": 1.1071341612621344,
      "learning_rate": 2.510846407305603e-06,
      "loss": 0.2731557130813599,
      "num_input_tokens_seen": 323919944,
      "step": 19780,
      "train_runtime": 38069.8799,
      "train_tokens_per_second": 8508.562
    },
    {
      "epoch": 1.332143819014274,
      "grad_norm": 1.139051713810028,
      "learning_rate": 2.508553611161021e-06,
      "loss": 0.30778183937072756,
      "num_input_tokens_seen": 324001864,
      "step": 19785,
      "train_runtime": 38079.444,
      "train_tokens_per_second": 8508.577
    },
    {
      "epoch": 1.3324804740102343,
      "grad_norm": 0.9882130961960145,
      "learning_rate": 2.5062615117414773e-06,
      "loss": 0.326526403427124,
      "num_input_tokens_seen": 324083784,
      "step": 19790,
      "train_runtime": 38089.0318,
      "train_tokens_per_second": 8508.586
    },
    {
      "epoch": 1.3328171290061945,
      "grad_norm": 0.9292335381782956,
      "learning_rate": 2.50397010968795e-06,
      "loss": 0.27596614360809324,
      "num_input_tokens_seen": 324165704,
      "step": 19795,
      "train_runtime": 38098.6182,
      "train_tokens_per_second": 8508.595
    },
    {
      "epoch": 1.3331537840021546,
      "grad_norm": 1.2700830778784418,
      "learning_rate": 2.501679405641222e-06,
      "loss": 0.31825311183929444,
      "num_input_tokens_seen": 324247624,
      "step": 19800,
      "train_runtime": 38108.2192,
      "train_tokens_per_second": 8508.601
    },
    {
      "epoch": 1.3334904389981148,
      "grad_norm": 1.100605796498167,
      "learning_rate": 2.499389400241881e-06,
      "loss": 0.2931965351104736,
      "num_input_tokens_seen": 324329544,
      "step": 19805,
      "train_runtime": 38117.8,
      "train_tokens_per_second": 8508.611
    },
    {
      "epoch": 1.3338270939940748,
      "grad_norm": 1.074191596956675,
      "learning_rate": 2.497100094130321e-06,
      "loss": 0.2750261306762695,
      "num_input_tokens_seen": 324411464,
      "step": 19810,
      "train_runtime": 38127.4136,
      "train_tokens_per_second": 8508.614
    },
    {
      "epoch": 1.334163748990035,
      "grad_norm": 1.1406514053271188,
      "learning_rate": 2.494811487946737e-06,
      "loss": 0.27162761688232423,
      "num_input_tokens_seen": 324493384,
      "step": 19815,
      "train_runtime": 38137.014,
      "train_tokens_per_second": 8508.621
    },
    {
      "epoch": 1.3345004039859951,
      "grad_norm": 1.1188153516993695,
      "learning_rate": 2.492523582331135e-06,
      "loss": 0.3120009422302246,
      "num_input_tokens_seen": 324575304,
      "step": 19820,
      "train_runtime": 38146.6206,
      "train_tokens_per_second": 8508.625
    },
    {
      "epoch": 1.3348370589819554,
      "grad_norm": 1.0335860427814665,
      "learning_rate": 2.4902363779233134e-06,
      "loss": 0.27572193145751955,
      "num_input_tokens_seen": 324657224,
      "step": 19825,
      "train_runtime": 38156.2273,
      "train_tokens_per_second": 8508.63
    },
    {
      "epoch": 1.3351737139779154,
      "grad_norm": 1.060698261469215,
      "learning_rate": 2.4879498753628885e-06,
      "loss": 0.28606462478637695,
      "num_input_tokens_seen": 324739144,
      "step": 19830,
      "train_runtime": 38165.8326,
      "train_tokens_per_second": 8508.635
    },
    {
      "epoch": 1.3355103689738757,
      "grad_norm": 0.9452516140328966,
      "learning_rate": 2.4856640752892702e-06,
      "loss": 0.26226470470428465,
      "num_input_tokens_seen": 324821064,
      "step": 19835,
      "train_runtime": 38175.4068,
      "train_tokens_per_second": 8508.647
    },
    {
      "epoch": 1.3358470239698357,
      "grad_norm": 1.15741097978375,
      "learning_rate": 2.4833789783416757e-06,
      "loss": 0.2838031768798828,
      "num_input_tokens_seen": 324902984,
      "step": 19840,
      "train_runtime": 38185.0137,
      "train_tokens_per_second": 8508.652
    },
    {
      "epoch": 1.336183678965796,
      "grad_norm": 1.2032407207320412,
      "learning_rate": 2.4810945851591245e-06,
      "loss": 0.2916494607925415,
      "num_input_tokens_seen": 324984904,
      "step": 19845,
      "train_runtime": 38194.5986,
      "train_tokens_per_second": 8508.661
    },
    {
      "epoch": 1.336520333961756,
      "grad_norm": 1.076718852557263,
      "learning_rate": 2.4788108963804406e-06,
      "loss": 0.2840999126434326,
      "num_input_tokens_seen": 325066824,
      "step": 19850,
      "train_runtime": 38204.1688,
      "train_tokens_per_second": 8508.674
    },
    {
      "epoch": 1.3368569889577162,
      "grad_norm": 1.0702820187444744,
      "learning_rate": 2.4765279126442502e-06,
      "loss": 0.2793304920196533,
      "num_input_tokens_seen": 325148744,
      "step": 19855,
      "train_runtime": 38213.7587,
      "train_tokens_per_second": 8508.683
    },
    {
      "epoch": 1.3371936439536762,
      "grad_norm": 1.0339743834210828,
      "learning_rate": 2.4742456345889828e-06,
      "loss": 0.3019388675689697,
      "num_input_tokens_seen": 325230664,
      "step": 19860,
      "train_runtime": 38223.3468,
      "train_tokens_per_second": 8508.691
    },
    {
      "epoch": 1.3375302989496365,
      "grad_norm": 1.0896359669003268,
      "learning_rate": 2.471964062852867e-06,
      "loss": 0.2672716617584229,
      "num_input_tokens_seen": 325312584,
      "step": 19865,
      "train_runtime": 38232.9499,
      "train_tokens_per_second": 8508.697
    },
    {
      "epoch": 1.3378669539455965,
      "grad_norm": 0.9654133377613203,
      "learning_rate": 2.469683198073943e-06,
      "loss": 0.26177539825439455,
      "num_input_tokens_seen": 325394504,
      "step": 19870,
      "train_runtime": 38242.5529,
      "train_tokens_per_second": 8508.702
    },
    {
      "epoch": 1.3382036089415568,
      "grad_norm": 1.3140360751610844,
      "learning_rate": 2.4674030408900414e-06,
      "loss": 0.2915104866027832,
      "num_input_tokens_seen": 325476424,
      "step": 19875,
      "train_runtime": 38252.1059,
      "train_tokens_per_second": 8508.719
    },
    {
      "epoch": 1.3385402639375168,
      "grad_norm": 1.3830143516066686,
      "learning_rate": 2.465123591938805e-06,
      "loss": 0.2907625436782837,
      "num_input_tokens_seen": 325558344,
      "step": 19880,
      "train_runtime": 38261.6527,
      "train_tokens_per_second": 8508.737
    },
    {
      "epoch": 1.338876918933477,
      "grad_norm": 0.8981974615895131,
      "learning_rate": 2.462844851857673e-06,
      "loss": 0.26992318630218504,
      "num_input_tokens_seen": 325640264,
      "step": 19885,
      "train_runtime": 38271.2339,
      "train_tokens_per_second": 8508.747
    },
    {
      "epoch": 1.339213573929437,
      "grad_norm": 1.3806341608079404,
      "learning_rate": 2.4605668212838878e-06,
      "loss": 0.31442899703979493,
      "num_input_tokens_seen": 325722184,
      "step": 19890,
      "train_runtime": 38280.8178,
      "train_tokens_per_second": 8508.757
    },
    {
      "epoch": 1.3395502289253973,
      "grad_norm": 1.236392458709952,
      "learning_rate": 2.458289500854493e-06,
      "loss": 0.31676135063171384,
      "num_input_tokens_seen": 325804104,
      "step": 19895,
      "train_runtime": 38290.4414,
      "train_tokens_per_second": 8508.758
    },
    {
      "epoch": 1.3398868839213574,
      "grad_norm": 1.0018403605985713,
      "learning_rate": 2.4560128912063353e-06,
      "loss": 0.2926662445068359,
      "num_input_tokens_seen": 325886024,
      "step": 19900,
      "train_runtime": 38300.0598,
      "train_tokens_per_second": 8508.76
    },
    {
      "epoch": 1.3402235389173176,
      "grad_norm": 1.3785035139563433,
      "learning_rate": 2.4537369929760596e-06,
      "loss": 0.3222623825073242,
      "num_input_tokens_seen": 325967944,
      "step": 19905,
      "train_runtime": 38309.6532,
      "train_tokens_per_second": 8508.768
    },
    {
      "epoch": 1.3405601939132776,
      "grad_norm": 1.0561819773422751,
      "learning_rate": 2.4514618068001144e-06,
      "loss": 0.2692115068435669,
      "num_input_tokens_seen": 326049864,
      "step": 19910,
      "train_runtime": 38319.2195,
      "train_tokens_per_second": 8508.781
    },
    {
      "epoch": 1.3408968489092379,
      "grad_norm": 1.3377656196307073,
      "learning_rate": 2.4491873333147463e-06,
      "loss": 0.3076594352722168,
      "num_input_tokens_seen": 326131784,
      "step": 19915,
      "train_runtime": 38328.7819,
      "train_tokens_per_second": 8508.796
    },
    {
      "epoch": 1.341233503905198,
      "grad_norm": 1.0305263011137913,
      "learning_rate": 2.4469135731560106e-06,
      "loss": 0.2601524591445923,
      "num_input_tokens_seen": 326213704,
      "step": 19920,
      "train_runtime": 38338.4179,
      "train_tokens_per_second": 8508.794
    },
    {
      "epoch": 1.3415701589011582,
      "grad_norm": 1.2379652743368819,
      "learning_rate": 2.4446405269597494e-06,
      "loss": 0.3072859764099121,
      "num_input_tokens_seen": 326295624,
      "step": 19925,
      "train_runtime": 38347.9326,
      "train_tokens_per_second": 8508.819
    },
    {
      "epoch": 1.3419068138971182,
      "grad_norm": 0.9779500833544003,
      "learning_rate": 2.442368195361618e-06,
      "loss": 0.23801469802856445,
      "num_input_tokens_seen": 326377544,
      "step": 19930,
      "train_runtime": 38362.6371,
      "train_tokens_per_second": 8507.693
    },
    {
      "epoch": 1.3422434688930784,
      "grad_norm": 1.3392321658974773,
      "learning_rate": 2.4400965789970647e-06,
      "loss": 0.27005414962768554,
      "num_input_tokens_seen": 326459464,
      "step": 19935,
      "train_runtime": 38375.646,
      "train_tokens_per_second": 8506.944
    },
    {
      "epoch": 1.3425801238890385,
      "grad_norm": 1.152539037423255,
      "learning_rate": 2.437825678501339e-06,
      "loss": 0.2948033332824707,
      "num_input_tokens_seen": 326541384,
      "step": 19940,
      "train_runtime": 38385.4793,
      "train_tokens_per_second": 8506.899
    },
    {
      "epoch": 1.3429167788849987,
      "grad_norm": 1.205223466614004,
      "learning_rate": 2.4355554945094956e-06,
      "loss": 0.2931781768798828,
      "num_input_tokens_seen": 326623304,
      "step": 19945,
      "train_runtime": 38395.0027,
      "train_tokens_per_second": 8506.922
    },
    {
      "epoch": 1.3432534338809587,
      "grad_norm": 0.9508206884543404,
      "learning_rate": 2.4332860276563763e-06,
      "loss": 0.2902604341506958,
      "num_input_tokens_seen": 326705224,
      "step": 19950,
      "train_runtime": 38404.5077,
      "train_tokens_per_second": 8506.95
    },
    {
      "epoch": 1.343590088876919,
      "grad_norm": 0.80675693913323,
      "learning_rate": 2.4310172785766396e-06,
      "loss": 0.27834396362304686,
      "num_input_tokens_seen": 326787144,
      "step": 19955,
      "train_runtime": 38414.015,
      "train_tokens_per_second": 8506.977
    },
    {
      "epoch": 1.343926743872879,
      "grad_norm": 1.3374739940883749,
      "learning_rate": 2.428749247904724e-06,
      "loss": 0.33186917304992675,
      "num_input_tokens_seen": 326869064,
      "step": 19960,
      "train_runtime": 38423.528,
      "train_tokens_per_second": 8507.003
    },
    {
      "epoch": 1.3442633988688393,
      "grad_norm": 1.3926313254931002,
      "learning_rate": 2.4264819362748856e-06,
      "loss": 0.29869627952575684,
      "num_input_tokens_seen": 326950984,
      "step": 19965,
      "train_runtime": 38433.0449,
      "train_tokens_per_second": 8507.028
    },
    {
      "epoch": 1.3446000538647993,
      "grad_norm": 0.9640835889188475,
      "learning_rate": 2.4242153443211663e-06,
      "loss": 0.3085618257522583,
      "num_input_tokens_seen": 327032904,
      "step": 19970,
      "train_runtime": 38442.708,
      "train_tokens_per_second": 8507.02
    },
    {
      "epoch": 1.3449367088607596,
      "grad_norm": 0.9395346219429236,
      "learning_rate": 2.4219494726774124e-06,
      "loss": 0.24863572120666505,
      "num_input_tokens_seen": 327114824,
      "step": 19975,
      "train_runtime": 38452.3144,
      "train_tokens_per_second": 8507.026
    },
    {
      "epoch": 1.3452733638567196,
      "grad_norm": 1.031530413201419,
      "learning_rate": 2.419684321977268e-06,
      "loss": 0.27347278594970703,
      "num_input_tokens_seen": 327196744,
      "step": 19980,
      "train_runtime": 38462.1115,
      "train_tokens_per_second": 8506.989
    },
    {
      "epoch": 1.3456100188526798,
      "grad_norm": 1.2501149035174968,
      "learning_rate": 2.417419892854175e-06,
      "loss": 0.31364808082580564,
      "num_input_tokens_seen": 327278664,
      "step": 19985,
      "train_runtime": 38471.6295,
      "train_tokens_per_second": 8507.013
    },
    {
      "epoch": 1.3459466738486399,
      "grad_norm": 1.1325483554281577,
      "learning_rate": 2.415156185941371e-06,
      "loss": 0.2798783779144287,
      "num_input_tokens_seen": 327360584,
      "step": 19990,
      "train_runtime": 38481.1516,
      "train_tokens_per_second": 8507.037
    },
    {
      "epoch": 1.3462833288446001,
      "grad_norm": 1.1731420305413829,
      "learning_rate": 2.412893201871902e-06,
      "loss": 0.28121912479400635,
      "num_input_tokens_seen": 327442504,
      "step": 19995,
      "train_runtime": 38490.6705,
      "train_tokens_per_second": 8507.062
    },
    {
      "epoch": 1.3466199838405601,
      "grad_norm": 1.1001250083316714,
      "learning_rate": 2.410630941278595e-06,
      "loss": 0.29634456634521483,
      "num_input_tokens_seen": 327524424,
      "step": 20000,
      "train_runtime": 38500.1794,
      "train_tokens_per_second": 8507.088
    },
    {
      "epoch": 1.3469566388365204,
      "grad_norm": 1.1725725795351156,
      "learning_rate": 2.408369404794092e-06,
      "loss": 0.2876330614089966,
      "num_input_tokens_seen": 327606344,
      "step": 20005,
      "train_runtime": 38509.7009,
      "train_tokens_per_second": 8507.112
    },
    {
      "epoch": 1.3472932938324804,
      "grad_norm": 1.096973612655097,
      "learning_rate": 2.4061085930508176e-06,
      "loss": 0.31005043983459474,
      "num_input_tokens_seen": 327688264,
      "step": 20010,
      "train_runtime": 38519.2063,
      "train_tokens_per_second": 8507.14
    },
    {
      "epoch": 1.3476299488284407,
      "grad_norm": 0.9933805753964283,
      "learning_rate": 2.4038485066810058e-06,
      "loss": 0.2744114637374878,
      "num_input_tokens_seen": 327770184,
      "step": 20015,
      "train_runtime": 38528.7221,
      "train_tokens_per_second": 8507.165
    },
    {
      "epoch": 1.3479666038244007,
      "grad_norm": 1.1520199802715085,
      "learning_rate": 2.40158914631668e-06,
      "loss": 0.27313957214355467,
      "num_input_tokens_seen": 327851624,
      "step": 20020,
      "train_runtime": 38538.5498,
      "train_tokens_per_second": 8507.108
    },
    {
      "epoch": 1.348303258820361,
      "grad_norm": 0.916038892093924,
      "learning_rate": 2.399330512589664e-06,
      "loss": 0.2725252866744995,
      "num_input_tokens_seen": 327933544,
      "step": 20025,
      "train_runtime": 38548.059,
      "train_tokens_per_second": 8507.135
    },
    {
      "epoch": 1.348639913816321,
      "grad_norm": 1.0251684875743707,
      "learning_rate": 2.3970726061315775e-06,
      "loss": 0.29237291812896726,
      "num_input_tokens_seen": 328015296,
      "step": 20030,
      "train_runtime": 38558.1314,
      "train_tokens_per_second": 8507.033
    },
    {
      "epoch": 1.3489765688122812,
      "grad_norm": 0.9953390627229036,
      "learning_rate": 2.394815427573836e-06,
      "loss": 0.2649536609649658,
      "num_input_tokens_seen": 328097216,
      "step": 20035,
      "train_runtime": 38567.7391,
      "train_tokens_per_second": 8507.038
    },
    {
      "epoch": 1.3493132238082413,
      "grad_norm": 1.0236812798180701,
      "learning_rate": 2.3925589775476516e-06,
      "loss": 0.26931424140930177,
      "num_input_tokens_seen": 328179136,
      "step": 20040,
      "train_runtime": 38577.3111,
      "train_tokens_per_second": 8507.051
    },
    {
      "epoch": 1.3496498788042015,
      "grad_norm": 1.0453857185380402,
      "learning_rate": 2.3903032566840377e-06,
      "loss": 0.2750478029251099,
      "num_input_tokens_seen": 328261056,
      "step": 20045,
      "train_runtime": 38586.9023,
      "train_tokens_per_second": 8507.059
    },
    {
      "epoch": 1.3499865338001615,
      "grad_norm": 1.1282896722870186,
      "learning_rate": 2.3880482656137926e-06,
      "loss": 0.3120352029800415,
      "num_input_tokens_seen": 328342976,
      "step": 20050,
      "train_runtime": 38596.4701,
      "train_tokens_per_second": 8507.073
    },
    {
      "epoch": 1.3503231887961218,
      "grad_norm": 1.0028924875996132,
      "learning_rate": 2.3857940049675254e-06,
      "loss": 0.27841644287109374,
      "num_input_tokens_seen": 328424896,
      "step": 20055,
      "train_runtime": 38606.0598,
      "train_tokens_per_second": 8507.081
    },
    {
      "epoch": 1.3506598437920818,
      "grad_norm": 0.9060657074054138,
      "learning_rate": 2.383540475375625e-06,
      "loss": 0.2584806203842163,
      "num_input_tokens_seen": 328506816,
      "step": 20060,
      "train_runtime": 38615.6609,
      "train_tokens_per_second": 8507.088
    },
    {
      "epoch": 1.350996498788042,
      "grad_norm": 1.291377506301538,
      "learning_rate": 2.3812876774682886e-06,
      "loss": 0.2812903165817261,
      "num_input_tokens_seen": 328588736,
      "step": 20065,
      "train_runtime": 38625.283,
      "train_tokens_per_second": 8507.089
    },
    {
      "epoch": 1.351333153784002,
      "grad_norm": 1.171964560790837,
      "learning_rate": 2.379035611875503e-06,
      "loss": 0.28804779052734375,
      "num_input_tokens_seen": 328670656,
      "step": 20070,
      "train_runtime": 38634.8188,
      "train_tokens_per_second": 8507.11
    },
    {
      "epoch": 1.3516698087799623,
      "grad_norm": 1.0639119504589916,
      "learning_rate": 2.376784279227051e-06,
      "loss": 0.29665288925170896,
      "num_input_tokens_seen": 328752576,
      "step": 20075,
      "train_runtime": 38644.3412,
      "train_tokens_per_second": 8507.134
    },
    {
      "epoch": 1.3520064637759224,
      "grad_norm": 1.076819606884958,
      "learning_rate": 2.3745336801525105e-06,
      "loss": 0.3064706802368164,
      "num_input_tokens_seen": 328834496,
      "step": 20080,
      "train_runtime": 38653.9123,
      "train_tokens_per_second": 8507.147
    },
    {
      "epoch": 1.3523431187718826,
      "grad_norm": 1.092085855708561,
      "learning_rate": 2.3722838152812543e-06,
      "loss": 0.29843754768371583,
      "num_input_tokens_seen": 328916416,
      "step": 20085,
      "train_runtime": 38663.511,
      "train_tokens_per_second": 8507.153
    },
    {
      "epoch": 1.3526797737678427,
      "grad_norm": 1.036858403553154,
      "learning_rate": 2.3700346852424485e-06,
      "loss": 0.2968432903289795,
      "num_input_tokens_seen": 328998336,
      "step": 20090,
      "train_runtime": 38673.0239,
      "train_tokens_per_second": 8507.179
    },
    {
      "epoch": 1.353016428763803,
      "grad_norm": 1.120029304970408,
      "learning_rate": 2.3677862906650594e-06,
      "loss": 0.3156294345855713,
      "num_input_tokens_seen": 329080256,
      "step": 20095,
      "train_runtime": 38682.5233,
      "train_tokens_per_second": 8507.208
    },
    {
      "epoch": 1.353353083759763,
      "grad_norm": 1.1810447461592943,
      "learning_rate": 2.365538632177837e-06,
      "loss": 0.2945725440979004,
      "num_input_tokens_seen": 329162176,
      "step": 20100,
      "train_runtime": 38697.4209,
      "train_tokens_per_second": 8506.049
    },
    {
      "epoch": 1.3536897387557232,
      "grad_norm": 1.1397968955163793,
      "learning_rate": 2.36329171040934e-06,
      "loss": 0.2696204900741577,
      "num_input_tokens_seen": 329244096,
      "step": 20105,
      "train_runtime": 38718.8566,
      "train_tokens_per_second": 8503.456
    },
    {
      "epoch": 1.3540263937516832,
      "grad_norm": 1.074944223282233,
      "learning_rate": 2.3610455259879045e-06,
      "loss": 0.2592494964599609,
      "num_input_tokens_seen": 329326016,
      "step": 20110,
      "train_runtime": 38728.491,
      "train_tokens_per_second": 8503.456
    },
    {
      "epoch": 1.3543630487476435,
      "grad_norm": 1.310297955073439,
      "learning_rate": 2.358800079541674e-06,
      "loss": 0.29919564723968506,
      "num_input_tokens_seen": 329407936,
      "step": 20115,
      "train_runtime": 38738.0084,
      "train_tokens_per_second": 8503.481
    },
    {
      "epoch": 1.3546997037436035,
      "grad_norm": 1.0778211717302844,
      "learning_rate": 2.3565553716985794e-06,
      "loss": 0.31352849006652833,
      "num_input_tokens_seen": 329489856,
      "step": 20120,
      "train_runtime": 38747.5179,
      "train_tokens_per_second": 8503.509
    },
    {
      "epoch": 1.3550363587395637,
      "grad_norm": 1.0561371078795858,
      "learning_rate": 2.3543114030863457e-06,
      "loss": 0.260768723487854,
      "num_input_tokens_seen": 329571776,
      "step": 20125,
      "train_runtime": 38757.0324,
      "train_tokens_per_second": 8503.535
    },
    {
      "epoch": 1.3553730137355238,
      "grad_norm": 1.196773653435015,
      "learning_rate": 2.352068174332491e-06,
      "loss": 0.28005821704864503,
      "num_input_tokens_seen": 329653696,
      "step": 20130,
      "train_runtime": 38766.688,
      "train_tokens_per_second": 8503.53
    },
    {
      "epoch": 1.355709668731484,
      "grad_norm": 1.0915890685586747,
      "learning_rate": 2.349825686064329e-06,
      "loss": 0.262341570854187,
      "num_input_tokens_seen": 329735616,
      "step": 20135,
      "train_runtime": 38776.1928,
      "train_tokens_per_second": 8503.558
    },
    {
      "epoch": 1.356046323727444,
      "grad_norm": 1.1836947900131214,
      "learning_rate": 2.3475839389089605e-06,
      "loss": 0.3065411806106567,
      "num_input_tokens_seen": 329817536,
      "step": 20140,
      "train_runtime": 38785.8119,
      "train_tokens_per_second": 8503.561
    },
    {
      "epoch": 1.3563829787234043,
      "grad_norm": 1.1872694503652947,
      "learning_rate": 2.345342933493289e-06,
      "loss": 0.27569217681884767,
      "num_input_tokens_seen": 329899456,
      "step": 20145,
      "train_runtime": 38795.433,
      "train_tokens_per_second": 8503.564
    },
    {
      "epoch": 1.3567196337193643,
      "grad_norm": 1.079917231920891,
      "learning_rate": 2.343102670443998e-06,
      "loss": 0.29011390209197996,
      "num_input_tokens_seen": 329981376,
      "step": 20150,
      "train_runtime": 38805.124,
      "train_tokens_per_second": 8503.552
    },
    {
      "epoch": 1.3570562887153246,
      "grad_norm": 1.252450181109306,
      "learning_rate": 2.3408631503875763e-06,
      "loss": 0.31160783767700195,
      "num_input_tokens_seen": 330063296,
      "step": 20155,
      "train_runtime": 38814.8008,
      "train_tokens_per_second": 8503.542
    },
    {
      "epoch": 1.3573929437112846,
      "grad_norm": 1.1647687745606283,
      "learning_rate": 2.338624373950291e-06,
      "loss": 0.24740099906921387,
      "num_input_tokens_seen": 330145216,
      "step": 20160,
      "train_runtime": 38824.471,
      "train_tokens_per_second": 8503.534
    },
    {
      "epoch": 1.3577295987072449,
      "grad_norm": 1.1743901713855187,
      "learning_rate": 2.336386341758216e-06,
      "loss": 0.2837717533111572,
      "num_input_tokens_seen": 330227136,
      "step": 20165,
      "train_runtime": 38834.1176,
      "train_tokens_per_second": 8503.531
    },
    {
      "epoch": 1.3580662537032049,
      "grad_norm": 1.0456221808966666,
      "learning_rate": 2.3341490544372063e-06,
      "loss": 0.28386507034301756,
      "num_input_tokens_seen": 330309056,
      "step": 20170,
      "train_runtime": 38843.7976,
      "train_tokens_per_second": 8503.521
    },
    {
      "epoch": 1.3584029086991651,
      "grad_norm": 1.2713824902490265,
      "learning_rate": 2.3319125126129128e-06,
      "loss": 0.28914268016815187,
      "num_input_tokens_seen": 330390976,
      "step": 20175,
      "train_runtime": 38853.4821,
      "train_tokens_per_second": 8503.51
    },
    {
      "epoch": 1.3587395636951252,
      "grad_norm": 1.1268387363345755,
      "learning_rate": 2.329676716910777e-06,
      "loss": 0.313228964805603,
      "num_input_tokens_seen": 330472896,
      "step": 20180,
      "train_runtime": 38863.0599,
      "train_tokens_per_second": 8503.522
    },
    {
      "epoch": 1.3590762186910854,
      "grad_norm": 0.9922074730584804,
      "learning_rate": 2.3274416679560325e-06,
      "loss": 0.28026437759399414,
      "num_input_tokens_seen": 330554816,
      "step": 20185,
      "train_runtime": 38872.5686,
      "train_tokens_per_second": 8503.55
    },
    {
      "epoch": 1.3594128736870454,
      "grad_norm": 1.0109129562899035,
      "learning_rate": 2.3252073663737016e-06,
      "loss": 0.3010397911071777,
      "num_input_tokens_seen": 330636736,
      "step": 20190,
      "train_runtime": 38882.1897,
      "train_tokens_per_second": 8503.552
    },
    {
      "epoch": 1.3597495286830057,
      "grad_norm": 3.096005267145079,
      "learning_rate": 2.3229738127886054e-06,
      "loss": 0.2961612939834595,
      "num_input_tokens_seen": 330718656,
      "step": 20195,
      "train_runtime": 38891.7909,
      "train_tokens_per_second": 8503.559
    },
    {
      "epoch": 1.3600861836789657,
      "grad_norm": 1.0700395025610527,
      "learning_rate": 2.3207410078253416e-06,
      "loss": 0.30752248764038087,
      "num_input_tokens_seen": 330800576,
      "step": 20200,
      "train_runtime": 38901.6567,
      "train_tokens_per_second": 8503.509
    },
    {
      "epoch": 1.360422838674926,
      "grad_norm": 1.1047015848709822,
      "learning_rate": 2.3185089521083154e-06,
      "loss": 0.2994969367980957,
      "num_input_tokens_seen": 330882496,
      "step": 20205,
      "train_runtime": 38911.3294,
      "train_tokens_per_second": 8503.5
    },
    {
      "epoch": 1.360759493670886,
      "grad_norm": 0.9982638140352244,
      "learning_rate": 2.3162776462617067e-06,
      "loss": 0.2759389400482178,
      "num_input_tokens_seen": 330964416,
      "step": 20210,
      "train_runtime": 38920.8708,
      "train_tokens_per_second": 8503.52
    },
    {
      "epoch": 1.3610961486668463,
      "grad_norm": 1.261649962653299,
      "learning_rate": 2.314047090909498e-06,
      "loss": 0.28277628421783446,
      "num_input_tokens_seen": 331046336,
      "step": 20215,
      "train_runtime": 38930.9082,
      "train_tokens_per_second": 8503.432
    },
    {
      "epoch": 1.3614328036628063,
      "grad_norm": 1.2095216421718202,
      "learning_rate": 2.3118172866754557e-06,
      "loss": 0.28890290260314944,
      "num_input_tokens_seen": 331128256,
      "step": 20220,
      "train_runtime": 38940.5119,
      "train_tokens_per_second": 8503.439
    },
    {
      "epoch": 1.3617694586587665,
      "grad_norm": 1.0800674687272585,
      "learning_rate": 2.309588234183137e-06,
      "loss": 0.2877105474472046,
      "num_input_tokens_seen": 331210176,
      "step": 20225,
      "train_runtime": 38950.1105,
      "train_tokens_per_second": 8503.446
    },
    {
      "epoch": 1.3621061136547266,
      "grad_norm": 1.0376758799043524,
      "learning_rate": 2.3073599340558896e-06,
      "loss": 0.28927059173583985,
      "num_input_tokens_seen": 331291792,
      "step": 20230,
      "train_runtime": 38960.0894,
      "train_tokens_per_second": 8503.363
    },
    {
      "epoch": 1.3624427686506868,
      "grad_norm": 1.3520265271479872,
      "learning_rate": 2.3051323869168506e-06,
      "loss": 0.32587695121765137,
      "num_input_tokens_seen": 331373712,
      "step": 20235,
      "train_runtime": 38969.6607,
      "train_tokens_per_second": 8503.377
    },
    {
      "epoch": 1.3627794236466468,
      "grad_norm": 0.9339949492161149,
      "learning_rate": 2.3029055933889443e-06,
      "loss": 0.28050918579101564,
      "num_input_tokens_seen": 331455632,
      "step": 20240,
      "train_runtime": 38979.2291,
      "train_tokens_per_second": 8503.391
    },
    {
      "epoch": 1.363116078642607,
      "grad_norm": 0.9175458244709642,
      "learning_rate": 2.300679554094892e-06,
      "loss": 0.2609947919845581,
      "num_input_tokens_seen": 331537552,
      "step": 20245,
      "train_runtime": 38988.7485,
      "train_tokens_per_second": 8503.416
    },
    {
      "epoch": 1.3634527336385671,
      "grad_norm": 1.0626514797486546,
      "learning_rate": 2.298454269657191e-06,
      "loss": 0.285076642036438,
      "num_input_tokens_seen": 331619472,
      "step": 20250,
      "train_runtime": 38998.3586,
      "train_tokens_per_second": 8503.421
    },
    {
      "epoch": 1.3637893886345274,
      "grad_norm": 0.9511557150414881,
      "learning_rate": 2.2962297406981432e-06,
      "loss": 0.2643630027770996,
      "num_input_tokens_seen": 331701392,
      "step": 20255,
      "train_runtime": 39007.94,
      "train_tokens_per_second": 8503.433
    },
    {
      "epoch": 1.3641260436304874,
      "grad_norm": 1.0870697671566518,
      "learning_rate": 2.2940059678398223e-06,
      "loss": 0.26644349098205566,
      "num_input_tokens_seen": 331783312,
      "step": 20260,
      "train_runtime": 39017.557,
      "train_tokens_per_second": 8503.436
    },
    {
      "epoch": 1.3644626986264476,
      "grad_norm": 1.0367850890565442,
      "learning_rate": 2.291782951704106e-06,
      "loss": 0.2782252073287964,
      "num_input_tokens_seen": 331865232,
      "step": 20265,
      "train_runtime": 39027.3666,
      "train_tokens_per_second": 8503.398
    },
    {
      "epoch": 1.3647993536224077,
      "grad_norm": 1.0422049600609444,
      "learning_rate": 2.289560692912651e-06,
      "loss": 0.2870193958282471,
      "num_input_tokens_seen": 331947152,
      "step": 20270,
      "train_runtime": 39037.1864,
      "train_tokens_per_second": 8503.358
    },
    {
      "epoch": 1.365136008618368,
      "grad_norm": 1.0321208795641441,
      "learning_rate": 2.287339192086906e-06,
      "loss": 0.29428997039794924,
      "num_input_tokens_seen": 332029072,
      "step": 20275,
      "train_runtime": 39047.0314,
      "train_tokens_per_second": 8503.312
    },
    {
      "epoch": 1.365472663614328,
      "grad_norm": 1.0148428403877536,
      "learning_rate": 2.2851184498481056e-06,
      "loss": 0.2796201467514038,
      "num_input_tokens_seen": 332110992,
      "step": 20280,
      "train_runtime": 39056.8045,
      "train_tokens_per_second": 8503.281
    },
    {
      "epoch": 1.3658093186102882,
      "grad_norm": 1.1782058152173354,
      "learning_rate": 2.2828984668172732e-06,
      "loss": 0.2733154773712158,
      "num_input_tokens_seen": 332192912,
      "step": 20285,
      "train_runtime": 39066.4417,
      "train_tokens_per_second": 8503.28
    },
    {
      "epoch": 1.3661459736062485,
      "grad_norm": 1.0761898841856345,
      "learning_rate": 2.2806792436152185e-06,
      "loss": 0.29717919826507566,
      "num_input_tokens_seen": 332274832,
      "step": 20290,
      "train_runtime": 39076.2215,
      "train_tokens_per_second": 8503.249
    },
    {
      "epoch": 1.3664826286022085,
      "grad_norm": 0.9365584657803665,
      "learning_rate": 2.2784607808625463e-06,
      "loss": 0.2956104278564453,
      "num_input_tokens_seen": 332356752,
      "step": 20295,
      "train_runtime": 39085.7828,
      "train_tokens_per_second": 8503.265
    },
    {
      "epoch": 1.3668192835981685,
      "grad_norm": 1.1864223860202132,
      "learning_rate": 2.276243079179634e-06,
      "loss": 0.2922065734863281,
      "num_input_tokens_seen": 332438288,
      "step": 20300,
      "train_runtime": 39095.3623,
      "train_tokens_per_second": 8503.267
    },
    {
      "epoch": 1.3671559385941288,
      "grad_norm": 1.1189074822217413,
      "learning_rate": 2.2740261391866634e-06,
      "loss": 0.3420071601867676,
      "num_input_tokens_seen": 332520208,
      "step": 20305,
      "train_runtime": 39105.0343,
      "train_tokens_per_second": 8503.258
    },
    {
      "epoch": 1.367492593590089,
      "grad_norm": 0.9245085155575448,
      "learning_rate": 2.2718099615035865e-06,
      "loss": 0.2515967845916748,
      "num_input_tokens_seen": 332602128,
      "step": 20310,
      "train_runtime": 39114.554,
      "train_tokens_per_second": 8503.283
    },
    {
      "epoch": 1.367829248586049,
      "grad_norm": 1.1419268124808108,
      "learning_rate": 2.2695945467501567e-06,
      "loss": 0.28384244441986084,
      "num_input_tokens_seen": 332684048,
      "step": 20315,
      "train_runtime": 39124.2269,
      "train_tokens_per_second": 8503.275
    },
    {
      "epoch": 1.368165903582009,
      "grad_norm": 0.9871857454578977,
      "learning_rate": 2.267379895545905e-06,
      "loss": 0.27557475566864015,
      "num_input_tokens_seen": 332765968,
      "step": 20320,
      "train_runtime": 39133.9019,
      "train_tokens_per_second": 8503.266
    },
    {
      "epoch": 1.3685025585779693,
      "grad_norm": 0.9519140360892416,
      "learning_rate": 2.2651660085101513e-06,
      "loss": 0.3000530004501343,
      "num_input_tokens_seen": 332847888,
      "step": 20325,
      "train_runtime": 39143.6187,
      "train_tokens_per_second": 8503.248
    },
    {
      "epoch": 1.3688392135739296,
      "grad_norm": 0.9568078604784133,
      "learning_rate": 2.262952886262003e-06,
      "loss": 0.30855140686035154,
      "num_input_tokens_seen": 332929808,
      "step": 20330,
      "train_runtime": 39153.3118,
      "train_tokens_per_second": 8503.235
    },
    {
      "epoch": 1.3691758685698896,
      "grad_norm": 1.0722501365708328,
      "learning_rate": 2.260740529420352e-06,
      "loss": 0.2950451850891113,
      "num_input_tokens_seen": 333011728,
      "step": 20335,
      "train_runtime": 39162.9768,
      "train_tokens_per_second": 8503.228
    },
    {
      "epoch": 1.3695125235658496,
      "grad_norm": 1.2105741867916173,
      "learning_rate": 2.2585289386038754e-06,
      "loss": 0.29564263820648196,
      "num_input_tokens_seen": 333093648,
      "step": 20340,
      "train_runtime": 39172.7107,
      "train_tokens_per_second": 8503.207
    },
    {
      "epoch": 1.3698491785618099,
      "grad_norm": 1.1131939258690537,
      "learning_rate": 2.256318114431043e-06,
      "loss": 0.2966634273529053,
      "num_input_tokens_seen": 333175568,
      "step": 20345,
      "train_runtime": 39182.2187,
      "train_tokens_per_second": 8503.234
    },
    {
      "epoch": 1.3701858335577701,
      "grad_norm": 1.1980860697047249,
      "learning_rate": 2.2541080575200973e-06,
      "loss": 0.27890493869781496,
      "num_input_tokens_seen": 333256920,
      "step": 20350,
      "train_runtime": 39192.0234,
      "train_tokens_per_second": 8503.182
    },
    {
      "epoch": 1.3705224885537302,
      "grad_norm": 1.2598141619751697,
      "learning_rate": 2.251898768489081e-06,
      "loss": 0.3184329032897949,
      "num_input_tokens_seen": 333338840,
      "step": 20355,
      "train_runtime": 39201.7219,
      "train_tokens_per_second": 8503.168
    },
    {
      "epoch": 1.3708591435496902,
      "grad_norm": 1.0863454072511183,
      "learning_rate": 2.249690247955807e-06,
      "loss": 0.31045458316802976,
      "num_input_tokens_seen": 333420760,
      "step": 20360,
      "train_runtime": 39211.4145,
      "train_tokens_per_second": 8503.156
    },
    {
      "epoch": 1.3711957985456504,
      "grad_norm": 1.0414956208266635,
      "learning_rate": 2.247482496537887e-06,
      "loss": 0.31400671005249026,
      "num_input_tokens_seen": 333502680,
      "step": 20365,
      "train_runtime": 39221.0526,
      "train_tokens_per_second": 8503.155
    },
    {
      "epoch": 1.3715324535416107,
      "grad_norm": 0.9615615583862244,
      "learning_rate": 2.24527551485271e-06,
      "loss": 0.28610663414001464,
      "num_input_tokens_seen": 333584600,
      "step": 20370,
      "train_runtime": 39230.5657,
      "train_tokens_per_second": 8503.181
    },
    {
      "epoch": 1.3718691085375707,
      "grad_norm": 1.2548088326891689,
      "learning_rate": 2.243069303517451e-06,
      "loss": 0.3129443645477295,
      "num_input_tokens_seen": 333666520,
      "step": 20375,
      "train_runtime": 39240.1929,
      "train_tokens_per_second": 8503.182
    },
    {
      "epoch": 1.3722057635335307,
      "grad_norm": 1.0361179147115098,
      "learning_rate": 2.2408638631490696e-06,
      "loss": 0.24504492282867432,
      "num_input_tokens_seen": 333748440,
      "step": 20380,
      "train_runtime": 39249.7985,
      "train_tokens_per_second": 8503.189
    },
    {
      "epoch": 1.372542418529491,
      "grad_norm": 0.9324414743521089,
      "learning_rate": 2.2386591943643117e-06,
      "loss": 0.2600815534591675,
      "num_input_tokens_seen": 333830360,
      "step": 20385,
      "train_runtime": 39259.4565,
      "train_tokens_per_second": 8503.183
    },
    {
      "epoch": 1.3728790735254512,
      "grad_norm": 1.3617642393153413,
      "learning_rate": 2.2364552977797023e-06,
      "loss": 0.29639520645141604,
      "num_input_tokens_seen": 333912280,
      "step": 20390,
      "train_runtime": 39269.1599,
      "train_tokens_per_second": 8503.168
    },
    {
      "epoch": 1.3732157285214113,
      "grad_norm": 1.1252229510944924,
      "learning_rate": 2.23425217401156e-06,
      "loss": 0.26079583168029785,
      "num_input_tokens_seen": 333994200,
      "step": 20395,
      "train_runtime": 39278.6687,
      "train_tokens_per_second": 8503.196
    },
    {
      "epoch": 1.3735523835173713,
      "grad_norm": 1.7339344524583429,
      "learning_rate": 2.2320498236759746e-06,
      "loss": 0.29015583992004396,
      "num_input_tokens_seen": 334076120,
      "step": 20400,
      "train_runtime": 39288.1825,
      "train_tokens_per_second": 8503.222
    },
    {
      "epoch": 1.3738890385133316,
      "grad_norm": 0.9834738951129008,
      "learning_rate": 2.2298482473888335e-06,
      "loss": 0.2970778703689575,
      "num_input_tokens_seen": 334158040,
      "step": 20405,
      "train_runtime": 39297.8464,
      "train_tokens_per_second": 8503.215
    },
    {
      "epoch": 1.3742256935092918,
      "grad_norm": 1.0408829919724945,
      "learning_rate": 2.227647445765792e-06,
      "loss": 0.2815716743469238,
      "num_input_tokens_seen": 334239960,
      "step": 20410,
      "train_runtime": 39307.5296,
      "train_tokens_per_second": 8503.204
    },
    {
      "epoch": 1.3745623485052518,
      "grad_norm": 1.103205341371261,
      "learning_rate": 2.225447419422305e-06,
      "loss": 0.2687265396118164,
      "num_input_tokens_seen": 334321880,
      "step": 20415,
      "train_runtime": 39317.2416,
      "train_tokens_per_second": 8503.188
    },
    {
      "epoch": 1.3748990035012119,
      "grad_norm": 0.9977402312381675,
      "learning_rate": 2.223248168973599e-06,
      "loss": 0.25371999740600587,
      "num_input_tokens_seen": 334403800,
      "step": 20420,
      "train_runtime": 39326.9428,
      "train_tokens_per_second": 8503.173
    },
    {
      "epoch": 1.375235658497172,
      "grad_norm": 1.00722559496768,
      "learning_rate": 2.221049695034688e-06,
      "loss": 0.26945621967315675,
      "num_input_tokens_seen": 334485720,
      "step": 20425,
      "train_runtime": 39336.6235,
      "train_tokens_per_second": 8503.163
    },
    {
      "epoch": 1.3755723134931324,
      "grad_norm": 1.172350100404657,
      "learning_rate": 2.218851998220369e-06,
      "loss": 0.3090240955352783,
      "num_input_tokens_seen": 334567640,
      "step": 20430,
      "train_runtime": 39346.4058,
      "train_tokens_per_second": 8503.131
    },
    {
      "epoch": 1.3759089684890924,
      "grad_norm": 1.1516870354329238,
      "learning_rate": 2.21665507914522e-06,
      "loss": 0.28406033515930174,
      "num_input_tokens_seen": 334649560,
      "step": 20435,
      "train_runtime": 39356.2028,
      "train_tokens_per_second": 8503.096
    },
    {
      "epoch": 1.3762456234850524,
      "grad_norm": 1.0624039226850266,
      "learning_rate": 2.214458938423601e-06,
      "loss": 0.29752354621887206,
      "num_input_tokens_seen": 334731480,
      "step": 20440,
      "train_runtime": 39365.7207,
      "train_tokens_per_second": 8503.121
    },
    {
      "epoch": 1.3765822784810127,
      "grad_norm": 1.0137579945190411,
      "learning_rate": 2.212263576669661e-06,
      "loss": 0.26182172298431394,
      "num_input_tokens_seen": 334813400,
      "step": 20445,
      "train_runtime": 39375.345,
      "train_tokens_per_second": 8503.123
    },
    {
      "epoch": 1.376918933476973,
      "grad_norm": 1.1779157138202674,
      "learning_rate": 2.2100689944973182e-06,
      "loss": 0.2673488616943359,
      "num_input_tokens_seen": 334895320,
      "step": 20450,
      "train_runtime": 39385.0192,
      "train_tokens_per_second": 8503.114
    },
    {
      "epoch": 1.377255588472933,
      "grad_norm": 1.0870429237019916,
      "learning_rate": 2.2078751925202886e-06,
      "loss": 0.31511197090148924,
      "num_input_tokens_seen": 334977240,
      "step": 20455,
      "train_runtime": 39394.6684,
      "train_tokens_per_second": 8503.111
    },
    {
      "epoch": 1.377592243468893,
      "grad_norm": 1.1970198484035885,
      "learning_rate": 2.205682171352054e-06,
      "loss": 0.31320767402648925,
      "num_input_tokens_seen": 335059160,
      "step": 20460,
      "train_runtime": 39404.1906,
      "train_tokens_per_second": 8503.135
    },
    {
      "epoch": 1.3779288984648532,
      "grad_norm": 1.0675953868677575,
      "learning_rate": 2.2034899316058926e-06,
      "loss": 0.245933198928833,
      "num_input_tokens_seen": 335141080,
      "step": 20465,
      "train_runtime": 39413.8744,
      "train_tokens_per_second": 8503.124
    },
    {
      "epoch": 1.3782655534608135,
      "grad_norm": 1.0381099295170795,
      "learning_rate": 2.201298473894854e-06,
      "loss": 0.2956599950790405,
      "num_input_tokens_seen": 335223000,
      "step": 20470,
      "train_runtime": 39423.3786,
      "train_tokens_per_second": 8503.153
    },
    {
      "epoch": 1.3786022084567735,
      "grad_norm": 1.044400124941879,
      "learning_rate": 2.1991077988317723e-06,
      "loss": 0.27970571517944337,
      "num_input_tokens_seen": 335304920,
      "step": 20475,
      "train_runtime": 39432.9046,
      "train_tokens_per_second": 8503.176
    },
    {
      "epoch": 1.3789388634527335,
      "grad_norm": 1.1231337153307246,
      "learning_rate": 2.1969179070292637e-06,
      "loss": 0.2824363708496094,
      "num_input_tokens_seen": 335386840,
      "step": 20480,
      "train_runtime": 39442.5694,
      "train_tokens_per_second": 8503.169
    },
    {
      "epoch": 1.3792755184486938,
      "grad_norm": 1.2650301651388625,
      "learning_rate": 2.1947287990997236e-06,
      "loss": 0.2763112783432007,
      "num_input_tokens_seen": 335468760,
      "step": 20485,
      "train_runtime": 39452.2227,
      "train_tokens_per_second": 8503.165
    },
    {
      "epoch": 1.379612173444654,
      "grad_norm": 1.1881543549144502,
      "learning_rate": 2.192540475655328e-06,
      "loss": 0.286287784576416,
      "num_input_tokens_seen": 335550680,
      "step": 20490,
      "train_runtime": 39461.8393,
      "train_tokens_per_second": 8503.169
    },
    {
      "epoch": 1.379948828440614,
      "grad_norm": 1.127790946384757,
      "learning_rate": 2.1903529373080397e-06,
      "loss": 0.2858565330505371,
      "num_input_tokens_seen": 335632600,
      "step": 20495,
      "train_runtime": 39471.5708,
      "train_tokens_per_second": 8503.148
    },
    {
      "epoch": 1.380285483436574,
      "grad_norm": 1.1523588858585696,
      "learning_rate": 2.1881661846695895e-06,
      "loss": 0.2674875736236572,
      "num_input_tokens_seen": 335714520,
      "step": 20500,
      "train_runtime": 39481.2555,
      "train_tokens_per_second": 8503.137
    },
    {
      "epoch": 1.3806221384325343,
      "grad_norm": 1.174631934812137,
      "learning_rate": 2.185980218351503e-06,
      "loss": 0.32117903232574463,
      "num_input_tokens_seen": 335796440,
      "step": 20505,
      "train_runtime": 39490.9466,
      "train_tokens_per_second": 8503.125
    },
    {
      "epoch": 1.3809587934284946,
      "grad_norm": 1.0456248860182638,
      "learning_rate": 2.1837950389650714e-06,
      "loss": 0.27762575149536134,
      "num_input_tokens_seen": 335878360,
      "step": 20510,
      "train_runtime": 39500.6153,
      "train_tokens_per_second": 8503.117
    },
    {
      "epoch": 1.3812954484244546,
      "grad_norm": 0.9409745924548237,
      "learning_rate": 2.1816106471213787e-06,
      "loss": 0.2843282222747803,
      "num_input_tokens_seen": 335960280,
      "step": 20515,
      "train_runtime": 39510.1319,
      "train_tokens_per_second": 8503.142
    },
    {
      "epoch": 1.3816321034204146,
      "grad_norm": 1.1626164856069061,
      "learning_rate": 2.179427043431282e-06,
      "loss": 0.3279150724411011,
      "num_input_tokens_seen": 336042200,
      "step": 20520,
      "train_runtime": 39519.6514,
      "train_tokens_per_second": 8503.167
    },
    {
      "epoch": 1.381968758416375,
      "grad_norm": 0.9541928187559964,
      "learning_rate": 2.177244228505418e-06,
      "loss": 0.2927178144454956,
      "num_input_tokens_seen": 336124120,
      "step": 20525,
      "train_runtime": 39529.1541,
      "train_tokens_per_second": 8503.195
    },
    {
      "epoch": 1.3823054134123351,
      "grad_norm": 1.27400397760255,
      "learning_rate": 2.1750622029542047e-06,
      "loss": 0.302512526512146,
      "num_input_tokens_seen": 336206040,
      "step": 20530,
      "train_runtime": 39538.6601,
      "train_tokens_per_second": 8503.223
    },
    {
      "epoch": 1.3826420684082952,
      "grad_norm": 1.0314911243781888,
      "learning_rate": 2.1728809673878374e-06,
      "loss": 0.26551499366760256,
      "num_input_tokens_seen": 336287960,
      "step": 20535,
      "train_runtime": 39548.1765,
      "train_tokens_per_second": 8503.248
    },
    {
      "epoch": 1.3829787234042552,
      "grad_norm": 0.9718426141423865,
      "learning_rate": 2.1707005224162915e-06,
      "loss": 0.2943101167678833,
      "num_input_tokens_seen": 336369880,
      "step": 20540,
      "train_runtime": 39557.735,
      "train_tokens_per_second": 8503.264
    },
    {
      "epoch": 1.3833153784002155,
      "grad_norm": 1.0284648472767843,
      "learning_rate": 2.1685208686493257e-06,
      "loss": 0.2680203437805176,
      "num_input_tokens_seen": 336450960,
      "step": 20545,
      "train_runtime": 39567.5127,
      "train_tokens_per_second": 8503.212
    },
    {
      "epoch": 1.3836520333961757,
      "grad_norm": 1.0846672174779364,
      "learning_rate": 2.166342006696467e-06,
      "loss": 0.3101027965545654,
      "num_input_tokens_seen": 336532880,
      "step": 20550,
      "train_runtime": 39577.1177,
      "train_tokens_per_second": 8503.219
    },
    {
      "epoch": 1.3839886883921357,
      "grad_norm": 1.1634548394496567,
      "learning_rate": 2.1641639371670335e-06,
      "loss": 0.2748044490814209,
      "num_input_tokens_seen": 336614800,
      "step": 20555,
      "train_runtime": 39586.7629,
      "train_tokens_per_second": 8503.216
    },
    {
      "epoch": 1.3843253433880958,
      "grad_norm": 1.2667141107426991,
      "learning_rate": 2.161986660670109e-06,
      "loss": 0.31628665924072263,
      "num_input_tokens_seen": 336696720,
      "step": 20560,
      "train_runtime": 39596.4494,
      "train_tokens_per_second": 8503.205
    },
    {
      "epoch": 1.384661998384056,
      "grad_norm": 0.9692186324702895,
      "learning_rate": 2.159810177814566e-06,
      "loss": 0.3167509317398071,
      "num_input_tokens_seen": 336778640,
      "step": 20565,
      "train_runtime": 39606.0947,
      "train_tokens_per_second": 8503.202
    },
    {
      "epoch": 1.3849986533800163,
      "grad_norm": 0.9555591886798385,
      "learning_rate": 2.1576344892090507e-06,
      "loss": 0.3029730796813965,
      "num_input_tokens_seen": 336860560,
      "step": 20570,
      "train_runtime": 39615.7587,
      "train_tokens_per_second": 8503.196
    },
    {
      "epoch": 1.3853353083759763,
      "grad_norm": 0.960954477482284,
      "learning_rate": 2.1554595954619866e-06,
      "loss": 0.30451412200927735,
      "num_input_tokens_seen": 336942480,
      "step": 20575,
      "train_runtime": 39625.4473,
      "train_tokens_per_second": 8503.184
    },
    {
      "epoch": 1.3856719633719363,
      "grad_norm": 1.0724401445326277,
      "learning_rate": 2.1532854971815763e-06,
      "loss": 0.2957176208496094,
      "num_input_tokens_seen": 337024400,
      "step": 20580,
      "train_runtime": 39635.1016,
      "train_tokens_per_second": 8503.18
    },
    {
      "epoch": 1.3860086183678966,
      "grad_norm": 0.9307748576628532,
      "learning_rate": 2.151112194975799e-06,
      "loss": 0.2615334987640381,
      "num_input_tokens_seen": 337106320,
      "step": 20585,
      "train_runtime": 39644.6025,
      "train_tokens_per_second": 8503.208
    },
    {
      "epoch": 1.3863452733638568,
      "grad_norm": 1.1119031762842575,
      "learning_rate": 2.1489396894524106e-06,
      "loss": 0.28707118034362794,
      "num_input_tokens_seen": 337188240,
      "step": 20590,
      "train_runtime": 39654.2403,
      "train_tokens_per_second": 8503.208
    },
    {
      "epoch": 1.3866819283598169,
      "grad_norm": 1.113548615214425,
      "learning_rate": 2.1467679812189497e-06,
      "loss": 0.3124527454376221,
      "num_input_tokens_seen": 337269824,
      "step": 20595,
      "train_runtime": 39664.1804,
      "train_tokens_per_second": 8503.134
    },
    {
      "epoch": 1.3870185833557769,
      "grad_norm": 1.0803336891762545,
      "learning_rate": 2.1445970708827213e-06,
      "loss": 0.2988314628601074,
      "num_input_tokens_seen": 337351744,
      "step": 20600,
      "train_runtime": 39673.7317,
      "train_tokens_per_second": 8503.151
    },
    {
      "epoch": 1.3873552383517371,
      "grad_norm": 1.0623018490920808,
      "learning_rate": 2.14242695905082e-06,
      "loss": 0.29074921607971194,
      "num_input_tokens_seen": 337433664,
      "step": 20605,
      "train_runtime": 39683.346,
      "train_tokens_per_second": 8503.156
    },
    {
      "epoch": 1.3876918933476974,
      "grad_norm": 1.2239200705841793,
      "learning_rate": 2.1402576463301044e-06,
      "loss": 0.28074538707733154,
      "num_input_tokens_seen": 337515584,
      "step": 20610,
      "train_runtime": 39692.8625,
      "train_tokens_per_second": 8503.181
    },
    {
      "epoch": 1.3880285483436574,
      "grad_norm": 1.4591969846854473,
      "learning_rate": 2.13808913332722e-06,
      "loss": 0.2863782405853271,
      "num_input_tokens_seen": 337597504,
      "step": 20615,
      "train_runtime": 39702.4658,
      "train_tokens_per_second": 8503.187
    },
    {
      "epoch": 1.3883652033396174,
      "grad_norm": 1.1811583477519254,
      "learning_rate": 2.1359214206485845e-06,
      "loss": 0.2797116279602051,
      "num_input_tokens_seen": 337679424,
      "step": 20620,
      "train_runtime": 39711.978,
      "train_tokens_per_second": 8503.213
    },
    {
      "epoch": 1.3887018583355777,
      "grad_norm": 0.9033543325059183,
      "learning_rate": 2.1337545089003905e-06,
      "loss": 0.24509794712066652,
      "num_input_tokens_seen": 337761344,
      "step": 20625,
      "train_runtime": 39721.5747,
      "train_tokens_per_second": 8503.221
    },
    {
      "epoch": 1.389038513331538,
      "grad_norm": 1.1289455276899407,
      "learning_rate": 2.131588398688608e-06,
      "loss": 0.31659817695617676,
      "num_input_tokens_seen": 337843264,
      "step": 20630,
      "train_runtime": 39731.0847,
      "train_tokens_per_second": 8503.248
    },
    {
      "epoch": 1.389375168327498,
      "grad_norm": 1.0056551752399856,
      "learning_rate": 2.1294230906189833e-06,
      "loss": 0.2789276599884033,
      "num_input_tokens_seen": 337925184,
      "step": 20635,
      "train_runtime": 39740.6866,
      "train_tokens_per_second": 8503.255
    },
    {
      "epoch": 1.389711823323458,
      "grad_norm": 0.9608135851750244,
      "learning_rate": 2.1272585852970358e-06,
      "loss": 0.2735706329345703,
      "num_input_tokens_seen": 338007104,
      "step": 20640,
      "train_runtime": 39750.3633,
      "train_tokens_per_second": 8503.246
    },
    {
      "epoch": 1.3900484783194182,
      "grad_norm": 1.3748218622200499,
      "learning_rate": 2.125094883328068e-06,
      "loss": 0.270487380027771,
      "num_input_tokens_seen": 338089024,
      "step": 20645,
      "train_runtime": 39760.0675,
      "train_tokens_per_second": 8503.231
    },
    {
      "epoch": 1.3903851333153785,
      "grad_norm": 0.8864248616038778,
      "learning_rate": 2.1229319853171447e-06,
      "loss": 0.2712353229522705,
      "num_input_tokens_seen": 338170944,
      "step": 20650,
      "train_runtime": 39769.6701,
      "train_tokens_per_second": 8503.237
    },
    {
      "epoch": 1.3907217883113385,
      "grad_norm": 1.1544959149581708,
      "learning_rate": 2.1207698918691217e-06,
      "loss": 0.2869921445846558,
      "num_input_tokens_seen": 338252864,
      "step": 20655,
      "train_runtime": 39779.2147,
      "train_tokens_per_second": 8503.256
    },
    {
      "epoch": 1.3910584433072986,
      "grad_norm": 1.044826833834973,
      "learning_rate": 2.1186086035886133e-06,
      "loss": 0.2974970579147339,
      "num_input_tokens_seen": 338334784,
      "step": 20660,
      "train_runtime": 39788.8862,
      "train_tokens_per_second": 8503.248
    },
    {
      "epoch": 1.3913950983032588,
      "grad_norm": 1.2827629440026724,
      "learning_rate": 2.116448121080023e-06,
      "loss": 0.29353647232055663,
      "num_input_tokens_seen": 338416704,
      "step": 20665,
      "train_runtime": 39798.5341,
      "train_tokens_per_second": 8503.245
    },
    {
      "epoch": 1.391731753299219,
      "grad_norm": 0.910229609221341,
      "learning_rate": 2.1142884449475205e-06,
      "loss": 0.2853943586349487,
      "num_input_tokens_seen": 338498624,
      "step": 20670,
      "train_runtime": 39808.1781,
      "train_tokens_per_second": 8503.243
    },
    {
      "epoch": 1.392068408295179,
      "grad_norm": 1.135585575695245,
      "learning_rate": 2.112129575795052e-06,
      "loss": 0.2631541728973389,
      "num_input_tokens_seen": 338580544,
      "step": 20675,
      "train_runtime": 39817.721,
      "train_tokens_per_second": 8503.263
    },
    {
      "epoch": 1.3924050632911391,
      "grad_norm": 0.9404120307877434,
      "learning_rate": 2.1099715142263394e-06,
      "loss": 0.25963003635406495,
      "num_input_tokens_seen": 338662464,
      "step": 20680,
      "train_runtime": 39827.3151,
      "train_tokens_per_second": 8503.271
    },
    {
      "epoch": 1.3927417182870994,
      "grad_norm": 0.9210875737199792,
      "learning_rate": 2.1078142608448773e-06,
      "loss": 0.2956238269805908,
      "num_input_tokens_seen": 338744384,
      "step": 20685,
      "train_runtime": 39836.9906,
      "train_tokens_per_second": 8503.262
    },
    {
      "epoch": 1.3930783732830596,
      "grad_norm": 1.3454942864988426,
      "learning_rate": 2.1056578162539327e-06,
      "loss": 0.29031124114990237,
      "num_input_tokens_seen": 338826304,
      "step": 20690,
      "train_runtime": 39846.5194,
      "train_tokens_per_second": 8503.285
    },
    {
      "epoch": 1.3934150282790196,
      "grad_norm": 1.1679739325552345,
      "learning_rate": 2.103502181056554e-06,
      "loss": 0.2989187240600586,
      "num_input_tokens_seen": 338908224,
      "step": 20695,
      "train_runtime": 39856.2679,
      "train_tokens_per_second": 8503.26
    },
    {
      "epoch": 1.3937516832749797,
      "grad_norm": 1.214666194679275,
      "learning_rate": 2.1013473558555503e-06,
      "loss": 0.27613234519958496,
      "num_input_tokens_seen": 338990144,
      "step": 20700,
      "train_runtime": 39865.7823,
      "train_tokens_per_second": 8503.286
    },
    {
      "epoch": 1.39408833827094,
      "grad_norm": 1.00010191503137,
      "learning_rate": 2.099193341253519e-06,
      "loss": 0.29334297180175783,
      "num_input_tokens_seen": 339072064,
      "step": 20705,
      "train_runtime": 39875.485,
      "train_tokens_per_second": 8503.271
    },
    {
      "epoch": 1.3944249932669002,
      "grad_norm": 1.1877072727341202,
      "learning_rate": 2.0970401378528167e-06,
      "loss": 0.27960925102233886,
      "num_input_tokens_seen": 339153984,
      "step": 20710,
      "train_runtime": 39885.1096,
      "train_tokens_per_second": 8503.273
    },
    {
      "epoch": 1.3947616482628602,
      "grad_norm": 1.0541783587774272,
      "learning_rate": 2.0948877462555844e-06,
      "loss": 0.27819778919219973,
      "num_input_tokens_seen": 339235904,
      "step": 20715,
      "train_runtime": 39894.8023,
      "train_tokens_per_second": 8503.261
    },
    {
      "epoch": 1.3950983032588202,
      "grad_norm": 1.0752685904681911,
      "learning_rate": 2.09273616706373e-06,
      "loss": 0.3038111925125122,
      "num_input_tokens_seen": 339317824,
      "step": 20720,
      "train_runtime": 39904.4955,
      "train_tokens_per_second": 8503.248
    },
    {
      "epoch": 1.3954349582547805,
      "grad_norm": 1.1678638567284796,
      "learning_rate": 2.090585400878936e-06,
      "loss": 0.2759523868560791,
      "num_input_tokens_seen": 339399744,
      "step": 20725,
      "train_runtime": 39914.1888,
      "train_tokens_per_second": 8503.235
    },
    {
      "epoch": 1.3957716132507407,
      "grad_norm": 1.3080459297936746,
      "learning_rate": 2.0884354483026576e-06,
      "loss": 0.29725263118743894,
      "num_input_tokens_seen": 339481664,
      "step": 20730,
      "train_runtime": 39923.8847,
      "train_tokens_per_second": 8503.222
    },
    {
      "epoch": 1.3961082682467008,
      "grad_norm": 0.9611990735541144,
      "learning_rate": 2.0862863099361214e-06,
      "loss": 0.3007199764251709,
      "num_input_tokens_seen": 339563584,
      "step": 20735,
      "train_runtime": 39933.5998,
      "train_tokens_per_second": 8503.205
    },
    {
      "epoch": 1.3964449232426608,
      "grad_norm": 1.207759747206983,
      "learning_rate": 2.084137986380326e-06,
      "loss": 0.32167530059814453,
      "num_input_tokens_seen": 339645504,
      "step": 20740,
      "train_runtime": 39943.2692,
      "train_tokens_per_second": 8503.197
    },
    {
      "epoch": 1.396781578238621,
      "grad_norm": 1.0325809933393846,
      "learning_rate": 2.081990478236048e-06,
      "loss": 0.27014613151550293,
      "num_input_tokens_seen": 339727424,
      "step": 20745,
      "train_runtime": 39952.9418,
      "train_tokens_per_second": 8503.189
    },
    {
      "epoch": 1.3971182332345813,
      "grad_norm": 1.0841348131634796,
      "learning_rate": 2.079843786103824e-06,
      "loss": 0.28767056465148927,
      "num_input_tokens_seen": 339809344,
      "step": 20750,
      "train_runtime": 39962.6734,
      "train_tokens_per_second": 8503.168
    },
    {
      "epoch": 1.3974548882305413,
      "grad_norm": 1.2618651957464946,
      "learning_rate": 2.0776979105839774e-06,
      "loss": 0.2852169513702393,
      "num_input_tokens_seen": 339891264,
      "step": 20755,
      "train_runtime": 39972.401,
      "train_tokens_per_second": 8503.149
    },
    {
      "epoch": 1.3977915432265013,
      "grad_norm": 1.3980398062081714,
      "learning_rate": 2.075552852276588e-06,
      "loss": 0.32133636474609373,
      "num_input_tokens_seen": 339973184,
      "step": 20760,
      "train_runtime": 39982.0759,
      "train_tokens_per_second": 8503.14
    },
    {
      "epoch": 1.3981281982224616,
      "grad_norm": 1.2230230023825384,
      "learning_rate": 2.07340861178152e-06,
      "loss": 0.27669601440429686,
      "num_input_tokens_seen": 340055104,
      "step": 20765,
      "train_runtime": 39991.5884,
      "train_tokens_per_second": 8503.166
    },
    {
      "epoch": 1.3984648532184218,
      "grad_norm": 1.1011216635169079,
      "learning_rate": 2.0712651896984013e-06,
      "loss": 0.2967384338378906,
      "num_input_tokens_seen": 340137024,
      "step": 20770,
      "train_runtime": 40001.0919,
      "train_tokens_per_second": 8503.193
    },
    {
      "epoch": 1.3988015082143819,
      "grad_norm": 1.1099599321455556,
      "learning_rate": 2.0691225866266335e-06,
      "loss": 0.3062562942504883,
      "num_input_tokens_seen": 340218944,
      "step": 20775,
      "train_runtime": 40010.6078,
      "train_tokens_per_second": 8503.219
    },
    {
      "epoch": 1.3991381632103421,
      "grad_norm": 1.0690166184307255,
      "learning_rate": 2.066980803165388e-06,
      "loss": 0.3056901454925537,
      "num_input_tokens_seen": 340300864,
      "step": 20780,
      "train_runtime": 40020.3287,
      "train_tokens_per_second": 8503.2
    },
    {
      "epoch": 1.3994748182063022,
      "grad_norm": 1.024755137422685,
      "learning_rate": 2.0648398399136088e-06,
      "loss": 0.29085416793823243,
      "num_input_tokens_seen": 340382784,
      "step": 20785,
      "train_runtime": 40030.0393,
      "train_tokens_per_second": 8503.184
    },
    {
      "epoch": 1.3998114732022624,
      "grad_norm": 0.9325613179814055,
      "learning_rate": 2.062699697470007e-06,
      "loss": 0.2706148147583008,
      "num_input_tokens_seen": 340464704,
      "step": 20790,
      "train_runtime": 40039.6979,
      "train_tokens_per_second": 8503.179
    },
    {
      "epoch": 1.4001481281982224,
      "grad_norm": 1.3562329900018903,
      "learning_rate": 2.060560376433072e-06,
      "loss": 0.3233614683151245,
      "num_input_tokens_seen": 340546624,
      "step": 20795,
      "train_runtime": 40049.4242,
      "train_tokens_per_second": 8503.159
    },
    {
      "epoch": 1.4004847831941827,
      "grad_norm": 1.093201993274004,
      "learning_rate": 2.05842187740105e-06,
      "loss": 0.30796046257019044,
      "num_input_tokens_seen": 340628544,
      "step": 20800,
      "train_runtime": 40059.1408,
      "train_tokens_per_second": 8503.142
    },
    {
      "epoch": 1.4008214381901427,
      "grad_norm": 1.0920461174675127,
      "learning_rate": 2.0562842009719745e-06,
      "loss": 0.2582095623016357,
      "num_input_tokens_seen": 340710464,
      "step": 20805,
      "train_runtime": 40068.8359,
      "train_tokens_per_second": 8503.129
    },
    {
      "epoch": 1.401158093186103,
      "grad_norm": 1.1905508025536338,
      "learning_rate": 2.0541473477436317e-06,
      "loss": 0.2721451759338379,
      "num_input_tokens_seen": 340792384,
      "step": 20810,
      "train_runtime": 40078.4759,
      "train_tokens_per_second": 8503.127
    },
    {
      "epoch": 1.401494748182063,
      "grad_norm": 1.3600389972278308,
      "learning_rate": 2.052011318313591e-06,
      "loss": 0.2704375505447388,
      "num_input_tokens_seen": 340874304,
      "step": 20815,
      "train_runtime": 40088.179,
      "train_tokens_per_second": 8503.113
    },
    {
      "epoch": 1.4018314031780232,
      "grad_norm": 1.2992810544789717,
      "learning_rate": 2.049876113279185e-06,
      "loss": 0.32093856334686277,
      "num_input_tokens_seen": 340956224,
      "step": 20820,
      "train_runtime": 40097.8559,
      "train_tokens_per_second": 8503.104
    },
    {
      "epoch": 1.4021680581739833,
      "grad_norm": 1.1531468550964081,
      "learning_rate": 2.0477417332375165e-06,
      "loss": 0.2657993078231812,
      "num_input_tokens_seen": 341038144,
      "step": 20825,
      "train_runtime": 40107.5497,
      "train_tokens_per_second": 8503.091
    },
    {
      "epoch": 1.4025047131699435,
      "grad_norm": 1.0686440083879016,
      "learning_rate": 2.045608178785458e-06,
      "loss": 0.28987321853637693,
      "num_input_tokens_seen": 341120064,
      "step": 20830,
      "train_runtime": 40117.0605,
      "train_tokens_per_second": 8503.117
    },
    {
      "epoch": 1.4028413681659035,
      "grad_norm": 1.0784124211147081,
      "learning_rate": 2.043475450519652e-06,
      "loss": 0.2633155584335327,
      "num_input_tokens_seen": 341201984,
      "step": 20835,
      "train_runtime": 40126.6918,
      "train_tokens_per_second": 8503.118
    },
    {
      "epoch": 1.4031780231618638,
      "grad_norm": 1.1221291611337239,
      "learning_rate": 2.0413435490365064e-06,
      "loss": 0.2776984453201294,
      "num_input_tokens_seen": 341283904,
      "step": 20840,
      "train_runtime": 40136.3623,
      "train_tokens_per_second": 8503.11
    },
    {
      "epoch": 1.4035146781578238,
      "grad_norm": 1.289499094023469,
      "learning_rate": 2.0392124749322064e-06,
      "loss": 0.31078810691833497,
      "num_input_tokens_seen": 341365824,
      "step": 20845,
      "train_runtime": 40145.8778,
      "train_tokens_per_second": 8503.135
    },
    {
      "epoch": 1.403851333153784,
      "grad_norm": 0.9910855013499533,
      "learning_rate": 2.037082228802693e-06,
      "loss": 0.26931712627410886,
      "num_input_tokens_seen": 341447744,
      "step": 20850,
      "train_runtime": 40155.5803,
      "train_tokens_per_second": 8503.121
    },
    {
      "epoch": 1.404187988149744,
      "grad_norm": 0.9999701601997574,
      "learning_rate": 2.03495281124369e-06,
      "loss": 0.2399512767791748,
      "num_input_tokens_seen": 341529392,
      "step": 20855,
      "train_runtime": 40165.5016,
      "train_tokens_per_second": 8503.053
    },
    {
      "epoch": 1.4045246431457044,
      "grad_norm": 1.016721415951789,
      "learning_rate": 2.0328242228506746e-06,
      "loss": 0.26685926914215086,
      "num_input_tokens_seen": 341611312,
      "step": 20860,
      "train_runtime": 40175.1355,
      "train_tokens_per_second": 8503.053
    },
    {
      "epoch": 1.4048612981416644,
      "grad_norm": 1.078202407923467,
      "learning_rate": 2.0306964642189053e-06,
      "loss": 0.26873958110809326,
      "num_input_tokens_seen": 341693232,
      "step": 20865,
      "train_runtime": 40184.697,
      "train_tokens_per_second": 8503.069
    },
    {
      "epoch": 1.4051979531376246,
      "grad_norm": 0.9729505625582356,
      "learning_rate": 2.028569535943402e-06,
      "loss": 0.2764342546463013,
      "num_input_tokens_seen": 341775152,
      "step": 20870,
      "train_runtime": 40194.3384,
      "train_tokens_per_second": 8503.067
    },
    {
      "epoch": 1.4055346081335847,
      "grad_norm": 1.2910163354145474,
      "learning_rate": 2.0264434386189524e-06,
      "loss": 0.3126168727874756,
      "num_input_tokens_seen": 341857072,
      "step": 20875,
      "train_runtime": 40204.0281,
      "train_tokens_per_second": 8503.055
    },
    {
      "epoch": 1.405871263129545,
      "grad_norm": 1.0804424539296846,
      "learning_rate": 2.0243181728401136e-06,
      "loss": 0.2837043046951294,
      "num_input_tokens_seen": 341938992,
      "step": 20880,
      "train_runtime": 40213.7004,
      "train_tokens_per_second": 8503.047
    },
    {
      "epoch": 1.406207918125505,
      "grad_norm": 1.151608754382011,
      "learning_rate": 2.022193739201208e-06,
      "loss": 0.3077866554260254,
      "num_input_tokens_seen": 342020912,
      "step": 20885,
      "train_runtime": 40223.3547,
      "train_tokens_per_second": 8503.043
    },
    {
      "epoch": 1.4065445731214652,
      "grad_norm": 1.0643949259727785,
      "learning_rate": 2.0200701382963267e-06,
      "loss": 0.2711671829223633,
      "num_input_tokens_seen": 342102832,
      "step": 20890,
      "train_runtime": 40233.0296,
      "train_tokens_per_second": 8503.034
    },
    {
      "epoch": 1.4068812281174252,
      "grad_norm": 1.0400038634458473,
      "learning_rate": 2.0179473707193324e-06,
      "loss": 0.28544976711273196,
      "num_input_tokens_seen": 342184752,
      "step": 20895,
      "train_runtime": 40242.7519,
      "train_tokens_per_second": 8503.016
    },
    {
      "epoch": 1.4072178831133855,
      "grad_norm": 0.948731816143397,
      "learning_rate": 2.0158254370638435e-06,
      "loss": 0.26539196968078616,
      "num_input_tokens_seen": 342266672,
      "step": 20900,
      "train_runtime": 40252.4706,
      "train_tokens_per_second": 8502.998
    },
    {
      "epoch": 1.4075545381093455,
      "grad_norm": 1.0468560747169977,
      "learning_rate": 2.0137043379232585e-06,
      "loss": 0.30677156448364257,
      "num_input_tokens_seen": 342348592,
      "step": 20905,
      "train_runtime": 40262.1761,
      "train_tokens_per_second": 8502.983
    },
    {
      "epoch": 1.4078911931053057,
      "grad_norm": 0.9433999893242844,
      "learning_rate": 2.0115840738907293e-06,
      "loss": 0.25561976432800293,
      "num_input_tokens_seen": 342430512,
      "step": 20910,
      "train_runtime": 40271.8883,
      "train_tokens_per_second": 8502.966
    },
    {
      "epoch": 1.4082278481012658,
      "grad_norm": 0.9535378268401242,
      "learning_rate": 2.0094646455591865e-06,
      "loss": 0.25310192108154295,
      "num_input_tokens_seen": 342512432,
      "step": 20915,
      "train_runtime": 40281.407,
      "train_tokens_per_second": 8502.991
    },
    {
      "epoch": 1.408564503097226,
      "grad_norm": 1.0872199723247586,
      "learning_rate": 2.007346053521319e-06,
      "loss": 0.29708964824676515,
      "num_input_tokens_seen": 342594352,
      "step": 20920,
      "train_runtime": 40290.9315,
      "train_tokens_per_second": 8503.014
    },
    {
      "epoch": 1.408901158093186,
      "grad_norm": 1.463508994278999,
      "learning_rate": 2.005228298369583e-06,
      "loss": 0.3043045997619629,
      "num_input_tokens_seen": 342675184,
      "step": 20925,
      "train_runtime": 40300.5837,
      "train_tokens_per_second": 8502.983
    },
    {
      "epoch": 1.4092378130891463,
      "grad_norm": 1.2423866386930509,
      "learning_rate": 2.003111380696208e-06,
      "loss": 0.2791585922241211,
      "num_input_tokens_seen": 342757104,
      "step": 20930,
      "train_runtime": 40310.1038,
      "train_tokens_per_second": 8503.007
    },
    {
      "epoch": 1.4095744680851063,
      "grad_norm": 1.2118579056131134,
      "learning_rate": 2.0009953010931747e-06,
      "loss": 0.25297889709472654,
      "num_input_tokens_seen": 342839024,
      "step": 20935,
      "train_runtime": 40319.787,
      "train_tokens_per_second": 8502.997
    },
    {
      "epoch": 1.4099111230810666,
      "grad_norm": 1.2510049534521885,
      "learning_rate": 1.998880060152244e-06,
      "loss": 0.3038308620452881,
      "num_input_tokens_seen": 342920944,
      "step": 20940,
      "train_runtime": 40329.492,
      "train_tokens_per_second": 8502.982
    },
    {
      "epoch": 1.4102477780770266,
      "grad_norm": 1.0861225353857356,
      "learning_rate": 1.9967656584649354e-06,
      "loss": 0.27735276222229005,
      "num_input_tokens_seen": 343002864,
      "step": 20945,
      "train_runtime": 40339.1703,
      "train_tokens_per_second": 8502.973
    },
    {
      "epoch": 1.4105844330729869,
      "grad_norm": 1.0480835077424546,
      "learning_rate": 1.994652096622533e-06,
      "loss": 0.2725769281387329,
      "num_input_tokens_seen": 343084784,
      "step": 20950,
      "train_runtime": 40348.899,
      "train_tokens_per_second": 8502.953
    },
    {
      "epoch": 1.410921088068947,
      "grad_norm": 1.0590169107679366,
      "learning_rate": 1.992539375216089e-06,
      "loss": 0.30078878402709963,
      "num_input_tokens_seen": 343166704,
      "step": 20955,
      "train_runtime": 40358.592,
      "train_tokens_per_second": 8502.94
    },
    {
      "epoch": 1.4112577430649071,
      "grad_norm": 1.1244427758229933,
      "learning_rate": 1.9904274948364184e-06,
      "loss": 0.2854571580886841,
      "num_input_tokens_seen": 343248624,
      "step": 20960,
      "train_runtime": 40368.2606,
      "train_tokens_per_second": 8502.933
    },
    {
      "epoch": 1.4115943980608672,
      "grad_norm": 1.1228040286404688,
      "learning_rate": 1.988316456074102e-06,
      "loss": 0.2861645698547363,
      "num_input_tokens_seen": 343330544,
      "step": 20965,
      "train_runtime": 40377.7821,
      "train_tokens_per_second": 8502.957
    },
    {
      "epoch": 1.4119310530568274,
      "grad_norm": 1.157338443997102,
      "learning_rate": 1.9862062595194854e-06,
      "loss": 0.30047001838684084,
      "num_input_tokens_seen": 343412464,
      "step": 20970,
      "train_runtime": 40387.4806,
      "train_tokens_per_second": 8502.943
    },
    {
      "epoch": 1.4122677080527875,
      "grad_norm": 1.2904167031243008,
      "learning_rate": 1.984096905762676e-06,
      "loss": 0.27982091903686523,
      "num_input_tokens_seen": 343494384,
      "step": 20975,
      "train_runtime": 40397.1495,
      "train_tokens_per_second": 8502.936
    },
    {
      "epoch": 1.4126043630487477,
      "grad_norm": 1.1918691584705918,
      "learning_rate": 1.9819883953935533e-06,
      "loss": 0.29467360973358153,
      "num_input_tokens_seen": 343576304,
      "step": 20980,
      "train_runtime": 40406.8144,
      "train_tokens_per_second": 8502.93
    },
    {
      "epoch": 1.4129410180447077,
      "grad_norm": 1.2608667917080194,
      "learning_rate": 1.9798807290017485e-06,
      "loss": 0.29174056053161623,
      "num_input_tokens_seen": 343658224,
      "step": 20985,
      "train_runtime": 40416.5453,
      "train_tokens_per_second": 8502.909
    },
    {
      "epoch": 1.413277673040668,
      "grad_norm": 0.8810069252056133,
      "learning_rate": 1.9777739071766693e-06,
      "loss": 0.26865057945251464,
      "num_input_tokens_seen": 343740144,
      "step": 20990,
      "train_runtime": 40426.2332,
      "train_tokens_per_second": 8502.898
    },
    {
      "epoch": 1.413614328036628,
      "grad_norm": 1.1226327112707002,
      "learning_rate": 1.97566793050748e-06,
      "loss": 0.27738997936248777,
      "num_input_tokens_seen": 343822064,
      "step": 20995,
      "train_runtime": 40435.8842,
      "train_tokens_per_second": 8502.895
    },
    {
      "epoch": 1.4139509830325883,
      "grad_norm": 1.042140867829353,
      "learning_rate": 1.9735627995831096e-06,
      "loss": 0.2787912368774414,
      "num_input_tokens_seen": 343903984,
      "step": 21000,
      "train_runtime": 40445.3896,
      "train_tokens_per_second": 8502.922
    },
    {
      "epoch": 1.4142876380285483,
      "grad_norm": 1.2694217863284474,
      "learning_rate": 1.9714585149922523e-06,
      "loss": 0.31298246383666994,
      "num_input_tokens_seen": 343985904,
      "step": 21005,
      "train_runtime": 40454.8967,
      "train_tokens_per_second": 8502.949
    },
    {
      "epoch": 1.4146242930245085,
      "grad_norm": 1.109865772962812,
      "learning_rate": 1.9693550773233634e-06,
      "loss": 0.2940122127532959,
      "num_input_tokens_seen": 344067824,
      "step": 21010,
      "train_runtime": 40464.42,
      "train_tokens_per_second": 8502.972
    },
    {
      "epoch": 1.4149609480204686,
      "grad_norm": 1.1209148925097987,
      "learning_rate": 1.967252487164663e-06,
      "loss": 0.27232358455657957,
      "num_input_tokens_seen": 344149744,
      "step": 21015,
      "train_runtime": 40473.9328,
      "train_tokens_per_second": 8502.997
    },
    {
      "epoch": 1.4152976030164288,
      "grad_norm": 1.0794614731364984,
      "learning_rate": 1.9651507451041344e-06,
      "loss": 0.3012054443359375,
      "num_input_tokens_seen": 344231664,
      "step": 21020,
      "train_runtime": 40483.442,
      "train_tokens_per_second": 8503.024
    },
    {
      "epoch": 1.4156342580123888,
      "grad_norm": 1.245075064981434,
      "learning_rate": 1.9630498517295204e-06,
      "loss": 0.30307128429412844,
      "num_input_tokens_seen": 344313584,
      "step": 21025,
      "train_runtime": 40493.09,
      "train_tokens_per_second": 8503.021
    },
    {
      "epoch": 1.415970913008349,
      "grad_norm": 1.136001603317287,
      "learning_rate": 1.960949807628335e-06,
      "loss": 0.26766195297241213,
      "num_input_tokens_seen": 344395504,
      "step": 21030,
      "train_runtime": 40502.6601,
      "train_tokens_per_second": 8503.034
    },
    {
      "epoch": 1.4163075680043091,
      "grad_norm": 1.1222989200512286,
      "learning_rate": 1.9588506133878415e-06,
      "loss": 0.3262138843536377,
      "num_input_tokens_seen": 344477424,
      "step": 21035,
      "train_runtime": 40512.2587,
      "train_tokens_per_second": 8503.042
    },
    {
      "epoch": 1.4166442230002694,
      "grad_norm": 0.9151454697492999,
      "learning_rate": 1.956752269595078e-06,
      "loss": 0.3093678951263428,
      "num_input_tokens_seen": 344559344,
      "step": 21040,
      "train_runtime": 40521.9068,
      "train_tokens_per_second": 8503.039
    },
    {
      "epoch": 1.4169808779962294,
      "grad_norm": 1.1770915973434828,
      "learning_rate": 1.9546547768368384e-06,
      "loss": 0.28089218139648436,
      "num_input_tokens_seen": 344641264,
      "step": 21045,
      "train_runtime": 40531.4661,
      "train_tokens_per_second": 8503.054
    },
    {
      "epoch": 1.4173175329921897,
      "grad_norm": 1.4932300524629571,
      "learning_rate": 1.9525581356996793e-06,
      "loss": 0.31073789596557616,
      "num_input_tokens_seen": 344723184,
      "step": 21050,
      "train_runtime": 40541.198,
      "train_tokens_per_second": 8503.034
    },
    {
      "epoch": 1.4176541879881497,
      "grad_norm": 1.2221366083661183,
      "learning_rate": 1.95046234676992e-06,
      "loss": 0.27796814441680906,
      "num_input_tokens_seen": 344805104,
      "step": 21055,
      "train_runtime": 40551.086,
      "train_tokens_per_second": 8502.981
    },
    {
      "epoch": 1.41799084298411,
      "grad_norm": 1.0502331640958875,
      "learning_rate": 1.9483674106336416e-06,
      "loss": 0.2733154296875,
      "num_input_tokens_seen": 344885776,
      "step": 21060,
      "train_runtime": 40560.6676,
      "train_tokens_per_second": 8502.961
    },
    {
      "epoch": 1.41832749798007,
      "grad_norm": 0.9775576610521729,
      "learning_rate": 1.9462733278766853e-06,
      "loss": 0.28854491710662844,
      "num_input_tokens_seen": 344967696,
      "step": 21065,
      "train_runtime": 40570.5379,
      "train_tokens_per_second": 8502.912
    },
    {
      "epoch": 1.4186641529760302,
      "grad_norm": 1.1390936492899755,
      "learning_rate": 1.9441800990846553e-06,
      "loss": 0.31240310668945315,
      "num_input_tokens_seen": 345049616,
      "step": 21070,
      "train_runtime": 40580.2162,
      "train_tokens_per_second": 8502.902
    },
    {
      "epoch": 1.4190008079719902,
      "grad_norm": 0.9818740870956513,
      "learning_rate": 1.942087724842915e-06,
      "loss": 0.3064249515533447,
      "num_input_tokens_seen": 345131536,
      "step": 21075,
      "train_runtime": 40589.8785,
      "train_tokens_per_second": 8502.896
    },
    {
      "epoch": 1.4193374629679505,
      "grad_norm": 1.1050654619703673,
      "learning_rate": 1.9399962057365944e-06,
      "loss": 0.28597373962402345,
      "num_input_tokens_seen": 345213456,
      "step": 21080,
      "train_runtime": 40599.3844,
      "train_tokens_per_second": 8502.923
    },
    {
      "epoch": 1.4196741179639105,
      "grad_norm": 1.138002081495229,
      "learning_rate": 1.937905542350574e-06,
      "loss": 0.28817660808563234,
      "num_input_tokens_seen": 345295376,
      "step": 21085,
      "train_runtime": 40608.974,
      "train_tokens_per_second": 8502.933
    },
    {
      "epoch": 1.4200107729598708,
      "grad_norm": 1.2191549259829826,
      "learning_rate": 1.935815735269506e-06,
      "loss": 0.3029794454574585,
      "num_input_tokens_seen": 345377008,
      "step": 21090,
      "train_runtime": 40619.0031,
      "train_tokens_per_second": 8502.843
    },
    {
      "epoch": 1.4203474279558308,
      "grad_norm": 1.1392070279248423,
      "learning_rate": 1.933726785077797e-06,
      "loss": 0.3002621650695801,
      "num_input_tokens_seen": 345458928,
      "step": 21095,
      "train_runtime": 40628.6773,
      "train_tokens_per_second": 8502.835
    },
    {
      "epoch": 1.420684082951791,
      "grad_norm": 1.1185366979919187,
      "learning_rate": 1.9316386923596146e-06,
      "loss": 0.27973902225494385,
      "num_input_tokens_seen": 345540848,
      "step": 21100,
      "train_runtime": 40638.2549,
      "train_tokens_per_second": 8502.847
    },
    {
      "epoch": 1.421020737947751,
      "grad_norm": 1.2747834710087769,
      "learning_rate": 1.929551457698888e-06,
      "loss": 0.31074569225311277,
      "num_input_tokens_seen": 345622768,
      "step": 21105,
      "train_runtime": 40647.7643,
      "train_tokens_per_second": 8502.873
    },
    {
      "epoch": 1.4213573929437113,
      "grad_norm": 1.1360241601863816,
      "learning_rate": 1.9274650816793044e-06,
      "loss": 0.27662522792816163,
      "num_input_tokens_seen": 345704688,
      "step": 21110,
      "train_runtime": 40657.3967,
      "train_tokens_per_second": 8502.873
    },
    {
      "epoch": 1.4216940479396714,
      "grad_norm": 1.1669747423355687,
      "learning_rate": 1.9253795648843136e-06,
      "loss": 0.3050896406173706,
      "num_input_tokens_seen": 345785808,
      "step": 21115,
      "train_runtime": 40667.1668,
      "train_tokens_per_second": 8502.825
    },
    {
      "epoch": 1.4220307029356316,
      "grad_norm": 1.0984057908870652,
      "learning_rate": 1.9232949078971236e-06,
      "loss": 0.290204381942749,
      "num_input_tokens_seen": 345867728,
      "step": 21120,
      "train_runtime": 40676.86,
      "train_tokens_per_second": 8502.813
    },
    {
      "epoch": 1.4223673579315916,
      "grad_norm": 1.0731434700836506,
      "learning_rate": 1.9212111113006998e-06,
      "loss": 0.2815673828125,
      "num_input_tokens_seen": 345949648,
      "step": 21125,
      "train_runtime": 40686.514,
      "train_tokens_per_second": 8502.809
    },
    {
      "epoch": 1.4227040129275519,
      "grad_norm": 1.1344484299148034,
      "learning_rate": 1.919128175677775e-06,
      "loss": 0.2847961187362671,
      "num_input_tokens_seen": 346031568,
      "step": 21130,
      "train_runtime": 40696.0584,
      "train_tokens_per_second": 8502.828
    },
    {
      "epoch": 1.423040667923512,
      "grad_norm": 1.3379077031895568,
      "learning_rate": 1.917046101610827e-06,
      "loss": 0.29941248893737793,
      "num_input_tokens_seen": 346113488,
      "step": 21135,
      "train_runtime": 40705.6497,
      "train_tokens_per_second": 8502.837
    },
    {
      "epoch": 1.4233773229194722,
      "grad_norm": 0.9720019100476166,
      "learning_rate": 1.9149648896821087e-06,
      "loss": 0.26144018173217776,
      "num_input_tokens_seen": 346195408,
      "step": 21140,
      "train_runtime": 40715.2868,
      "train_tokens_per_second": 8502.836
    },
    {
      "epoch": 1.4237139779154322,
      "grad_norm": 1.1565881467122157,
      "learning_rate": 1.912884540473621e-06,
      "loss": 0.2833715438842773,
      "num_input_tokens_seen": 346277328,
      "step": 21145,
      "train_runtime": 40724.983,
      "train_tokens_per_second": 8502.823
    },
    {
      "epoch": 1.4240506329113924,
      "grad_norm": 1.3033352981588253,
      "learning_rate": 1.9108050545671283e-06,
      "loss": 0.25452227592468263,
      "num_input_tokens_seen": 346359248,
      "step": 21150,
      "train_runtime": 40734.7105,
      "train_tokens_per_second": 8502.804
    },
    {
      "epoch": 1.4243872879073525,
      "grad_norm": 1.220490135666286,
      "learning_rate": 1.9087264325441507e-06,
      "loss": 0.2502012014389038,
      "num_input_tokens_seen": 346441168,
      "step": 21155,
      "train_runtime": 40744.3259,
      "train_tokens_per_second": 8502.808
    },
    {
      "epoch": 1.4247239429033127,
      "grad_norm": 0.9607112428021203,
      "learning_rate": 1.9066486749859686e-06,
      "loss": 0.2900082111358643,
      "num_input_tokens_seen": 346523088,
      "step": 21160,
      "train_runtime": 40754.1068,
      "train_tokens_per_second": 8502.777
    },
    {
      "epoch": 1.4250605978992728,
      "grad_norm": 1.0070835787589332,
      "learning_rate": 1.9045717824736209e-06,
      "loss": 0.2908333301544189,
      "num_input_tokens_seen": 346605008,
      "step": 21165,
      "train_runtime": 40763.6281,
      "train_tokens_per_second": 8502.801
    },
    {
      "epoch": 1.425397252895233,
      "grad_norm": 1.3366794298987563,
      "learning_rate": 1.9024957555879037e-06,
      "loss": 0.29026241302490235,
      "num_input_tokens_seen": 346686520,
      "step": 21170,
      "train_runtime": 40773.5936,
      "train_tokens_per_second": 8502.722
    },
    {
      "epoch": 1.425733907891193,
      "grad_norm": 1.4036008406066889,
      "learning_rate": 1.9004205949093695e-06,
      "loss": 0.28227312564849855,
      "num_input_tokens_seen": 346768440,
      "step": 21175,
      "train_runtime": 40783.2041,
      "train_tokens_per_second": 8502.727
    },
    {
      "epoch": 1.4260705628871533,
      "grad_norm": 1.0997537803608328,
      "learning_rate": 1.8983463010183355e-06,
      "loss": 0.28099989891052246,
      "num_input_tokens_seen": 346850360,
      "step": 21180,
      "train_runtime": 40792.7033,
      "train_tokens_per_second": 8502.755
    },
    {
      "epoch": 1.4264072178831133,
      "grad_norm": 1.1174079688303145,
      "learning_rate": 1.8962728744948644e-06,
      "loss": 0.29959964752197266,
      "num_input_tokens_seen": 346932280,
      "step": 21185,
      "train_runtime": 40802.2041,
      "train_tokens_per_second": 8502.783
    },
    {
      "epoch": 1.4267438728790736,
      "grad_norm": 1.2431158058163907,
      "learning_rate": 1.894200315918789e-06,
      "loss": 0.29094223976135253,
      "num_input_tokens_seen": 347014200,
      "step": 21190,
      "train_runtime": 40811.7038,
      "train_tokens_per_second": 8502.811
    },
    {
      "epoch": 1.4270805278750336,
      "grad_norm": 1.1171312640831568,
      "learning_rate": 1.8921286258696913e-06,
      "loss": 0.28380594253540037,
      "num_input_tokens_seen": 347095872,
      "step": 21195,
      "train_runtime": 40821.7159,
      "train_tokens_per_second": 8502.726
    },
    {
      "epoch": 1.4274171828709938,
      "grad_norm": 0.9070523515075717,
      "learning_rate": 1.890057804926913e-06,
      "loss": 0.28193728923797606,
      "num_input_tokens_seen": 347177792,
      "step": 21200,
      "train_runtime": 40831.2501,
      "train_tokens_per_second": 8502.747
    },
    {
      "epoch": 1.4277538378669539,
      "grad_norm": 1.096058587595634,
      "learning_rate": 1.887987853669553e-06,
      "loss": 0.26970372200012205,
      "num_input_tokens_seen": 347259712,
      "step": 21205,
      "train_runtime": 40840.8806,
      "train_tokens_per_second": 8502.748
    },
    {
      "epoch": 1.4280904928629141,
      "grad_norm": 1.4016346221475993,
      "learning_rate": 1.8859187726764654e-06,
      "loss": 0.2884895086288452,
      "num_input_tokens_seen": 347341632,
      "step": 21210,
      "train_runtime": 40850.3891,
      "train_tokens_per_second": 8502.774
    },
    {
      "epoch": 1.4284271478588741,
      "grad_norm": 0.9128978935796704,
      "learning_rate": 1.8838505625262638e-06,
      "loss": 0.2800004005432129,
      "num_input_tokens_seen": 347423552,
      "step": 21215,
      "train_runtime": 40860.0424,
      "train_tokens_per_second": 8502.77
    },
    {
      "epoch": 1.4287638028548344,
      "grad_norm": 1.0727088190772542,
      "learning_rate": 1.8817832237973154e-06,
      "loss": 0.28690874576568604,
      "num_input_tokens_seen": 347505472,
      "step": 21220,
      "train_runtime": 40869.9346,
      "train_tokens_per_second": 8502.717
    },
    {
      "epoch": 1.4291004578507944,
      "grad_norm": 1.2943188642941437,
      "learning_rate": 1.8797167570677439e-06,
      "loss": 0.24705922603607178,
      "num_input_tokens_seen": 347586912,
      "step": 21225,
      "train_runtime": 40879.905,
      "train_tokens_per_second": 8502.635
    },
    {
      "epoch": 1.4294371128467547,
      "grad_norm": 1.212306636337695,
      "learning_rate": 1.8776511629154342e-06,
      "loss": 0.3032437801361084,
      "num_input_tokens_seen": 347668832,
      "step": 21230,
      "train_runtime": 40889.6201,
      "train_tokens_per_second": 8502.618
    },
    {
      "epoch": 1.4297737678427147,
      "grad_norm": 0.9905164799385953,
      "learning_rate": 1.8755864419180176e-06,
      "loss": 0.2882194995880127,
      "num_input_tokens_seen": 347750752,
      "step": 21235,
      "train_runtime": 40899.2812,
      "train_tokens_per_second": 8502.613
    },
    {
      "epoch": 1.430110422838675,
      "grad_norm": 1.044964423645327,
      "learning_rate": 1.8735225946528906e-06,
      "loss": 0.27943053245544436,
      "num_input_tokens_seen": 347832672,
      "step": 21240,
      "train_runtime": 40908.9589,
      "train_tokens_per_second": 8502.604
    },
    {
      "epoch": 1.430447077834635,
      "grad_norm": 1.299452762741033,
      "learning_rate": 1.8714596216972008e-06,
      "loss": 0.2924210071563721,
      "num_input_tokens_seen": 347914592,
      "step": 21245,
      "train_runtime": 40918.4718,
      "train_tokens_per_second": 8502.629
    },
    {
      "epoch": 1.4307837328305952,
      "grad_norm": 1.128605180522319,
      "learning_rate": 1.8693975236278512e-06,
      "loss": 0.2697469711303711,
      "num_input_tokens_seen": 347996512,
      "step": 21250,
      "train_runtime": 40927.9809,
      "train_tokens_per_second": 8502.655
    },
    {
      "epoch": 1.4311203878265553,
      "grad_norm": 1.1913529229756588,
      "learning_rate": 1.8673363010215017e-06,
      "loss": 0.3148850917816162,
      "num_input_tokens_seen": 348078432,
      "step": 21255,
      "train_runtime": 40937.5858,
      "train_tokens_per_second": 8502.661
    },
    {
      "epoch": 1.4314570428225155,
      "grad_norm": 1.154263357286751,
      "learning_rate": 1.8652759544545661e-06,
      "loss": 0.31554527282714845,
      "num_input_tokens_seen": 348160352,
      "step": 21260,
      "train_runtime": 40947.295,
      "train_tokens_per_second": 8502.646
    },
    {
      "epoch": 1.4317936978184758,
      "grad_norm": 0.9980746605907357,
      "learning_rate": 1.863216484503214e-06,
      "loss": 0.310725736618042,
      "num_input_tokens_seen": 348241720,
      "step": 21265,
      "train_runtime": 40957.4509,
      "train_tokens_per_second": 8502.524
    },
    {
      "epoch": 1.4321303528144358,
      "grad_norm": 1.2720987784758266,
      "learning_rate": 1.8611578917433698e-06,
      "loss": 0.30761873722076416,
      "num_input_tokens_seen": 348323640,
      "step": 21270,
      "train_runtime": 40967.1083,
      "train_tokens_per_second": 8502.52
    },
    {
      "epoch": 1.4324670078103958,
      "grad_norm": 1.1006550126404868,
      "learning_rate": 1.85910017675071e-06,
      "loss": 0.29059224128723143,
      "num_input_tokens_seen": 348405560,
      "step": 21275,
      "train_runtime": 40976.6146,
      "train_tokens_per_second": 8502.546
    },
    {
      "epoch": 1.432803662806356,
      "grad_norm": 0.9714228682677487,
      "learning_rate": 1.8570433401006737e-06,
      "loss": 0.24787790775299073,
      "num_input_tokens_seen": 348487480,
      "step": 21280,
      "train_runtime": 40986.1287,
      "train_tokens_per_second": 8502.571
    },
    {
      "epoch": 1.4331403178023163,
      "grad_norm": 0.9933506121569569,
      "learning_rate": 1.8549873823684418e-06,
      "loss": 0.2726050853729248,
      "num_input_tokens_seen": 348569400,
      "step": 21285,
      "train_runtime": 40995.6396,
      "train_tokens_per_second": 8502.597
    },
    {
      "epoch": 1.4334769727982763,
      "grad_norm": 1.0608899268133805,
      "learning_rate": 1.8529323041289616e-06,
      "loss": 0.2936979055404663,
      "num_input_tokens_seen": 348651320,
      "step": 21290,
      "train_runtime": 41005.2687,
      "train_tokens_per_second": 8502.598
    },
    {
      "epoch": 1.4338136277942364,
      "grad_norm": 1.259182294459717,
      "learning_rate": 1.850878105956927e-06,
      "loss": 0.28909618854522706,
      "num_input_tokens_seen": 348733240,
      "step": 21295,
      "train_runtime": 41014.8919,
      "train_tokens_per_second": 8502.6
    },
    {
      "epoch": 1.4341502827901966,
      "grad_norm": 0.936476766346519,
      "learning_rate": 1.8488247884267885e-06,
      "loss": 0.2853731632232666,
      "num_input_tokens_seen": 348815160,
      "step": 21300,
      "train_runtime": 41024.4017,
      "train_tokens_per_second": 8502.626
    },
    {
      "epoch": 1.4344869377861569,
      "grad_norm": 1.328102988856949,
      "learning_rate": 1.8467723521127495e-06,
      "loss": 0.28882012367248533,
      "num_input_tokens_seen": 348897080,
      "step": 21305,
      "train_runtime": 41033.9124,
      "train_tokens_per_second": 8502.652
    },
    {
      "epoch": 1.434823592782117,
      "grad_norm": 1.1601565887340375,
      "learning_rate": 1.8447207975887671e-06,
      "loss": 0.27324662208557127,
      "num_input_tokens_seen": 348979000,
      "step": 21310,
      "train_runtime": 41043.4213,
      "train_tokens_per_second": 8502.678
    },
    {
      "epoch": 1.435160247778077,
      "grad_norm": 1.237337237955694,
      "learning_rate": 1.8426701254285522e-06,
      "loss": 0.2772764444351196,
      "num_input_tokens_seen": 349060920,
      "step": 21315,
      "train_runtime": 41053.1283,
      "train_tokens_per_second": 8502.663
    },
    {
      "epoch": 1.4354969027740372,
      "grad_norm": 1.3584426497638649,
      "learning_rate": 1.840620336205569e-06,
      "loss": 0.3235848665237427,
      "num_input_tokens_seen": 349142840,
      "step": 21320,
      "train_runtime": 41062.8343,
      "train_tokens_per_second": 8502.648
    },
    {
      "epoch": 1.4358335577699974,
      "grad_norm": 0.9556010044358614,
      "learning_rate": 1.8385714304930324e-06,
      "loss": 0.25226128101348877,
      "num_input_tokens_seen": 349224760,
      "step": 21325,
      "train_runtime": 41072.6417,
      "train_tokens_per_second": 8502.613
    },
    {
      "epoch": 1.4361702127659575,
      "grad_norm": 1.0064817782860225,
      "learning_rate": 1.8365234088639178e-06,
      "loss": 0.2911026954650879,
      "num_input_tokens_seen": 349306680,
      "step": 21330,
      "train_runtime": 41082.3983,
      "train_tokens_per_second": 8502.587
    },
    {
      "epoch": 1.4365068677619175,
      "grad_norm": 1.0549650199390996,
      "learning_rate": 1.834476271890941e-06,
      "loss": 0.2374704122543335,
      "num_input_tokens_seen": 349388600,
      "step": 21335,
      "train_runtime": 41091.9118,
      "train_tokens_per_second": 8502.612
    },
    {
      "epoch": 1.4368435227578777,
      "grad_norm": 1.3485963286661307,
      "learning_rate": 1.8324300201465823e-06,
      "loss": 0.2609537601470947,
      "num_input_tokens_seen": 349470520,
      "step": 21340,
      "train_runtime": 41101.4165,
      "train_tokens_per_second": 8502.639
    },
    {
      "epoch": 1.437180177753838,
      "grad_norm": 1.0934743308323869,
      "learning_rate": 1.8303846542030674e-06,
      "loss": 0.29326372146606444,
      "num_input_tokens_seen": 349552440,
      "step": 21345,
      "train_runtime": 41111.2233,
      "train_tokens_per_second": 8502.604
    },
    {
      "epoch": 1.437516832749798,
      "grad_norm": 0.9914767911556388,
      "learning_rate": 1.828340174632377e-06,
      "loss": 0.2747200489044189,
      "num_input_tokens_seen": 349634360,
      "step": 21350,
      "train_runtime": 41120.7554,
      "train_tokens_per_second": 8502.625
    },
    {
      "epoch": 1.437853487745758,
      "grad_norm": 1.133526384206998,
      "learning_rate": 1.8262965820062434e-06,
      "loss": 0.268083381652832,
      "num_input_tokens_seen": 349716280,
      "step": 21355,
      "train_runtime": 41130.3962,
      "train_tokens_per_second": 8502.624
    },
    {
      "epoch": 1.4381901427417183,
      "grad_norm": 0.9503857713240197,
      "learning_rate": 1.8242538768961499e-06,
      "loss": 0.26226134300231935,
      "num_input_tokens_seen": 349798200,
      "step": 21360,
      "train_runtime": 41139.9313,
      "train_tokens_per_second": 8502.644
    },
    {
      "epoch": 1.4385267977376786,
      "grad_norm": 1.339900670109416,
      "learning_rate": 1.8222120598733317e-06,
      "loss": 0.266936731338501,
      "num_input_tokens_seen": 349880120,
      "step": 21365,
      "train_runtime": 41149.5569,
      "train_tokens_per_second": 8502.646
    },
    {
      "epoch": 1.4388634527336386,
      "grad_norm": 1.0645139507148187,
      "learning_rate": 1.820171131508781e-06,
      "loss": 0.305302095413208,
      "num_input_tokens_seen": 349962040,
      "step": 21370,
      "train_runtime": 41159.1699,
      "train_tokens_per_second": 8502.651
    },
    {
      "epoch": 1.4392001077295986,
      "grad_norm": 1.1204105158013753,
      "learning_rate": 1.8181310923732304e-06,
      "loss": 0.27611346244812013,
      "num_input_tokens_seen": 350043960,
      "step": 21375,
      "train_runtime": 41168.9461,
      "train_tokens_per_second": 8502.621
    },
    {
      "epoch": 1.4395367627255589,
      "grad_norm": 0.9866048546850186,
      "learning_rate": 1.8160919430371764e-06,
      "loss": 0.26640374660491944,
      "num_input_tokens_seen": 350125880,
      "step": 21380,
      "train_runtime": 41178.4506,
      "train_tokens_per_second": 8502.648
    },
    {
      "epoch": 1.439873417721519,
      "grad_norm": 1.1837615742768344,
      "learning_rate": 1.814053684070855e-06,
      "loss": 0.281670069694519,
      "num_input_tokens_seen": 350207800,
      "step": 21385,
      "train_runtime": 41187.956,
      "train_tokens_per_second": 8502.675
    },
    {
      "epoch": 1.4402100727174791,
      "grad_norm": 1.0612851142365518,
      "learning_rate": 1.8120163160442633e-06,
      "loss": 0.2608105897903442,
      "num_input_tokens_seen": 350289720,
      "step": 21390,
      "train_runtime": 41198.0351,
      "train_tokens_per_second": 8502.583
    },
    {
      "epoch": 1.4405467277134392,
      "grad_norm": 1.17434097147536,
      "learning_rate": 1.8099798395271428e-06,
      "loss": 0.30204272270202637,
      "num_input_tokens_seen": 350371640,
      "step": 21395,
      "train_runtime": 41208.013,
      "train_tokens_per_second": 8502.512
    },
    {
      "epoch": 1.4408833827093994,
      "grad_norm": 1.2729909445712444,
      "learning_rate": 1.807944255088988e-06,
      "loss": 0.2573745012283325,
      "num_input_tokens_seen": 350453560,
      "step": 21400,
      "train_runtime": 41217.5211,
      "train_tokens_per_second": 8502.538
    },
    {
      "epoch": 1.4412200377053597,
      "grad_norm": 1.170365013245679,
      "learning_rate": 1.8059095632990437e-06,
      "loss": 0.262644362449646,
      "num_input_tokens_seen": 350535480,
      "step": 21405,
      "train_runtime": 41227.1521,
      "train_tokens_per_second": 8502.539
    },
    {
      "epoch": 1.4415566927013197,
      "grad_norm": 1.3459007282723985,
      "learning_rate": 1.8038757647263045e-06,
      "loss": 0.30420923233032227,
      "num_input_tokens_seen": 350617400,
      "step": 21410,
      "train_runtime": 41236.8271,
      "train_tokens_per_second": 8502.531
    },
    {
      "epoch": 1.4418933476972797,
      "grad_norm": 1.088463007154926,
      "learning_rate": 1.8018428599395143e-06,
      "loss": 0.28276500701904295,
      "num_input_tokens_seen": 350698656,
      "step": 21415,
      "train_runtime": 41246.7002,
      "train_tokens_per_second": 8502.466
    },
    {
      "epoch": 1.44223000269324,
      "grad_norm": 1.119925706808944,
      "learning_rate": 1.7998108495071743e-06,
      "loss": 0.2805616855621338,
      "num_input_tokens_seen": 350780576,
      "step": 21420,
      "train_runtime": 41256.3564,
      "train_tokens_per_second": 8502.461
    },
    {
      "epoch": 1.4425666576892002,
      "grad_norm": 1.0608958038918925,
      "learning_rate": 1.797779733997521e-06,
      "loss": 0.27525150775909424,
      "num_input_tokens_seen": 350862496,
      "step": 21425,
      "train_runtime": 41266.1509,
      "train_tokens_per_second": 8502.428
    },
    {
      "epoch": 1.4429033126851603,
      "grad_norm": 1.2132427673824375,
      "learning_rate": 1.7957495139785585e-06,
      "loss": 0.2842967987060547,
      "num_input_tokens_seen": 350944416,
      "step": 21430,
      "train_runtime": 41275.7025,
      "train_tokens_per_second": 8502.446
    },
    {
      "epoch": 1.4432399676811203,
      "grad_norm": 1.2113235296993325,
      "learning_rate": 1.7937201900180228e-06,
      "loss": 0.2967234134674072,
      "num_input_tokens_seen": 351026336,
      "step": 21435,
      "train_runtime": 41285.3509,
      "train_tokens_per_second": 8502.443
    },
    {
      "epoch": 1.4435766226770805,
      "grad_norm": 1.315302668469846,
      "learning_rate": 1.7916917626834134e-06,
      "loss": 0.30446538925170896,
      "num_input_tokens_seen": 351108256,
      "step": 21440,
      "train_runtime": 41294.9768,
      "train_tokens_per_second": 8502.445
    },
    {
      "epoch": 1.4439132776730408,
      "grad_norm": 1.2271832581324629,
      "learning_rate": 1.7896642325419722e-06,
      "loss": 0.2923828840255737,
      "num_input_tokens_seen": 351190176,
      "step": 21445,
      "train_runtime": 41304.4949,
      "train_tokens_per_second": 8502.469
    },
    {
      "epoch": 1.4442499326690008,
      "grad_norm": 1.2475810535728642,
      "learning_rate": 1.7876376001606915e-06,
      "loss": 0.28530044555664064,
      "num_input_tokens_seen": 351271904,
      "step": 21450,
      "train_runtime": 41314.5933,
      "train_tokens_per_second": 8502.369
    },
    {
      "epoch": 1.4445865876649608,
      "grad_norm": 1.0309192713768183,
      "learning_rate": 1.7856118661063128e-06,
      "loss": 0.26799807548522947,
      "num_input_tokens_seen": 351353824,
      "step": 21455,
      "train_runtime": 41324.446,
      "train_tokens_per_second": 8502.324
    },
    {
      "epoch": 1.444923242660921,
      "grad_norm": 1.1686131885925055,
      "learning_rate": 1.7835870309453251e-06,
      "loss": 0.30089716911315917,
      "num_input_tokens_seen": 351435744,
      "step": 21460,
      "train_runtime": 41334.0499,
      "train_tokens_per_second": 8502.33
    },
    {
      "epoch": 1.4452598976568813,
      "grad_norm": 1.2216276796978862,
      "learning_rate": 1.7815630952439667e-06,
      "loss": 0.2982743740081787,
      "num_input_tokens_seen": 351517664,
      "step": 21465,
      "train_runtime": 41343.5558,
      "train_tokens_per_second": 8502.357
    },
    {
      "epoch": 1.4455965526528414,
      "grad_norm": 0.940445160699712,
      "learning_rate": 1.7795400595682289e-06,
      "loss": 0.2787323474884033,
      "num_input_tokens_seen": 351599584,
      "step": 21470,
      "train_runtime": 41353.0627,
      "train_tokens_per_second": 8502.383
    },
    {
      "epoch": 1.4459332076488014,
      "grad_norm": 1.154648552012513,
      "learning_rate": 1.7775179244838408e-06,
      "loss": 0.2992808103561401,
      "num_input_tokens_seen": 351681504,
      "step": 21475,
      "train_runtime": 41362.5891,
      "train_tokens_per_second": 8502.405
    },
    {
      "epoch": 1.4462698626447617,
      "grad_norm": 0.9575326971130665,
      "learning_rate": 1.7754966905562925e-06,
      "loss": 0.26952669620513914,
      "num_input_tokens_seen": 351763424,
      "step": 21480,
      "train_runtime": 41372.1633,
      "train_tokens_per_second": 8502.418
    },
    {
      "epoch": 1.446606517640722,
      "grad_norm": 1.3539246180426268,
      "learning_rate": 1.7734763583508098e-06,
      "loss": 0.29204320907592773,
      "num_input_tokens_seen": 351845344,
      "step": 21485,
      "train_runtime": 41381.7062,
      "train_tokens_per_second": 8502.437
    },
    {
      "epoch": 1.446943172636682,
      "grad_norm": 1.3053456391546985,
      "learning_rate": 1.7714569284323757e-06,
      "loss": 0.30841612815856934,
      "num_input_tokens_seen": 351927264,
      "step": 21490,
      "train_runtime": 41391.3077,
      "train_tokens_per_second": 8502.444
    },
    {
      "epoch": 1.447279827632642,
      "grad_norm": 1.1281648295174207,
      "learning_rate": 1.7694384013657173e-06,
      "loss": 0.30423603057861326,
      "num_input_tokens_seen": 352009184,
      "step": 21495,
      "train_runtime": 41400.8696,
      "train_tokens_per_second": 8502.459
    },
    {
      "epoch": 1.4476164826286022,
      "grad_norm": 1.2754411523382045,
      "learning_rate": 1.767420777715308e-06,
      "loss": 0.3178918123245239,
      "num_input_tokens_seen": 352091104,
      "step": 21500,
      "train_runtime": 41410.3893,
      "train_tokens_per_second": 8502.482
    },
    {
      "epoch": 1.4479531376245625,
      "grad_norm": 1.1881120404629506,
      "learning_rate": 1.7654040580453702e-06,
      "loss": 0.2827815294265747,
      "num_input_tokens_seen": 352173024,
      "step": 21505,
      "train_runtime": 41420.0223,
      "train_tokens_per_second": 8502.483
    },
    {
      "epoch": 1.4482897926205225,
      "grad_norm": 1.1078183438181148,
      "learning_rate": 1.7633882429198734e-06,
      "loss": 0.3010611295700073,
      "num_input_tokens_seen": 352254944,
      "step": 21510,
      "train_runtime": 41429.7021,
      "train_tokens_per_second": 8502.474
    },
    {
      "epoch": 1.4486264476164825,
      "grad_norm": 1.0174854792086492,
      "learning_rate": 1.7613733329025318e-06,
      "loss": 0.2876286029815674,
      "num_input_tokens_seen": 352336864,
      "step": 21515,
      "train_runtime": 41439.3404,
      "train_tokens_per_second": 8502.473
    },
    {
      "epoch": 1.4489631026124428,
      "grad_norm": 1.091269407645099,
      "learning_rate": 1.759359328556814e-06,
      "loss": 0.28769640922546386,
      "num_input_tokens_seen": 352418784,
      "step": 21520,
      "train_runtime": 41448.854,
      "train_tokens_per_second": 8502.498
    },
    {
      "epoch": 1.449299757608403,
      "grad_norm": 1.0933036223707964,
      "learning_rate": 1.7573462304459227e-06,
      "loss": 0.290163516998291,
      "num_input_tokens_seen": 352500704,
      "step": 21525,
      "train_runtime": 41458.3849,
      "train_tokens_per_second": 8502.519
    },
    {
      "epoch": 1.449636412604363,
      "grad_norm": 1.1954377156775493,
      "learning_rate": 1.7553340391328216e-06,
      "loss": 0.30649919509887696,
      "num_input_tokens_seen": 352582624,
      "step": 21530,
      "train_runtime": 41467.9036,
      "train_tokens_per_second": 8502.543
    },
    {
      "epoch": 1.449973067600323,
      "grad_norm": 1.0501895602622489,
      "learning_rate": 1.7533227551802068e-06,
      "loss": 0.2755756855010986,
      "num_input_tokens_seen": 352664544,
      "step": 21535,
      "train_runtime": 41477.4242,
      "train_tokens_per_second": 8502.566
    },
    {
      "epoch": 1.4503097225962833,
      "grad_norm": 1.1881318839092503,
      "learning_rate": 1.7513123791505321e-06,
      "loss": 0.2933255672454834,
      "num_input_tokens_seen": 352746464,
      "step": 21540,
      "train_runtime": 41486.9303,
      "train_tokens_per_second": 8502.593
    },
    {
      "epoch": 1.4506463775922436,
      "grad_norm": 1.1515113594211743,
      "learning_rate": 1.7493029116059913e-06,
      "loss": 0.30164189338684083,
      "num_input_tokens_seen": 352828384,
      "step": 21545,
      "train_runtime": 41496.4403,
      "train_tokens_per_second": 8502.618
    },
    {
      "epoch": 1.4509830325882036,
      "grad_norm": 1.1458487499878933,
      "learning_rate": 1.7472943531085252e-06,
      "loss": 0.29973411560058594,
      "num_input_tokens_seen": 352910304,
      "step": 21550,
      "train_runtime": 41506.0511,
      "train_tokens_per_second": 8502.623
    },
    {
      "epoch": 1.4513196875841636,
      "grad_norm": 1.2235519553407372,
      "learning_rate": 1.7452867042198212e-06,
      "loss": 0.2876373291015625,
      "num_input_tokens_seen": 352992224,
      "step": 21555,
      "train_runtime": 41515.5869,
      "train_tokens_per_second": 8502.643
    },
    {
      "epoch": 1.4516563425801239,
      "grad_norm": 1.0160799995604497,
      "learning_rate": 1.7432799655013122e-06,
      "loss": 0.28609814643859866,
      "num_input_tokens_seen": 353074144,
      "step": 21560,
      "train_runtime": 41525.0901,
      "train_tokens_per_second": 8502.67
    },
    {
      "epoch": 1.4519929975760841,
      "grad_norm": 0.9660820724588104,
      "learning_rate": 1.741274137514174e-06,
      "loss": 0.2958077907562256,
      "num_input_tokens_seen": 353156064,
      "step": 21565,
      "train_runtime": 41534.7154,
      "train_tokens_per_second": 8502.672
    },
    {
      "epoch": 1.4523296525720442,
      "grad_norm": 1.0558766586154111,
      "learning_rate": 1.7392692208193351e-06,
      "loss": 0.26708106994628905,
      "num_input_tokens_seen": 353237984,
      "step": 21570,
      "train_runtime": 41544.2296,
      "train_tokens_per_second": 8502.697
    },
    {
      "epoch": 1.4526663075680042,
      "grad_norm": 0.9758061649967609,
      "learning_rate": 1.7372652159774578e-06,
      "loss": 0.23765718936920166,
      "num_input_tokens_seen": 353319904,
      "step": 21575,
      "train_runtime": 41553.7392,
      "train_tokens_per_second": 8502.722
    },
    {
      "epoch": 1.4530029625639644,
      "grad_norm": 1.2200985514633393,
      "learning_rate": 1.7352621235489625e-06,
      "loss": 0.2883406639099121,
      "num_input_tokens_seen": 353401824,
      "step": 21580,
      "train_runtime": 41563.2571,
      "train_tokens_per_second": 8502.746
    },
    {
      "epoch": 1.4533396175599247,
      "grad_norm": 1.3038563365704916,
      "learning_rate": 1.733259944094e-06,
      "loss": 0.2930554151535034,
      "num_input_tokens_seen": 353483744,
      "step": 21585,
      "train_runtime": 41572.7838,
      "train_tokens_per_second": 8502.768
    },
    {
      "epoch": 1.4536762725558847,
      "grad_norm": 1.1322645960269873,
      "learning_rate": 1.7312586781724789e-06,
      "loss": 0.2922735929489136,
      "num_input_tokens_seen": 353565664,
      "step": 21590,
      "train_runtime": 41582.2899,
      "train_tokens_per_second": 8502.794
    },
    {
      "epoch": 1.4540129275518447,
      "grad_norm": 1.0528593095636969,
      "learning_rate": 1.7292583263440454e-06,
      "loss": 0.29834661483764646,
      "num_input_tokens_seen": 353647584,
      "step": 21595,
      "train_runtime": 41591.9816,
      "train_tokens_per_second": 8502.783
    },
    {
      "epoch": 1.454349582547805,
      "grad_norm": 1.1986580743319342,
      "learning_rate": 1.7272588891680914e-06,
      "loss": 0.2700026273727417,
      "num_input_tokens_seen": 353729504,
      "step": 21600,
      "train_runtime": 41601.6449,
      "train_tokens_per_second": 8502.777
    },
    {
      "epoch": 1.4546862375437652,
      "grad_norm": 1.2738452501774136,
      "learning_rate": 1.7252603672037527e-06,
      "loss": 0.2709773540496826,
      "num_input_tokens_seen": 353811424,
      "step": 21605,
      "train_runtime": 41611.3713,
      "train_tokens_per_second": 8502.758
    },
    {
      "epoch": 1.4550228925397253,
      "grad_norm": 0.9714324158112965,
      "learning_rate": 1.72326276100991e-06,
      "loss": 0.2531388998031616,
      "num_input_tokens_seen": 353893344,
      "step": 21610,
      "train_runtime": 41621.043,
      "train_tokens_per_second": 8502.75
    },
    {
      "epoch": 1.4553595475356853,
      "grad_norm": 1.0760848849597338,
      "learning_rate": 1.721266071145186e-06,
      "loss": 0.2927816867828369,
      "num_input_tokens_seen": 353975264,
      "step": 21615,
      "train_runtime": 41630.5526,
      "train_tokens_per_second": 8502.776
    },
    {
      "epoch": 1.4556962025316456,
      "grad_norm": 1.03140061028391,
      "learning_rate": 1.7192702981679531e-06,
      "loss": 0.2983665943145752,
      "num_input_tokens_seen": 354057184,
      "step": 21620,
      "train_runtime": 41640.0861,
      "train_tokens_per_second": 8502.797
    },
    {
      "epoch": 1.4560328575276058,
      "grad_norm": 1.0466268051295025,
      "learning_rate": 1.717275442636317e-06,
      "loss": 0.24576122760772706,
      "num_input_tokens_seen": 354139104,
      "step": 21625,
      "train_runtime": 41649.7445,
      "train_tokens_per_second": 8502.792
    },
    {
      "epoch": 1.4563695125235658,
      "grad_norm": 1.24351436563361,
      "learning_rate": 1.7152815051081384e-06,
      "loss": 0.31062285900115966,
      "num_input_tokens_seen": 354221024,
      "step": 21630,
      "train_runtime": 41659.3047,
      "train_tokens_per_second": 8502.807
    },
    {
      "epoch": 1.4567061675195259,
      "grad_norm": 1.0627171513249698,
      "learning_rate": 1.7132884861410093e-06,
      "loss": 0.2888061761856079,
      "num_input_tokens_seen": 354302944,
      "step": 21635,
      "train_runtime": 41668.8879,
      "train_tokens_per_second": 8502.817
    },
    {
      "epoch": 1.4570428225154861,
      "grad_norm": 1.0358144134738323,
      "learning_rate": 1.7112963862922766e-06,
      "loss": 0.2651897668838501,
      "num_input_tokens_seen": 354384864,
      "step": 21640,
      "train_runtime": 41678.4376,
      "train_tokens_per_second": 8502.835
    },
    {
      "epoch": 1.4573794775114464,
      "grad_norm": 1.1329414524446382,
      "learning_rate": 1.7093052061190224e-06,
      "loss": 0.30678553581237794,
      "num_input_tokens_seen": 354466784,
      "step": 21645,
      "train_runtime": 41688.0816,
      "train_tokens_per_second": 8502.833
    },
    {
      "epoch": 1.4577161325074064,
      "grad_norm": 1.157487898877406,
      "learning_rate": 1.7073149461780742e-06,
      "loss": 0.293152379989624,
      "num_input_tokens_seen": 354548704,
      "step": 21650,
      "train_runtime": 41697.6463,
      "train_tokens_per_second": 8502.847
    },
    {
      "epoch": 1.4580527875033664,
      "grad_norm": 1.1312840604956367,
      "learning_rate": 1.705325607026001e-06,
      "loss": 0.296121883392334,
      "num_input_tokens_seen": 354630624,
      "step": 21655,
      "train_runtime": 41707.2011,
      "train_tokens_per_second": 8502.863
    },
    {
      "epoch": 1.4583894424993267,
      "grad_norm": 1.1366144463107635,
      "learning_rate": 1.7033371892191164e-06,
      "loss": 0.2634941577911377,
      "num_input_tokens_seen": 354712544,
      "step": 21660,
      "train_runtime": 41716.7104,
      "train_tokens_per_second": 8502.889
    },
    {
      "epoch": 1.458726097495287,
      "grad_norm": 1.2645867851995964,
      "learning_rate": 1.7013496933134726e-06,
      "loss": 0.29679107666015625,
      "num_input_tokens_seen": 354794464,
      "step": 21665,
      "train_runtime": 41726.3032,
      "train_tokens_per_second": 8502.897
    },
    {
      "epoch": 1.459062752491247,
      "grad_norm": 1.2083287851542626,
      "learning_rate": 1.6993631198648724e-06,
      "loss": 0.32359728813171384,
      "num_input_tokens_seen": 354876384,
      "step": 21670,
      "train_runtime": 41735.8177,
      "train_tokens_per_second": 8502.922
    },
    {
      "epoch": 1.459399407487207,
      "grad_norm": 1.29952583056424,
      "learning_rate": 1.697377469428847e-06,
      "loss": 0.28023695945739746,
      "num_input_tokens_seen": 354958304,
      "step": 21675,
      "train_runtime": 41745.3367,
      "train_tokens_per_second": 8502.945
    },
    {
      "epoch": 1.4597360624831672,
      "grad_norm": 1.1178970425252572,
      "learning_rate": 1.6953927425606842e-06,
      "loss": 0.293774938583374,
      "num_input_tokens_seen": 355040224,
      "step": 21680,
      "train_runtime": 41756.2389,
      "train_tokens_per_second": 8502.687
    },
    {
      "epoch": 1.4600727174791275,
      "grad_norm": 0.8490468697936622,
      "learning_rate": 1.6934089398154002e-06,
      "loss": 0.28152012825012207,
      "num_input_tokens_seen": 355122144,
      "step": 21685,
      "train_runtime": 41765.9289,
      "train_tokens_per_second": 8502.676
    },
    {
      "epoch": 1.4604093724750875,
      "grad_norm": 1.0551564100876276,
      "learning_rate": 1.6914260617477634e-06,
      "loss": 0.2740155220031738,
      "num_input_tokens_seen": 355204064,
      "step": 21690,
      "train_runtime": 41775.4506,
      "train_tokens_per_second": 8502.699
    },
    {
      "epoch": 1.4607460274710475,
      "grad_norm": 1.169363307613112,
      "learning_rate": 1.6894441089122788e-06,
      "loss": 0.2587131977081299,
      "num_input_tokens_seen": 355285984,
      "step": 21695,
      "train_runtime": 41784.957,
      "train_tokens_per_second": 8502.725
    },
    {
      "epoch": 1.4610826824670078,
      "grad_norm": 1.2129554572864572,
      "learning_rate": 1.6874630818631926e-06,
      "loss": 0.2967516899108887,
      "num_input_tokens_seen": 355367904,
      "step": 21700,
      "train_runtime": 41794.4817,
      "train_tokens_per_second": 8502.747
    },
    {
      "epoch": 1.461419337462968,
      "grad_norm": 0.9766516619285142,
      "learning_rate": 1.685482981154492e-06,
      "loss": 0.29266066551208497,
      "num_input_tokens_seen": 355449824,
      "step": 21705,
      "train_runtime": 41804.2049,
      "train_tokens_per_second": 8502.729
    },
    {
      "epoch": 1.461755992458928,
      "grad_norm": 1.1463448446269318,
      "learning_rate": 1.683503807339907e-06,
      "loss": 0.2547149658203125,
      "num_input_tokens_seen": 355531744,
      "step": 21710,
      "train_runtime": 41813.7926,
      "train_tokens_per_second": 8502.738
    },
    {
      "epoch": 1.462092647454888,
      "grad_norm": 1.041252269315621,
      "learning_rate": 1.6815255609729047e-06,
      "loss": 0.30172457695007326,
      "num_input_tokens_seen": 355613664,
      "step": 21715,
      "train_runtime": 41823.3065,
      "train_tokens_per_second": 8502.763
    },
    {
      "epoch": 1.4624293024508483,
      "grad_norm": 1.0308220919457516,
      "learning_rate": 1.6795482426067012e-06,
      "loss": 0.28999943733215333,
      "num_input_tokens_seen": 355695584,
      "step": 21720,
      "train_runtime": 41832.8166,
      "train_tokens_per_second": 8502.788
    },
    {
      "epoch": 1.4627659574468086,
      "grad_norm": 1.1641622513858516,
      "learning_rate": 1.6775718527942403e-06,
      "loss": 0.2746416091918945,
      "num_input_tokens_seen": 355777336,
      "step": 21725,
      "train_runtime": 41842.8391,
      "train_tokens_per_second": 8502.705
    },
    {
      "epoch": 1.4631026124427686,
      "grad_norm": 1.3526035803296634,
      "learning_rate": 1.675596392088219e-06,
      "loss": 0.2848746538162231,
      "num_input_tokens_seen": 355859256,
      "step": 21730,
      "train_runtime": 41852.4721,
      "train_tokens_per_second": 8502.706
    },
    {
      "epoch": 1.4634392674387287,
      "grad_norm": 1.144075913885207,
      "learning_rate": 1.6736218610410633e-06,
      "loss": 0.24697837829589844,
      "num_input_tokens_seen": 355941176,
      "step": 21735,
      "train_runtime": 41862.1482,
      "train_tokens_per_second": 8502.697
    },
    {
      "epoch": 1.463775922434689,
      "grad_norm": 1.1367883451583836,
      "learning_rate": 1.6716482602049483e-06,
      "loss": 0.2497769594192505,
      "num_input_tokens_seen": 356022800,
      "step": 21740,
      "train_runtime": 41872.1946,
      "train_tokens_per_second": 8502.607
    },
    {
      "epoch": 1.4641125774306492,
      "grad_norm": 1.3149507333170993,
      "learning_rate": 1.669675590131784e-06,
      "loss": 0.2974428415298462,
      "num_input_tokens_seen": 356104720,
      "step": 21745,
      "train_runtime": 41881.7175,
      "train_tokens_per_second": 8502.629
    },
    {
      "epoch": 1.4644492324266092,
      "grad_norm": 1.1186924461925694,
      "learning_rate": 1.6677038513732214e-06,
      "loss": 0.2924819469451904,
      "num_input_tokens_seen": 356186640,
      "step": 21750,
      "train_runtime": 41891.4166,
      "train_tokens_per_second": 8502.616
    },
    {
      "epoch": 1.4647858874225694,
      "grad_norm": 1.2781755408183828,
      "learning_rate": 1.6657330444806508e-06,
      "loss": 0.3112342119216919,
      "num_input_tokens_seen": 356268560,
      "step": 21755,
      "train_runtime": 41901.0576,
      "train_tokens_per_second": 8502.615
    },
    {
      "epoch": 1.4651225424185295,
      "grad_norm": 1.083852429334929,
      "learning_rate": 1.6637631700052021e-06,
      "loss": 0.31460549831390383,
      "num_input_tokens_seen": 356350480,
      "step": 21760,
      "train_runtime": 41910.7537,
      "train_tokens_per_second": 8502.603
    },
    {
      "epoch": 1.4654591974144897,
      "grad_norm": 1.0600052550052546,
      "learning_rate": 1.6617942284977428e-06,
      "loss": 0.26569571495056155,
      "num_input_tokens_seen": 356432400,
      "step": 21765,
      "train_runtime": 41920.4804,
      "train_tokens_per_second": 8502.584
    },
    {
      "epoch": 1.4657958524104497,
      "grad_norm": 1.2898810556946934,
      "learning_rate": 1.6598262205088862e-06,
      "loss": 0.3063030242919922,
      "num_input_tokens_seen": 356514320,
      "step": 21770,
      "train_runtime": 41930.1407,
      "train_tokens_per_second": 8502.579
    },
    {
      "epoch": 1.46613250740641,
      "grad_norm": 1.1598431007878416,
      "learning_rate": 1.6578591465889722e-06,
      "loss": 0.28462510108947753,
      "num_input_tokens_seen": 356596240,
      "step": 21775,
      "train_runtime": 41939.854,
      "train_tokens_per_second": 8502.563
    },
    {
      "epoch": 1.46646916240237,
      "grad_norm": 1.1598118171246534,
      "learning_rate": 1.6558930072880942e-06,
      "loss": 0.3454420566558838,
      "num_input_tokens_seen": 356678160,
      "step": 21780,
      "train_runtime": 41949.5058,
      "train_tokens_per_second": 8502.559
    },
    {
      "epoch": 1.4668058173983303,
      "grad_norm": 1.1688932745998422,
      "learning_rate": 1.6539278031560695e-06,
      "loss": 0.3000066041946411,
      "num_input_tokens_seen": 356760080,
      "step": 21785,
      "train_runtime": 41959.1746,
      "train_tokens_per_second": 8502.552
    },
    {
      "epoch": 1.4671424723942903,
      "grad_norm": 1.187875101954986,
      "learning_rate": 1.6519635347424672e-06,
      "loss": 0.2933598756790161,
      "num_input_tokens_seen": 356841336,
      "step": 21790,
      "train_runtime": 41969.0487,
      "train_tokens_per_second": 8502.488
    },
    {
      "epoch": 1.4674791273902505,
      "grad_norm": 1.1701773305527656,
      "learning_rate": 1.650000202596586e-06,
      "loss": 0.31423425674438477,
      "num_input_tokens_seen": 356923256,
      "step": 21795,
      "train_runtime": 41978.5821,
      "train_tokens_per_second": 8502.509
    },
    {
      "epoch": 1.4678157823862106,
      "grad_norm": 1.0480941306317795,
      "learning_rate": 1.6480378072674659e-06,
      "loss": 0.27893571853637694,
      "num_input_tokens_seen": 357005176,
      "step": 21800,
      "train_runtime": 41988.2346,
      "train_tokens_per_second": 8502.505
    },
    {
      "epoch": 1.4681524373821708,
      "grad_norm": 1.055796110148622,
      "learning_rate": 1.646076349303884e-06,
      "loss": 0.2814739227294922,
      "num_input_tokens_seen": 357087096,
      "step": 21805,
      "train_runtime": 41997.9465,
      "train_tokens_per_second": 8502.489
    },
    {
      "epoch": 1.4684890923781309,
      "grad_norm": 1.0643401967137072,
      "learning_rate": 1.6441158292543557e-06,
      "loss": 0.28167703151702883,
      "num_input_tokens_seen": 357169016,
      "step": 21810,
      "train_runtime": 42007.6602,
      "train_tokens_per_second": 8502.473
    },
    {
      "epoch": 1.468825747374091,
      "grad_norm": 1.149427023477955,
      "learning_rate": 1.6421562476671339e-06,
      "loss": 0.314646577835083,
      "num_input_tokens_seen": 357250936,
      "step": 21815,
      "train_runtime": 42017.342,
      "train_tokens_per_second": 8502.464
    },
    {
      "epoch": 1.4691624023700511,
      "grad_norm": 1.032253086874157,
      "learning_rate": 1.640197605090213e-06,
      "loss": 0.2818714141845703,
      "num_input_tokens_seen": 357332856,
      "step": 21820,
      "train_runtime": 42027.0179,
      "train_tokens_per_second": 8502.456
    },
    {
      "epoch": 1.4694990573660114,
      "grad_norm": 1.1050807537127376,
      "learning_rate": 1.6382399020713146e-06,
      "loss": 0.27606945037841796,
      "num_input_tokens_seen": 357414776,
      "step": 21825,
      "train_runtime": 42036.6797,
      "train_tokens_per_second": 8502.45
    },
    {
      "epoch": 1.4698357123619714,
      "grad_norm": 1.1248304357076981,
      "learning_rate": 1.6362831391579115e-06,
      "loss": 0.30437707901000977,
      "num_input_tokens_seen": 357496696,
      "step": 21830,
      "train_runtime": 42046.3414,
      "train_tokens_per_second": 8502.445
    },
    {
      "epoch": 1.4701723673579317,
      "grad_norm": 1.0796163516609754,
      "learning_rate": 1.634327316897199e-06,
      "loss": 0.24180641174316406,
      "num_input_tokens_seen": 357578616,
      "step": 21835,
      "train_runtime": 42056.0071,
      "train_tokens_per_second": 8502.439
    },
    {
      "epoch": 1.4705090223538917,
      "grad_norm": 1.0765883442004307,
      "learning_rate": 1.6323724358361216e-06,
      "loss": 0.26987504959106445,
      "num_input_tokens_seen": 357660536,
      "step": 21840,
      "train_runtime": 42065.5131,
      "train_tokens_per_second": 8502.465
    },
    {
      "epoch": 1.470845677349852,
      "grad_norm": 1.0886756534024884,
      "learning_rate": 1.6304184965213537e-06,
      "loss": 0.2683112621307373,
      "num_input_tokens_seen": 357742456,
      "step": 21845,
      "train_runtime": 42075.1601,
      "train_tokens_per_second": 8502.462
    },
    {
      "epoch": 1.471182332345812,
      "grad_norm": 1.2384091762582872,
      "learning_rate": 1.6284654994993088e-06,
      "loss": 0.2755014896392822,
      "num_input_tokens_seen": 357824376,
      "step": 21850,
      "train_runtime": 42084.695,
      "train_tokens_per_second": 8502.482
    },
    {
      "epoch": 1.4715189873417722,
      "grad_norm": 1.1079995850768836,
      "learning_rate": 1.6265134453161358e-06,
      "loss": 0.2803352355957031,
      "num_input_tokens_seen": 357906296,
      "step": 21855,
      "train_runtime": 42094.253,
      "train_tokens_per_second": 8502.498
    },
    {
      "epoch": 1.4718556423377323,
      "grad_norm": 0.9128045587261379,
      "learning_rate": 1.6245623345177203e-06,
      "loss": 0.2664696216583252,
      "num_input_tokens_seen": 357988216,
      "step": 21860,
      "train_runtime": 42103.8507,
      "train_tokens_per_second": 8502.505
    },
    {
      "epoch": 1.4721922973336925,
      "grad_norm": 1.0337699248829733,
      "learning_rate": 1.622612167649683e-06,
      "loss": 0.270680832862854,
      "num_input_tokens_seen": 358070136,
      "step": 21865,
      "train_runtime": 42113.3455,
      "train_tokens_per_second": 8502.534
    },
    {
      "epoch": 1.4725289523296525,
      "grad_norm": 1.0319324384290818,
      "learning_rate": 1.6206629452573868e-06,
      "loss": 0.26773581504821775,
      "num_input_tokens_seen": 358152056,
      "step": 21870,
      "train_runtime": 42122.9548,
      "train_tokens_per_second": 8502.539
    },
    {
      "epoch": 1.4728656073256128,
      "grad_norm": 1.204010146949133,
      "learning_rate": 1.618714667885919e-06,
      "loss": 0.2778221607208252,
      "num_input_tokens_seen": 358233976,
      "step": 21875,
      "train_runtime": 42132.4585,
      "train_tokens_per_second": 8502.565
    },
    {
      "epoch": 1.4732022623215728,
      "grad_norm": 1.123538385662638,
      "learning_rate": 1.6167673360801151e-06,
      "loss": 0.28377370834350585,
      "num_input_tokens_seen": 358315896,
      "step": 21880,
      "train_runtime": 42142.0901,
      "train_tokens_per_second": 8502.566
    },
    {
      "epoch": 1.473538917317533,
      "grad_norm": 1.1816618778790207,
      "learning_rate": 1.6148209503845336e-06,
      "loss": 0.2953511953353882,
      "num_input_tokens_seen": 358397816,
      "step": 21885,
      "train_runtime": 42151.5963,
      "train_tokens_per_second": 8502.592
    },
    {
      "epoch": 1.473875572313493,
      "grad_norm": 1.0729476882957012,
      "learning_rate": 1.6128755113434808e-06,
      "loss": 0.26287801265716554,
      "num_input_tokens_seen": 358479736,
      "step": 21890,
      "train_runtime": 42161.2169,
      "train_tokens_per_second": 8502.595
    },
    {
      "epoch": 1.4742122273094533,
      "grad_norm": 1.1554181245356527,
      "learning_rate": 1.6109310195009908e-06,
      "loss": 0.3114306926727295,
      "num_input_tokens_seen": 358561656,
      "step": 21895,
      "train_runtime": 42170.7979,
      "train_tokens_per_second": 8502.605
    },
    {
      "epoch": 1.4745488823054134,
      "grad_norm": 1.0746132610558412,
      "learning_rate": 1.6089874754008323e-06,
      "loss": 0.25333638191223146,
      "num_input_tokens_seen": 358643576,
      "step": 21900,
      "train_runtime": 42180.2969,
      "train_tokens_per_second": 8502.633
    },
    {
      "epoch": 1.4748855373013736,
      "grad_norm": 1.0079995737604448,
      "learning_rate": 1.6070448795865162e-06,
      "loss": 0.2871129274368286,
      "num_input_tokens_seen": 358725496,
      "step": 21905,
      "train_runtime": 42189.907,
      "train_tokens_per_second": 8502.638
    },
    {
      "epoch": 1.4752221922973336,
      "grad_norm": 0.9866940360067249,
      "learning_rate": 1.605103232601277e-06,
      "loss": 0.27859318256378174,
      "num_input_tokens_seen": 358807416,
      "step": 21910,
      "train_runtime": 42199.4625,
      "train_tokens_per_second": 8502.654
    },
    {
      "epoch": 1.475558847293294,
      "grad_norm": 0.8887146506935207,
      "learning_rate": 1.603162534988096e-06,
      "loss": 0.24658193588256835,
      "num_input_tokens_seen": 358889336,
      "step": 21915,
      "train_runtime": 42209.0669,
      "train_tokens_per_second": 8502.66
    },
    {
      "epoch": 1.475895502289254,
      "grad_norm": 1.142435711891372,
      "learning_rate": 1.6012227872896791e-06,
      "loss": 0.27772161960601804,
      "num_input_tokens_seen": 358971256,
      "step": 21920,
      "train_runtime": 42218.5941,
      "train_tokens_per_second": 8502.681
    },
    {
      "epoch": 1.4762321572852142,
      "grad_norm": 1.0264461670280915,
      "learning_rate": 1.599283990048472e-06,
      "loss": 0.2873664379119873,
      "num_input_tokens_seen": 359053176,
      "step": 21925,
      "train_runtime": 42228.2245,
      "train_tokens_per_second": 8502.682
    },
    {
      "epoch": 1.4765688122811742,
      "grad_norm": 1.229444720853511,
      "learning_rate": 1.5973461438066533e-06,
      "loss": 0.29301431179046633,
      "num_input_tokens_seen": 359135096,
      "step": 21930,
      "train_runtime": 42237.8625,
      "train_tokens_per_second": 8502.682
    },
    {
      "epoch": 1.4769054672771345,
      "grad_norm": 1.0512464741251597,
      "learning_rate": 1.5954092491061351e-06,
      "loss": 0.273439884185791,
      "num_input_tokens_seen": 359217016,
      "step": 21935,
      "train_runtime": 42247.688,
      "train_tokens_per_second": 8502.643
    },
    {
      "epoch": 1.4772421222730945,
      "grad_norm": 0.8388790860089982,
      "learning_rate": 1.5934733064885637e-06,
      "loss": 0.28354558944702146,
      "num_input_tokens_seen": 359298936,
      "step": 21940,
      "train_runtime": 42257.195,
      "train_tokens_per_second": 8502.669
    },
    {
      "epoch": 1.4775787772690547,
      "grad_norm": 1.2660091337846746,
      "learning_rate": 1.5915383164953203e-06,
      "loss": 0.2984761714935303,
      "num_input_tokens_seen": 359380856,
      "step": 21945,
      "train_runtime": 42266.7214,
      "train_tokens_per_second": 8502.691
    },
    {
      "epoch": 1.4779154322650148,
      "grad_norm": 0.9651973395933919,
      "learning_rate": 1.5896042796675155e-06,
      "loss": 0.2812242031097412,
      "num_input_tokens_seen": 359462776,
      "step": 21950,
      "train_runtime": 42276.3445,
      "train_tokens_per_second": 8502.693
    },
    {
      "epoch": 1.478252087260975,
      "grad_norm": 1.197402280520021,
      "learning_rate": 1.5876711965460023e-06,
      "loss": 0.2911383628845215,
      "num_input_tokens_seen": 359544696,
      "step": 21955,
      "train_runtime": 42285.9746,
      "train_tokens_per_second": 8502.694
    },
    {
      "epoch": 1.478588742256935,
      "grad_norm": 1.1689601459819363,
      "learning_rate": 1.5857390676713546e-06,
      "loss": 0.3180189371109009,
      "num_input_tokens_seen": 359625760,
      "step": 21960,
      "train_runtime": 42295.6803,
      "train_tokens_per_second": 8502.659
    },
    {
      "epoch": 1.4789253972528953,
      "grad_norm": 1.1014161223493393,
      "learning_rate": 1.5838078935838913e-06,
      "loss": 0.2497502326965332,
      "num_input_tokens_seen": 359707680,
      "step": 21965,
      "train_runtime": 42305.3482,
      "train_tokens_per_second": 8502.653
    },
    {
      "epoch": 1.4792620522488553,
      "grad_norm": 1.1492882892334084,
      "learning_rate": 1.581877674823657e-06,
      "loss": 0.28612394332885743,
      "num_input_tokens_seen": 359789600,
      "step": 21970,
      "train_runtime": 42314.9268,
      "train_tokens_per_second": 8502.664
    },
    {
      "epoch": 1.4795987072448156,
      "grad_norm": 1.0586770702063266,
      "learning_rate": 1.5799484119304308e-06,
      "loss": 0.25247650146484374,
      "num_input_tokens_seen": 359871248,
      "step": 21975,
      "train_runtime": 42324.4347,
      "train_tokens_per_second": 8502.683
    },
    {
      "epoch": 1.4799353622407756,
      "grad_norm": 1.2349697197280654,
      "learning_rate": 1.578020105443726e-06,
      "loss": 0.25716514587402345,
      "num_input_tokens_seen": 359953168,
      "step": 21980,
      "train_runtime": 42333.9514,
      "train_tokens_per_second": 8502.707
    },
    {
      "epoch": 1.4802720172367358,
      "grad_norm": 1.289746639663303,
      "learning_rate": 1.576092755902787e-06,
      "loss": 0.30326333045959475,
      "num_input_tokens_seen": 360035088,
      "step": 21985,
      "train_runtime": 42343.4622,
      "train_tokens_per_second": 8502.731
    },
    {
      "epoch": 1.4806086722326959,
      "grad_norm": 1.060037408732626,
      "learning_rate": 1.5741663638465903e-06,
      "loss": 0.3065227270126343,
      "num_input_tokens_seen": 360117008,
      "step": 21990,
      "train_runtime": 42352.9688,
      "train_tokens_per_second": 8502.757
    },
    {
      "epoch": 1.4809453272286561,
      "grad_norm": 1.032856164580151,
      "learning_rate": 1.5722409298138463e-06,
      "loss": 0.25143256187438967,
      "num_input_tokens_seen": 360198928,
      "step": 21995,
      "train_runtime": 42362.4938,
      "train_tokens_per_second": 8502.779
    },
    {
      "epoch": 1.4812819822246162,
      "grad_norm": 1.4130637762928453,
      "learning_rate": 1.570316454342994e-06,
      "loss": 0.3109081506729126,
      "num_input_tokens_seen": 360280848,
      "step": 22000,
      "train_runtime": 42372.173,
      "train_tokens_per_second": 8502.77
    },
    {
      "epoch": 1.4816186372205764,
      "grad_norm": 1.0004194887221276,
      "learning_rate": 1.568392937972212e-06,
      "loss": 0.29214727878570557,
      "num_input_tokens_seen": 360362768,
      "step": 22005,
      "train_runtime": 42381.8649,
      "train_tokens_per_second": 8502.759
    },
    {
      "epoch": 1.4819552922165364,
      "grad_norm": 1.096901434457282,
      "learning_rate": 1.5664703812393988e-06,
      "loss": 0.27561368942260744,
      "num_input_tokens_seen": 360444688,
      "step": 22010,
      "train_runtime": 42391.3878,
      "train_tokens_per_second": 8502.781
    },
    {
      "epoch": 1.4822919472124967,
      "grad_norm": 1.2185988601933906,
      "learning_rate": 1.5645487846821967e-06,
      "loss": 0.2896237850189209,
      "num_input_tokens_seen": 360526608,
      "step": 22015,
      "train_runtime": 42400.8963,
      "train_tokens_per_second": 8502.806
    },
    {
      "epoch": 1.4826286022084567,
      "grad_norm": 1.0908825987114383,
      "learning_rate": 1.5626281488379724e-06,
      "loss": 0.28876752853393556,
      "num_input_tokens_seen": 360608528,
      "step": 22020,
      "train_runtime": 42410.4104,
      "train_tokens_per_second": 8502.83
    },
    {
      "epoch": 1.482965257204417,
      "grad_norm": 1.0447199274948693,
      "learning_rate": 1.560708474243825e-06,
      "loss": 0.2555402278900146,
      "num_input_tokens_seen": 360690448,
      "step": 22025,
      "train_runtime": 42420.027,
      "train_tokens_per_second": 8502.834
    },
    {
      "epoch": 1.483301912200377,
      "grad_norm": 1.3426435772295116,
      "learning_rate": 1.5587897614365854e-06,
      "loss": 0.28762969970703123,
      "num_input_tokens_seen": 360772368,
      "step": 22030,
      "train_runtime": 42429.6825,
      "train_tokens_per_second": 8502.83
    },
    {
      "epoch": 1.4836385671963372,
      "grad_norm": 1.0451509365305915,
      "learning_rate": 1.5568720109528163e-06,
      "loss": 0.3094435214996338,
      "num_input_tokens_seen": 360854288,
      "step": 22035,
      "train_runtime": 42439.1978,
      "train_tokens_per_second": 8502.854
    },
    {
      "epoch": 1.4839752221922973,
      "grad_norm": 1.0216945842596055,
      "learning_rate": 1.5549552233288095e-06,
      "loss": 0.29816465377807616,
      "num_input_tokens_seen": 360935968,
      "step": 22040,
      "train_runtime": 42449.192,
      "train_tokens_per_second": 8502.776
    },
    {
      "epoch": 1.4843118771882575,
      "grad_norm": 1.126097682077238,
      "learning_rate": 1.5530393991005893e-06,
      "loss": 0.26206154823303224,
      "num_input_tokens_seen": 361017888,
      "step": 22045,
      "train_runtime": 42458.704,
      "train_tokens_per_second": 8502.8
    },
    {
      "epoch": 1.4846485321842176,
      "grad_norm": 1.0652335676138556,
      "learning_rate": 1.551124538803908e-06,
      "loss": 0.24426426887512206,
      "num_input_tokens_seen": 361099808,
      "step": 22050,
      "train_runtime": 42468.2146,
      "train_tokens_per_second": 8502.825
    },
    {
      "epoch": 1.4849851871801778,
      "grad_norm": 1.1986232050948178,
      "learning_rate": 1.5492106429742547e-06,
      "loss": 0.29026074409484864,
      "num_input_tokens_seen": 361181728,
      "step": 22055,
      "train_runtime": 42477.7255,
      "train_tokens_per_second": 8502.85
    },
    {
      "epoch": 1.4853218421761378,
      "grad_norm": 1.082281601839436,
      "learning_rate": 1.5472977121468386e-06,
      "loss": 0.3017944097518921,
      "num_input_tokens_seen": 361263648,
      "step": 22060,
      "train_runtime": 42487.2471,
      "train_tokens_per_second": 8502.873
    },
    {
      "epoch": 1.485658497172098,
      "grad_norm": 1.1693944189548384,
      "learning_rate": 1.545385746856609e-06,
      "loss": 0.27766034603118894,
      "num_input_tokens_seen": 361345568,
      "step": 22065,
      "train_runtime": 42496.7571,
      "train_tokens_per_second": 8502.897
    },
    {
      "epoch": 1.485995152168058,
      "grad_norm": 1.0127037583532756,
      "learning_rate": 1.5434747476382388e-06,
      "loss": 0.2432957887649536,
      "num_input_tokens_seen": 361427488,
      "step": 22070,
      "train_runtime": 42506.3968,
      "train_tokens_per_second": 8502.896
    },
    {
      "epoch": 1.4863318071640184,
      "grad_norm": 1.0577114439901143,
      "learning_rate": 1.5415647150261337e-06,
      "loss": 0.2636049747467041,
      "num_input_tokens_seen": 361509408,
      "step": 22075,
      "train_runtime": 42515.9064,
      "train_tokens_per_second": 8502.921
    },
    {
      "epoch": 1.4866684621599784,
      "grad_norm": 1.2064421754043584,
      "learning_rate": 1.539655649554428e-06,
      "loss": 0.30742831230163575,
      "num_input_tokens_seen": 361591328,
      "step": 22080,
      "train_runtime": 42525.4798,
      "train_tokens_per_second": 8502.934
    },
    {
      "epoch": 1.4870051171559386,
      "grad_norm": 1.1861836895181777,
      "learning_rate": 1.5377475517569856e-06,
      "loss": 0.26369714736938477,
      "num_input_tokens_seen": 361673248,
      "step": 22085,
      "train_runtime": 42535.1097,
      "train_tokens_per_second": 8502.934
    },
    {
      "epoch": 1.4873417721518987,
      "grad_norm": 1.1790320205656708,
      "learning_rate": 1.5358404221674e-06,
      "loss": 0.3031788349151611,
      "num_input_tokens_seen": 361755168,
      "step": 22090,
      "train_runtime": 42544.6174,
      "train_tokens_per_second": 8502.96
    },
    {
      "epoch": 1.487678427147859,
      "grad_norm": 1.1080285245475283,
      "learning_rate": 1.533934261318994e-06,
      "loss": 0.2941728115081787,
      "num_input_tokens_seen": 361837088,
      "step": 22095,
      "train_runtime": 42554.146,
      "train_tokens_per_second": 8502.981
    },
    {
      "epoch": 1.488015082143819,
      "grad_norm": 1.0911646009007776,
      "learning_rate": 1.5320290697448177e-06,
      "loss": 0.277821683883667,
      "num_input_tokens_seen": 361918896,
      "step": 22100,
      "train_runtime": 42564.3588,
      "train_tokens_per_second": 8502.863
    },
    {
      "epoch": 1.4883517371397792,
      "grad_norm": 1.1326259331689645,
      "learning_rate": 1.530124847977657e-06,
      "loss": 0.26886117458343506,
      "num_input_tokens_seen": 362000816,
      "step": 22105,
      "train_runtime": 42574.0265,
      "train_tokens_per_second": 8502.856
    },
    {
      "epoch": 1.4886883921357392,
      "grad_norm": 1.1032471907844885,
      "learning_rate": 1.528221596550014e-06,
      "loss": 0.2563521146774292,
      "num_input_tokens_seen": 362082736,
      "step": 22110,
      "train_runtime": 42583.5458,
      "train_tokens_per_second": 8502.879
    },
    {
      "epoch": 1.4890250471316995,
      "grad_norm": 1.2832380389004179,
      "learning_rate": 1.5263193159941331e-06,
      "loss": 0.2855405569076538,
      "num_input_tokens_seen": 362164656,
      "step": 22115,
      "train_runtime": 42593.058,
      "train_tokens_per_second": 8502.903
    },
    {
      "epoch": 1.4893617021276595,
      "grad_norm": 1.0928600586866812,
      "learning_rate": 1.5244180068419783e-06,
      "loss": 0.2710689067840576,
      "num_input_tokens_seen": 362246576,
      "step": 22120,
      "train_runtime": 42602.5604,
      "train_tokens_per_second": 8502.93
    },
    {
      "epoch": 1.4896983571236198,
      "grad_norm": 1.3665615037587777,
      "learning_rate": 1.5225176696252453e-06,
      "loss": 0.3211559295654297,
      "num_input_tokens_seen": 362328496,
      "step": 22125,
      "train_runtime": 42612.0717,
      "train_tokens_per_second": 8502.954
    },
    {
      "epoch": 1.4900350121195798,
      "grad_norm": 1.4326092361321159,
      "learning_rate": 1.5206183048753575e-06,
      "loss": 0.28330092430114745,
      "num_input_tokens_seen": 362410416,
      "step": 22130,
      "train_runtime": 42621.7566,
      "train_tokens_per_second": 8502.944
    },
    {
      "epoch": 1.49037166711554,
      "grad_norm": 1.3962174607920421,
      "learning_rate": 1.5187199131234653e-06,
      "loss": 0.33648455142974854,
      "num_input_tokens_seen": 362492336,
      "step": 22135,
      "train_runtime": 42631.3425,
      "train_tokens_per_second": 8502.954
    },
    {
      "epoch": 1.4907083221115,
      "grad_norm": 1.1211856352559308,
      "learning_rate": 1.5168224949004484e-06,
      "loss": 0.2812371015548706,
      "num_input_tokens_seen": 362574256,
      "step": 22140,
      "train_runtime": 42640.8548,
      "train_tokens_per_second": 8502.978
    },
    {
      "epoch": 1.4910449771074603,
      "grad_norm": 1.0932121936571797,
      "learning_rate": 1.514926050736914e-06,
      "loss": 0.26793386936187746,
      "num_input_tokens_seen": 362656176,
      "step": 22145,
      "train_runtime": 42650.3598,
      "train_tokens_per_second": 8503.004
    },
    {
      "epoch": 1.4913816321034203,
      "grad_norm": 1.163877428894745,
      "learning_rate": 1.5130305811631945e-06,
      "loss": 0.263005805015564,
      "num_input_tokens_seen": 362738096,
      "step": 22150,
      "train_runtime": 42659.8686,
      "train_tokens_per_second": 8503.029
    },
    {
      "epoch": 1.4917182870993806,
      "grad_norm": 1.266567139828509,
      "learning_rate": 1.5111360867093571e-06,
      "loss": 0.28897287845611574,
      "num_input_tokens_seen": 362820016,
      "step": 22155,
      "train_runtime": 42669.3756,
      "train_tokens_per_second": 8503.054
    },
    {
      "epoch": 1.4920549420953406,
      "grad_norm": 1.373737776943351,
      "learning_rate": 1.5092425679051848e-06,
      "loss": 0.26869151592254636,
      "num_input_tokens_seen": 362901936,
      "step": 22160,
      "train_runtime": 42678.891,
      "train_tokens_per_second": 8503.078
    },
    {
      "epoch": 1.4923915970913009,
      "grad_norm": 1.0980179704795345,
      "learning_rate": 1.5073500252801988e-06,
      "loss": 0.2838193416595459,
      "num_input_tokens_seen": 362983856,
      "step": 22165,
      "train_runtime": 42688.4123,
      "train_tokens_per_second": 8503.1
    },
    {
      "epoch": 1.492728252087261,
      "grad_norm": 1.1584966718593306,
      "learning_rate": 1.5054584593636412e-06,
      "loss": 0.28502635955810546,
      "num_input_tokens_seen": 363065776,
      "step": 22170,
      "train_runtime": 42699.272,
      "train_tokens_per_second": 8502.856
    },
    {
      "epoch": 1.4930649070832211,
      "grad_norm": 1.2046720196348113,
      "learning_rate": 1.5035678706844814e-06,
      "loss": 0.3015527009963989,
      "num_input_tokens_seen": 363147696,
      "step": 22175,
      "train_runtime": 42712.8021,
      "train_tokens_per_second": 8502.081
    },
    {
      "epoch": 1.4934015620791812,
      "grad_norm": 1.1119018075267437,
      "learning_rate": 1.501678259771418e-06,
      "loss": 0.29671053886413573,
      "num_input_tokens_seen": 363229616,
      "step": 22180,
      "train_runtime": 42725.557,
      "train_tokens_per_second": 8501.46
    },
    {
      "epoch": 1.4937382170751414,
      "grad_norm": 1.3142323089886538,
      "learning_rate": 1.499789627152874e-06,
      "loss": 0.28890061378479004,
      "num_input_tokens_seen": 363311536,
      "step": 22185,
      "train_runtime": 42736.5713,
      "train_tokens_per_second": 8501.186
    },
    {
      "epoch": 1.4940748720711015,
      "grad_norm": 1.1570091248210221,
      "learning_rate": 1.497901973356999e-06,
      "loss": 0.2772779941558838,
      "num_input_tokens_seen": 363393456,
      "step": 22190,
      "train_runtime": 42746.0707,
      "train_tokens_per_second": 8501.213
    },
    {
      "epoch": 1.4944115270670617,
      "grad_norm": 1.1167949599581064,
      "learning_rate": 1.4960152989116694e-06,
      "loss": 0.29124748706817627,
      "num_input_tokens_seen": 363475376,
      "step": 22195,
      "train_runtime": 42755.7179,
      "train_tokens_per_second": 8501.211
    },
    {
      "epoch": 1.4947481820630217,
      "grad_norm": 1.085067228736741,
      "learning_rate": 1.4941296043444869e-06,
      "loss": 0.27105400562286375,
      "num_input_tokens_seen": 363557296,
      "step": 22200,
      "train_runtime": 42765.3403,
      "train_tokens_per_second": 8501.214
    },
    {
      "epoch": 1.495084837058982,
      "grad_norm": 0.8432413368105534,
      "learning_rate": 1.4922448901827847e-06,
      "loss": 0.267771053314209,
      "num_input_tokens_seen": 363639216,
      "step": 22205,
      "train_runtime": 42775.0575,
      "train_tokens_per_second": 8501.198
    },
    {
      "epoch": 1.495421492054942,
      "grad_norm": 1.1686148361727295,
      "learning_rate": 1.4903611569536103e-06,
      "loss": 0.278275728225708,
      "num_input_tokens_seen": 363720976,
      "step": 22210,
      "train_runtime": 42785.124,
      "train_tokens_per_second": 8501.108
    },
    {
      "epoch": 1.4957581470509023,
      "grad_norm": 1.0809695619505972,
      "learning_rate": 1.488478405183748e-06,
      "loss": 0.3007126569747925,
      "num_input_tokens_seen": 363802896,
      "step": 22215,
      "train_runtime": 42802.6893,
      "train_tokens_per_second": 8499.534
    },
    {
      "epoch": 1.4960948020468623,
      "grad_norm": 1.320406651433031,
      "learning_rate": 1.4865966353997024e-06,
      "loss": 0.30350706577301023,
      "num_input_tokens_seen": 363884816,
      "step": 22220,
      "train_runtime": 42812.3237,
      "train_tokens_per_second": 8499.534
    },
    {
      "epoch": 1.4964314570428225,
      "grad_norm": 0.9175286807921906,
      "learning_rate": 1.4847158481277047e-06,
      "loss": 0.26477599143981934,
      "num_input_tokens_seen": 363966736,
      "step": 22225,
      "train_runtime": 42821.8341,
      "train_tokens_per_second": 8499.56
    },
    {
      "epoch": 1.4967681120387826,
      "grad_norm": 1.1070635835341065,
      "learning_rate": 1.48283604389371e-06,
      "loss": 0.26304116249084475,
      "num_input_tokens_seen": 364048656,
      "step": 22230,
      "train_runtime": 42831.3404,
      "train_tokens_per_second": 8499.586
    },
    {
      "epoch": 1.4971047670347428,
      "grad_norm": 0.9779877646012052,
      "learning_rate": 1.480957223223401e-06,
      "loss": 0.279502272605896,
      "num_input_tokens_seen": 364130576,
      "step": 22235,
      "train_runtime": 42840.8659,
      "train_tokens_per_second": 8499.608
    },
    {
      "epoch": 1.497441422030703,
      "grad_norm": 1.0076288002542797,
      "learning_rate": 1.4790793866421826e-06,
      "loss": 0.29399938583374025,
      "num_input_tokens_seen": 364212496,
      "step": 22240,
      "train_runtime": 42850.3697,
      "train_tokens_per_second": 8499.635
    },
    {
      "epoch": 1.497778077026663,
      "grad_norm": 1.1422531176340396,
      "learning_rate": 1.4772025346751862e-06,
      "loss": 0.3096950054168701,
      "num_input_tokens_seen": 364293688,
      "step": 22245,
      "train_runtime": 42860.2445,
      "train_tokens_per_second": 8499.571
    },
    {
      "epoch": 1.4981147320226231,
      "grad_norm": 1.1747935882825686,
      "learning_rate": 1.475326667847266e-06,
      "loss": 0.27430429458618166,
      "num_input_tokens_seen": 364375608,
      "step": 22250,
      "train_runtime": 42869.8108,
      "train_tokens_per_second": 8499.585
    },
    {
      "epoch": 1.4984513870185834,
      "grad_norm": 1.1897731409245267,
      "learning_rate": 1.4734517866830073e-06,
      "loss": 0.27205595970153806,
      "num_input_tokens_seen": 364457528,
      "step": 22255,
      "train_runtime": 42879.3193,
      "train_tokens_per_second": 8499.611
    },
    {
      "epoch": 1.4987880420145436,
      "grad_norm": 1.0622055624897144,
      "learning_rate": 1.4715778917067081e-06,
      "loss": 0.28821425437927245,
      "num_input_tokens_seen": 364539448,
      "step": 22260,
      "train_runtime": 42888.8318,
      "train_tokens_per_second": 8499.636
    },
    {
      "epoch": 1.4991246970105037,
      "grad_norm": 1.1110625451759943,
      "learning_rate": 1.4697049834424016e-06,
      "loss": 0.2725270509719849,
      "num_input_tokens_seen": 364621368,
      "step": 22265,
      "train_runtime": 42898.3867,
      "train_tokens_per_second": 8499.652
    },
    {
      "epoch": 1.4994613520064637,
      "grad_norm": 1.168587674864366,
      "learning_rate": 1.4678330624138387e-06,
      "loss": 0.2839771270751953,
      "num_input_tokens_seen": 364703288,
      "step": 22270,
      "train_runtime": 42907.9176,
      "train_tokens_per_second": 8499.673
    },
    {
      "epoch": 1.499798007002424,
      "grad_norm": 1.1207616356734864,
      "learning_rate": 1.465962129144497e-06,
      "loss": 0.2913990020751953,
      "num_input_tokens_seen": 364785208,
      "step": 22275,
      "train_runtime": 42917.5225,
      "train_tokens_per_second": 8499.68
    },
    {
      "epoch": 1.5001346619983842,
      "grad_norm": 1.3387407542455263,
      "learning_rate": 1.4640921841575766e-06,
      "loss": 0.29648404121398925,
      "num_input_tokens_seen": 364867128,
      "step": 22280,
      "train_runtime": 42927.1248,
      "train_tokens_per_second": 8499.687
    },
    {
      "epoch": 1.5004713169943442,
      "grad_norm": 1.09901649868177,
      "learning_rate": 1.4622232279760018e-06,
      "loss": 0.3071413278579712,
      "num_input_tokens_seen": 364949048,
      "step": 22285,
      "train_runtime": 42936.6261,
      "train_tokens_per_second": 8499.714
    },
    {
      "epoch": 1.5008079719903042,
      "grad_norm": 1.0058207853986838,
      "learning_rate": 1.4603552611224202e-06,
      "loss": 0.2682831287384033,
      "num_input_tokens_seen": 365030968,
      "step": 22290,
      "train_runtime": 42946.133,
      "train_tokens_per_second": 8499.74
    },
    {
      "epoch": 1.5011446269862645,
      "grad_norm": 1.117253547587297,
      "learning_rate": 1.4584882841192027e-06,
      "loss": 0.31122303009033203,
      "num_input_tokens_seen": 365112888,
      "step": 22295,
      "train_runtime": 42955.6455,
      "train_tokens_per_second": 8499.765
    },
    {
      "epoch": 1.5014812819822247,
      "grad_norm": 1.1632065672179843,
      "learning_rate": 1.456622297488442e-06,
      "loss": 0.25730180740356445,
      "num_input_tokens_seen": 365194808,
      "step": 22300,
      "train_runtime": 42965.2499,
      "train_tokens_per_second": 8499.772
    },
    {
      "epoch": 1.5018179369781848,
      "grad_norm": 1.1553802914697837,
      "learning_rate": 1.4547573017519595e-06,
      "loss": 0.28183574676513673,
      "num_input_tokens_seen": 365276728,
      "step": 22305,
      "train_runtime": 42974.8348,
      "train_tokens_per_second": 8499.782
    },
    {
      "epoch": 1.5021545919741448,
      "grad_norm": 1.1814167893766485,
      "learning_rate": 1.4528932974312897e-06,
      "loss": 0.25498619079589846,
      "num_input_tokens_seen": 365358648,
      "step": 22310,
      "train_runtime": 42984.3479,
      "train_tokens_per_second": 8499.807
    },
    {
      "epoch": 1.502491246970105,
      "grad_norm": 1.2595693922605797,
      "learning_rate": 1.4510302850477005e-06,
      "loss": 0.27510340213775636,
      "num_input_tokens_seen": 365440568,
      "step": 22315,
      "train_runtime": 42993.8612,
      "train_tokens_per_second": 8499.831
    },
    {
      "epoch": 1.5028279019660653,
      "grad_norm": 1.0403206559163887,
      "learning_rate": 1.4491682651221746e-06,
      "loss": 0.2630305767059326,
      "num_input_tokens_seen": 365522488,
      "step": 22320,
      "train_runtime": 43003.3879,
      "train_tokens_per_second": 8499.853
    },
    {
      "epoch": 1.5031645569620253,
      "grad_norm": 1.0886728034666344,
      "learning_rate": 1.447307238175421e-06,
      "loss": 0.27028467655181887,
      "num_input_tokens_seen": 365604408,
      "step": 22325,
      "train_runtime": 43012.9164,
      "train_tokens_per_second": 8499.875
    },
    {
      "epoch": 1.5035012119579854,
      "grad_norm": 1.0609054466039043,
      "learning_rate": 1.4454472047278688e-06,
      "loss": 0.273946475982666,
      "num_input_tokens_seen": 365686328,
      "step": 22330,
      "train_runtime": 43022.6223,
      "train_tokens_per_second": 8499.861
    },
    {
      "epoch": 1.5038378669539456,
      "grad_norm": 1.1608914058096986,
      "learning_rate": 1.443588165299672e-06,
      "loss": 0.27847537994384763,
      "num_input_tokens_seen": 365768248,
      "step": 22335,
      "train_runtime": 43032.3137,
      "train_tokens_per_second": 8499.851
    },
    {
      "epoch": 1.5041745219499059,
      "grad_norm": 1.0645271886143794,
      "learning_rate": 1.4417301204107037e-06,
      "loss": 0.2903605937957764,
      "num_input_tokens_seen": 365850168,
      "step": 22340,
      "train_runtime": 43041.9376,
      "train_tokens_per_second": 8499.854
    },
    {
      "epoch": 1.504511176945866,
      "grad_norm": 1.304960193172075,
      "learning_rate": 1.4398730705805607e-06,
      "loss": 0.28866167068481446,
      "num_input_tokens_seen": 365932088,
      "step": 22345,
      "train_runtime": 43051.5539,
      "train_tokens_per_second": 8499.858
    },
    {
      "epoch": 1.504847831941826,
      "grad_norm": 1.1579057726903736,
      "learning_rate": 1.43801701632856e-06,
      "loss": 0.28943533897399903,
      "num_input_tokens_seen": 366014008,
      "step": 22350,
      "train_runtime": 43061.0799,
      "train_tokens_per_second": 8499.88
    },
    {
      "epoch": 1.5051844869377862,
      "grad_norm": 1.2169441408859154,
      "learning_rate": 1.436161958173745e-06,
      "loss": 0.3021491527557373,
      "num_input_tokens_seen": 366095928,
      "step": 22355,
      "train_runtime": 43070.7486,
      "train_tokens_per_second": 8499.874
    },
    {
      "epoch": 1.5055211419337464,
      "grad_norm": 0.95028664278426,
      "learning_rate": 1.4343078966348718e-06,
      "loss": 0.2746077060699463,
      "num_input_tokens_seen": 366177848,
      "step": 22360,
      "train_runtime": 43080.412,
      "train_tokens_per_second": 8499.869
    },
    {
      "epoch": 1.5058577969297064,
      "grad_norm": 1.1499327556891057,
      "learning_rate": 1.432454832230426e-06,
      "loss": 0.27276661396026614,
      "num_input_tokens_seen": 366259768,
      "step": 22365,
      "train_runtime": 43090.1,
      "train_tokens_per_second": 8499.859
    },
    {
      "epoch": 1.5061944519256665,
      "grad_norm": 1.1888445978886288,
      "learning_rate": 1.4306027654786103e-06,
      "loss": 0.2776324272155762,
      "num_input_tokens_seen": 366341688,
      "step": 22370,
      "train_runtime": 43099.7566,
      "train_tokens_per_second": 8499.855
    },
    {
      "epoch": 1.5065311069216267,
      "grad_norm": 0.9961091516166566,
      "learning_rate": 1.4287516968973491e-06,
      "loss": 0.30149648189544676,
      "num_input_tokens_seen": 366423608,
      "step": 22375,
      "train_runtime": 43109.2714,
      "train_tokens_per_second": 8499.879
    },
    {
      "epoch": 1.506867761917587,
      "grad_norm": 1.2162599770211762,
      "learning_rate": 1.4269016270042884e-06,
      "loss": 0.299886417388916,
      "num_input_tokens_seen": 366505528,
      "step": 22380,
      "train_runtime": 43118.7829,
      "train_tokens_per_second": 8499.904
    },
    {
      "epoch": 1.507204416913547,
      "grad_norm": 1.2438789459485617,
      "learning_rate": 1.425052556316794e-06,
      "loss": 0.2520773887634277,
      "num_input_tokens_seen": 366587448,
      "step": 22385,
      "train_runtime": 43128.2984,
      "train_tokens_per_second": 8499.928
    },
    {
      "epoch": 1.507541071909507,
      "grad_norm": 1.0630096202192558,
      "learning_rate": 1.4232044853519522e-06,
      "loss": 0.29753541946411133,
      "num_input_tokens_seen": 366669368,
      "step": 22390,
      "train_runtime": 43137.8324,
      "train_tokens_per_second": 8499.949
    },
    {
      "epoch": 1.5078777269054673,
      "grad_norm": 0.9031040054887333,
      "learning_rate": 1.4213574146265708e-06,
      "loss": 0.2710383176803589,
      "num_input_tokens_seen": 366751288,
      "step": 22395,
      "train_runtime": 43147.3864,
      "train_tokens_per_second": 8499.965
    },
    {
      "epoch": 1.5082143819014275,
      "grad_norm": 1.0912564820261894,
      "learning_rate": 1.4195113446571752e-06,
      "loss": 0.2966273307800293,
      "num_input_tokens_seen": 366832160,
      "step": 22400,
      "train_runtime": 43156.9652,
      "train_tokens_per_second": 8499.953
    },
    {
      "epoch": 1.5085510368973876,
      "grad_norm": 1.2576906701489938,
      "learning_rate": 1.417666275960018e-06,
      "loss": 0.28866066932678225,
      "num_input_tokens_seen": 366914080,
      "step": 22405,
      "train_runtime": 43166.5242,
      "train_tokens_per_second": 8499.968
    },
    {
      "epoch": 1.5088876918933476,
      "grad_norm": 1.201524363314871,
      "learning_rate": 1.4158222090510605e-06,
      "loss": 0.2685322046279907,
      "num_input_tokens_seen": 366996000,
      "step": 22410,
      "train_runtime": 43176.1411,
      "train_tokens_per_second": 8499.972
    },
    {
      "epoch": 1.5092243468893078,
      "grad_norm": 1.0127163283160363,
      "learning_rate": 1.4139791444459945e-06,
      "loss": 0.28462848663330076,
      "num_input_tokens_seen": 367077920,
      "step": 22415,
      "train_runtime": 43185.7135,
      "train_tokens_per_second": 8499.985
    },
    {
      "epoch": 1.509561001885268,
      "grad_norm": 1.164968857923881,
      "learning_rate": 1.4121370826602254e-06,
      "loss": 0.2679532289505005,
      "num_input_tokens_seen": 367159840,
      "step": 22420,
      "train_runtime": 43195.3061,
      "train_tokens_per_second": 8499.994
    },
    {
      "epoch": 1.5098976568812281,
      "grad_norm": 1.1781199648837795,
      "learning_rate": 1.4102960242088803e-06,
      "loss": 0.2849021673202515,
      "num_input_tokens_seen": 367241760,
      "step": 22425,
      "train_runtime": 43204.8164,
      "train_tokens_per_second": 8500.019
    },
    {
      "epoch": 1.5102343118771882,
      "grad_norm": 1.1924369440705402,
      "learning_rate": 1.4084559696068057e-06,
      "loss": 0.2876799821853638,
      "num_input_tokens_seen": 367323680,
      "step": 22430,
      "train_runtime": 43214.3237,
      "train_tokens_per_second": 8500.045
    },
    {
      "epoch": 1.5105709668731484,
      "grad_norm": 1.077342561966995,
      "learning_rate": 1.406616919368567e-06,
      "loss": 0.29683678150177,
      "num_input_tokens_seen": 367405600,
      "step": 22435,
      "train_runtime": 43223.8306,
      "train_tokens_per_second": 8500.07
    },
    {
      "epoch": 1.5109076218691087,
      "grad_norm": 1.1111585687765502,
      "learning_rate": 1.4047788740084474e-06,
      "loss": 0.27262425422668457,
      "num_input_tokens_seen": 367487520,
      "step": 22440,
      "train_runtime": 43233.3412,
      "train_tokens_per_second": 8500.095
    },
    {
      "epoch": 1.5112442768650687,
      "grad_norm": 1.1434457693318583,
      "learning_rate": 1.402941834040452e-06,
      "loss": 0.2691437482833862,
      "num_input_tokens_seen": 367569440,
      "step": 22445,
      "train_runtime": 43242.8473,
      "train_tokens_per_second": 8500.121
    },
    {
      "epoch": 1.5115809318610287,
      "grad_norm": 0.9958250225569917,
      "learning_rate": 1.4011057999783006e-06,
      "loss": 0.26355345249176027,
      "num_input_tokens_seen": 367651128,
      "step": 22450,
      "train_runtime": 43252.3578,
      "train_tokens_per_second": 8500.141
    },
    {
      "epoch": 1.511917586856989,
      "grad_norm": 1.105138109835079,
      "learning_rate": 1.3992707723354398e-06,
      "loss": 0.25745244026184083,
      "num_input_tokens_seen": 367733048,
      "step": 22455,
      "train_runtime": 43261.8638,
      "train_tokens_per_second": 8500.167
    },
    {
      "epoch": 1.5122542418529492,
      "grad_norm": 0.972943418986154,
      "learning_rate": 1.397436751625022e-06,
      "loss": 0.26087088584899903,
      "num_input_tokens_seen": 367814968,
      "step": 22460,
      "train_runtime": 43271.368,
      "train_tokens_per_second": 8500.193
    },
    {
      "epoch": 1.5125908968489092,
      "grad_norm": 1.054602884957364,
      "learning_rate": 1.3956037383599308e-06,
      "loss": 0.28028876781463624,
      "num_input_tokens_seen": 367896888,
      "step": 22465,
      "train_runtime": 43280.8706,
      "train_tokens_per_second": 8500.219
    },
    {
      "epoch": 1.5129275518448693,
      "grad_norm": 0.9905543234957831,
      "learning_rate": 1.39377173305276e-06,
      "loss": 0.28860323429107665,
      "num_input_tokens_seen": 367978808,
      "step": 22470,
      "train_runtime": 43290.3743,
      "train_tokens_per_second": 8500.245
    },
    {
      "epoch": 1.5132642068408295,
      "grad_norm": 1.0559504955924768,
      "learning_rate": 1.3919407362158254e-06,
      "loss": 0.30568323135375974,
      "num_input_tokens_seen": 368060728,
      "step": 22475,
      "train_runtime": 43299.8839,
      "train_tokens_per_second": 8500.271
    },
    {
      "epoch": 1.5136008618367898,
      "grad_norm": 1.1696037634436272,
      "learning_rate": 1.3901107483611582e-06,
      "loss": 0.29840779304504395,
      "num_input_tokens_seen": 368142648,
      "step": 22480,
      "train_runtime": 43309.3955,
      "train_tokens_per_second": 8500.295
    },
    {
      "epoch": 1.5139375168327498,
      "grad_norm": 0.9277469608578434,
      "learning_rate": 1.3882817700005091e-06,
      "loss": 0.2569321632385254,
      "num_input_tokens_seen": 368224568,
      "step": 22485,
      "train_runtime": 43318.9028,
      "train_tokens_per_second": 8500.321
    },
    {
      "epoch": 1.5142741718287098,
      "grad_norm": 0.9802810894886128,
      "learning_rate": 1.3864538016453466e-06,
      "loss": 0.2612179756164551,
      "num_input_tokens_seen": 368306488,
      "step": 22490,
      "train_runtime": 43328.4208,
      "train_tokens_per_second": 8500.344
    },
    {
      "epoch": 1.51461082682467,
      "grad_norm": 1.1817329120804752,
      "learning_rate": 1.384626843806855e-06,
      "loss": 0.310501766204834,
      "num_input_tokens_seen": 368388408,
      "step": 22495,
      "train_runtime": 43338.0891,
      "train_tokens_per_second": 8500.338
    },
    {
      "epoch": 1.5149474818206303,
      "grad_norm": 1.0880003422025404,
      "learning_rate": 1.3828008969959367e-06,
      "loss": 0.30503115653991697,
      "num_input_tokens_seen": 368470328,
      "step": 22500,
      "train_runtime": 43347.685,
      "train_tokens_per_second": 8500.346
    },
    {
      "epoch": 1.5152841368165904,
      "grad_norm": 1.2685873206381129,
      "learning_rate": 1.3809759617232166e-06,
      "loss": 0.29964115619659426,
      "num_input_tokens_seen": 368552248,
      "step": 22505,
      "train_runtime": 43357.2009,
      "train_tokens_per_second": 8500.37
    },
    {
      "epoch": 1.5156207918125504,
      "grad_norm": 1.2899765269845478,
      "learning_rate": 1.3791520384990247e-06,
      "loss": 0.26866605281829836,
      "num_input_tokens_seen": 368634168,
      "step": 22510,
      "train_runtime": 43366.7034,
      "train_tokens_per_second": 8500.396
    },
    {
      "epoch": 1.5159574468085106,
      "grad_norm": 1.025602958555931,
      "learning_rate": 1.3773291278334216e-06,
      "loss": 0.2675811767578125,
      "num_input_tokens_seen": 368716088,
      "step": 22515,
      "train_runtime": 43376.1993,
      "train_tokens_per_second": 8500.424
    },
    {
      "epoch": 1.5162941018044709,
      "grad_norm": 1.1805016807289221,
      "learning_rate": 1.3755072302361754e-06,
      "loss": 0.31174893379211427,
      "num_input_tokens_seen": 368798008,
      "step": 22520,
      "train_runtime": 43385.7152,
      "train_tokens_per_second": 8500.448
    },
    {
      "epoch": 1.516630756800431,
      "grad_norm": 0.9308024332375736,
      "learning_rate": 1.3736863462167749e-06,
      "loss": 0.2503079414367676,
      "num_input_tokens_seen": 368879928,
      "step": 22525,
      "train_runtime": 43395.4204,
      "train_tokens_per_second": 8500.434
    },
    {
      "epoch": 1.516967411796391,
      "grad_norm": 0.9722833430878245,
      "learning_rate": 1.3718664762844242e-06,
      "loss": 0.26974759101867674,
      "num_input_tokens_seen": 368961848,
      "step": 22530,
      "train_runtime": 43405.1036,
      "train_tokens_per_second": 8500.425
    },
    {
      "epoch": 1.5173040667923512,
      "grad_norm": 1.0346526630949602,
      "learning_rate": 1.370047620948044e-06,
      "loss": 0.2975128650665283,
      "num_input_tokens_seen": 369043768,
      "step": 22535,
      "train_runtime": 43414.802,
      "train_tokens_per_second": 8500.413
    },
    {
      "epoch": 1.5176407217883114,
      "grad_norm": 1.7891685161454305,
      "learning_rate": 1.368229780716272e-06,
      "loss": 0.2878409385681152,
      "num_input_tokens_seen": 369125688,
      "step": 22540,
      "train_runtime": 43424.4855,
      "train_tokens_per_second": 8500.404
    },
    {
      "epoch": 1.5179773767842715,
      "grad_norm": 1.1916515972850632,
      "learning_rate": 1.3664129560974604e-06,
      "loss": 0.28114023208618166,
      "num_input_tokens_seen": 369207608,
      "step": 22545,
      "train_runtime": 43433.994,
      "train_tokens_per_second": 8500.43
    },
    {
      "epoch": 1.5183140317802315,
      "grad_norm": 1.3183891607347786,
      "learning_rate": 1.3645971475996777e-06,
      "loss": 0.2701871871948242,
      "num_input_tokens_seen": 369289528,
      "step": 22550,
      "train_runtime": 43443.4967,
      "train_tokens_per_second": 8500.456
    },
    {
      "epoch": 1.5186506867761917,
      "grad_norm": 1.2653605744826362,
      "learning_rate": 1.3627823557307135e-06,
      "loss": 0.29760422706604006,
      "num_input_tokens_seen": 369371448,
      "step": 22555,
      "train_runtime": 43453.0008,
      "train_tokens_per_second": 8500.482
    },
    {
      "epoch": 1.518987341772152,
      "grad_norm": 1.063843081285285,
      "learning_rate": 1.3609685809980616e-06,
      "loss": 0.27868084907531737,
      "num_input_tokens_seen": 369453368,
      "step": 22560,
      "train_runtime": 43462.6222,
      "train_tokens_per_second": 8500.485
    },
    {
      "epoch": 1.519323996768112,
      "grad_norm": 1.18397825242903,
      "learning_rate": 1.3591558239089443e-06,
      "loss": 0.2799037218093872,
      "num_input_tokens_seen": 369535288,
      "step": 22565,
      "train_runtime": 43472.2914,
      "train_tokens_per_second": 8500.479
    },
    {
      "epoch": 1.519660651764072,
      "grad_norm": 1.216148060961504,
      "learning_rate": 1.3573440849702902e-06,
      "loss": 0.3228152275085449,
      "num_input_tokens_seen": 369617208,
      "step": 22570,
      "train_runtime": 43481.7935,
      "train_tokens_per_second": 8500.505
    },
    {
      "epoch": 1.5199973067600323,
      "grad_norm": 1.0664755239223818,
      "learning_rate": 1.3555333646887476e-06,
      "loss": 0.2931244373321533,
      "num_input_tokens_seen": 369699128,
      "step": 22575,
      "train_runtime": 43491.2969,
      "train_tokens_per_second": 8500.531
    },
    {
      "epoch": 1.5203339617559926,
      "grad_norm": 1.0526272438376278,
      "learning_rate": 1.3537236635706779e-06,
      "loss": 0.2928951263427734,
      "num_input_tokens_seen": 369781048,
      "step": 22580,
      "train_runtime": 43500.8151,
      "train_tokens_per_second": 8500.554
    },
    {
      "epoch": 1.5206706167519526,
      "grad_norm": 0.966309191007683,
      "learning_rate": 1.351914982122159e-06,
      "loss": 0.2622663497924805,
      "num_input_tokens_seen": 369862968,
      "step": 22585,
      "train_runtime": 43510.3339,
      "train_tokens_per_second": 8500.578
    },
    {
      "epoch": 1.5210072717479126,
      "grad_norm": 1.1122274860697148,
      "learning_rate": 1.3501073208489823e-06,
      "loss": 0.2935344696044922,
      "num_input_tokens_seen": 369944888,
      "step": 22590,
      "train_runtime": 43519.9289,
      "train_tokens_per_second": 8500.586
    },
    {
      "epoch": 1.5213439267438729,
      "grad_norm": 1.062534166180158,
      "learning_rate": 1.3483006802566546e-06,
      "loss": 0.2917946815490723,
      "num_input_tokens_seen": 370026808,
      "step": 22595,
      "train_runtime": 43529.5265,
      "train_tokens_per_second": 8500.593
    },
    {
      "epoch": 1.5216805817398331,
      "grad_norm": 1.3093773914335534,
      "learning_rate": 1.3464950608503957e-06,
      "loss": 0.29051938056945803,
      "num_input_tokens_seen": 370108728,
      "step": 22600,
      "train_runtime": 43539.0232,
      "train_tokens_per_second": 8500.621
    },
    {
      "epoch": 1.5220172367357931,
      "grad_norm": 1.146922533618294,
      "learning_rate": 1.344690463135146e-06,
      "loss": 0.2924743413925171,
      "num_input_tokens_seen": 370190648,
      "step": 22605,
      "train_runtime": 43548.5213,
      "train_tokens_per_second": 8500.648
    },
    {
      "epoch": 1.5223538917317532,
      "grad_norm": 1.140635182780094,
      "learning_rate": 1.3428868876155493e-06,
      "loss": 0.3004067182540894,
      "num_input_tokens_seen": 370271664,
      "step": 22610,
      "train_runtime": 43558.1918,
      "train_tokens_per_second": 8500.621
    },
    {
      "epoch": 1.5226905467277134,
      "grad_norm": 1.0107773437064973,
      "learning_rate": 1.3410843347959745e-06,
      "loss": 0.25601868629455565,
      "num_input_tokens_seen": 370353584,
      "step": 22615,
      "train_runtime": 43567.6949,
      "train_tokens_per_second": 8500.647
    },
    {
      "epoch": 1.5230272017236737,
      "grad_norm": 1.1265452765483395,
      "learning_rate": 1.3392828051804979e-06,
      "loss": 0.3010770559310913,
      "num_input_tokens_seen": 370435504,
      "step": 22620,
      "train_runtime": 43577.199,
      "train_tokens_per_second": 8500.673
    },
    {
      "epoch": 1.5233638567196337,
      "grad_norm": 1.0157606781683068,
      "learning_rate": 1.3374822992729114e-06,
      "loss": 0.28783907890319826,
      "num_input_tokens_seen": 370517424,
      "step": 22625,
      "train_runtime": 43586.7024,
      "train_tokens_per_second": 8500.699
    },
    {
      "epoch": 1.5237005117155937,
      "grad_norm": 1.1845324283963883,
      "learning_rate": 1.3356828175767212e-06,
      "loss": 0.28117187023162843,
      "num_input_tokens_seen": 370599344,
      "step": 22630,
      "train_runtime": 43596.2059,
      "train_tokens_per_second": 8500.725
    },
    {
      "epoch": 1.524037166711554,
      "grad_norm": 1.081359771190128,
      "learning_rate": 1.3338843605951462e-06,
      "loss": 0.26826844215393064,
      "num_input_tokens_seen": 370681264,
      "step": 22635,
      "train_runtime": 43605.7088,
      "train_tokens_per_second": 8500.751
    },
    {
      "epoch": 1.5243738217075142,
      "grad_norm": 1.1290032992070105,
      "learning_rate": 1.33208692883112e-06,
      "loss": 0.2995465278625488,
      "num_input_tokens_seen": 370763184,
      "step": 22640,
      "train_runtime": 43615.2183,
      "train_tokens_per_second": 8500.776
    },
    {
      "epoch": 1.5247104767034743,
      "grad_norm": 1.105858260246011,
      "learning_rate": 1.3302905227872876e-06,
      "loss": 0.2792847156524658,
      "num_input_tokens_seen": 370845104,
      "step": 22645,
      "train_runtime": 43624.731,
      "train_tokens_per_second": 8500.8
    },
    {
      "epoch": 1.5250471316994343,
      "grad_norm": 1.0317389213585146,
      "learning_rate": 1.328495142966007e-06,
      "loss": 0.2750540256500244,
      "num_input_tokens_seen": 370927024,
      "step": 22650,
      "train_runtime": 43634.2343,
      "train_tokens_per_second": 8500.826
    },
    {
      "epoch": 1.5253837866953945,
      "grad_norm": 0.9892267764140114,
      "learning_rate": 1.3267007898693552e-06,
      "loss": 0.2678225517272949,
      "num_input_tokens_seen": 371008944,
      "step": 22655,
      "train_runtime": 43643.7375,
      "train_tokens_per_second": 8500.852
    },
    {
      "epoch": 1.5257204416913548,
      "grad_norm": 1.1013176825530646,
      "learning_rate": 1.324907463999111e-06,
      "loss": 0.27844958305358886,
      "num_input_tokens_seen": 371090864,
      "step": 22660,
      "train_runtime": 43653.2425,
      "train_tokens_per_second": 8500.877
    },
    {
      "epoch": 1.5260570966873148,
      "grad_norm": 1.248140260226189,
      "learning_rate": 1.3231151658567776e-06,
      "loss": 0.2942017078399658,
      "num_input_tokens_seen": 371172784,
      "step": 22665,
      "train_runtime": 43662.7784,
      "train_tokens_per_second": 8500.897
    },
    {
      "epoch": 1.5263937516832748,
      "grad_norm": 1.2056049687366943,
      "learning_rate": 1.321323895943563e-06,
      "loss": 0.3053530216217041,
      "num_input_tokens_seen": 371254704,
      "step": 22670,
      "train_runtime": 43672.5373,
      "train_tokens_per_second": 8500.873
    },
    {
      "epoch": 1.526730406679235,
      "grad_norm": 0.8869902028886826,
      "learning_rate": 1.3195336547603905e-06,
      "loss": 0.3113356590270996,
      "num_input_tokens_seen": 371336624,
      "step": 22675,
      "train_runtime": 43682.237,
      "train_tokens_per_second": 8500.861
    },
    {
      "epoch": 1.5270670616751953,
      "grad_norm": 1.185202686631269,
      "learning_rate": 1.3177444428078951e-06,
      "loss": 0.2716797113418579,
      "num_input_tokens_seen": 371418416,
      "step": 22680,
      "train_runtime": 43692.3471,
      "train_tokens_per_second": 8500.766
    },
    {
      "epoch": 1.5274037166711554,
      "grad_norm": 1.2746998497069288,
      "learning_rate": 1.3159562605864245e-06,
      "loss": 0.29289093017578127,
      "num_input_tokens_seen": 371500336,
      "step": 22685,
      "train_runtime": 43701.861,
      "train_tokens_per_second": 8500.79
    },
    {
      "epoch": 1.5277403716671154,
      "grad_norm": 1.2850826424800028,
      "learning_rate": 1.3141691085960385e-06,
      "loss": 0.28373935222625735,
      "num_input_tokens_seen": 371582256,
      "step": 22690,
      "train_runtime": 43711.5168,
      "train_tokens_per_second": 8500.786
    },
    {
      "epoch": 1.5280770266630757,
      "grad_norm": 1.098605134326676,
      "learning_rate": 1.3123829873365073e-06,
      "loss": 0.25984878540039064,
      "num_input_tokens_seen": 371663896,
      "step": 22695,
      "train_runtime": 43721.5567,
      "train_tokens_per_second": 8500.701
    },
    {
      "epoch": 1.528413681659036,
      "grad_norm": 0.9559256754995237,
      "learning_rate": 1.3105978973073136e-06,
      "loss": 0.25278477668762206,
      "num_input_tokens_seen": 371745816,
      "step": 22700,
      "train_runtime": 43731.0872,
      "train_tokens_per_second": 8500.722
    },
    {
      "epoch": 1.528750336654996,
      "grad_norm": 1.058413239500941,
      "learning_rate": 1.3088138390076565e-06,
      "loss": 0.2951428651809692,
      "num_input_tokens_seen": 371827736,
      "step": 22705,
      "train_runtime": 43740.8096,
      "train_tokens_per_second": 8500.705
    },
    {
      "epoch": 1.529086991650956,
      "grad_norm": 1.0528522959369062,
      "learning_rate": 1.3070308129364357e-06,
      "loss": 0.2629739999771118,
      "num_input_tokens_seen": 371909656,
      "step": 22710,
      "train_runtime": 43750.4019,
      "train_tokens_per_second": 8500.714
    },
    {
      "epoch": 1.5294236466469162,
      "grad_norm": 1.1214442748241866,
      "learning_rate": 1.3052488195922736e-06,
      "loss": 0.27296996116638184,
      "num_input_tokens_seen": 371991576,
      "step": 22715,
      "train_runtime": 43759.9098,
      "train_tokens_per_second": 8500.739
    },
    {
      "epoch": 1.5297603016428765,
      "grad_norm": 1.1915000251330927,
      "learning_rate": 1.3034678594734972e-06,
      "loss": 0.2727114200592041,
      "num_input_tokens_seen": 372073496,
      "step": 22720,
      "train_runtime": 43769.4218,
      "train_tokens_per_second": 8500.763
    },
    {
      "epoch": 1.5300969566388365,
      "grad_norm": 1.1370161539525032,
      "learning_rate": 1.3016879330781468e-06,
      "loss": 0.2925343751907349,
      "num_input_tokens_seen": 372155416,
      "step": 22725,
      "train_runtime": 43779.028,
      "train_tokens_per_second": 8500.769
    },
    {
      "epoch": 1.5304336116347965,
      "grad_norm": 1.208413379943275,
      "learning_rate": 1.2999090409039728e-06,
      "loss": 0.2850459814071655,
      "num_input_tokens_seen": 372237336,
      "step": 22730,
      "train_runtime": 43788.6469,
      "train_tokens_per_second": 8500.773
    },
    {
      "epoch": 1.5307702666307568,
      "grad_norm": 1.1339148582424468,
      "learning_rate": 1.2981311834484367e-06,
      "loss": 0.2952540397644043,
      "num_input_tokens_seen": 372319256,
      "step": 22735,
      "train_runtime": 43798.1562,
      "train_tokens_per_second": 8500.797
    },
    {
      "epoch": 1.531106921626717,
      "grad_norm": 1.0189224556525665,
      "learning_rate": 1.296354361208711e-06,
      "loss": 0.277122950553894,
      "num_input_tokens_seen": 372401176,
      "step": 22740,
      "train_runtime": 43807.6483,
      "train_tokens_per_second": 8500.826
    },
    {
      "epoch": 1.531443576622677,
      "grad_norm": 1.084584979508006,
      "learning_rate": 1.294578574681678e-06,
      "loss": 0.3034812927246094,
      "num_input_tokens_seen": 372483096,
      "step": 22745,
      "train_runtime": 43817.1708,
      "train_tokens_per_second": 8500.848
    },
    {
      "epoch": 1.531780231618637,
      "grad_norm": 1.3727238230586765,
      "learning_rate": 1.292803824363929e-06,
      "loss": 0.2749730348587036,
      "num_input_tokens_seen": 372565016,
      "step": 22750,
      "train_runtime": 43826.6801,
      "train_tokens_per_second": 8500.872
    },
    {
      "epoch": 1.5321168866145973,
      "grad_norm": 1.2719380104362283,
      "learning_rate": 1.2910301107517726e-06,
      "loss": 0.29363925457000734,
      "num_input_tokens_seen": 372646936,
      "step": 22755,
      "train_runtime": 43836.1971,
      "train_tokens_per_second": 8500.896
    },
    {
      "epoch": 1.5324535416105576,
      "grad_norm": 1.2671402979062028,
      "learning_rate": 1.2892574343412156e-06,
      "loss": 0.2791313171386719,
      "num_input_tokens_seen": 372728152,
      "step": 22760,
      "train_runtime": 43845.6947,
      "train_tokens_per_second": 8500.907
    },
    {
      "epoch": 1.5327901966065176,
      "grad_norm": 1.1492682940927457,
      "learning_rate": 1.2874857956279858e-06,
      "loss": 0.2878544330596924,
      "num_input_tokens_seen": 372810072,
      "step": 22765,
      "train_runtime": 43855.1975,
      "train_tokens_per_second": 8500.932
    },
    {
      "epoch": 1.5331268516024776,
      "grad_norm": 0.9827751624469278,
      "learning_rate": 1.2857151951075147e-06,
      "loss": 0.2682660579681396,
      "num_input_tokens_seen": 372891992,
      "step": 22770,
      "train_runtime": 43864.705,
      "train_tokens_per_second": 8500.957
    },
    {
      "epoch": 1.5334635065984379,
      "grad_norm": 1.0665022040071932,
      "learning_rate": 1.2839456332749456e-06,
      "loss": 0.29230360984802245,
      "num_input_tokens_seen": 372973912,
      "step": 22775,
      "train_runtime": 43874.2095,
      "train_tokens_per_second": 8500.983
    },
    {
      "epoch": 1.5338001615943981,
      "grad_norm": 1.255551397821721,
      "learning_rate": 1.2821771106251308e-06,
      "loss": 0.30268025398254395,
      "num_input_tokens_seen": 373055832,
      "step": 22780,
      "train_runtime": 43883.7112,
      "train_tokens_per_second": 8501.009
    },
    {
      "epoch": 1.5341368165903582,
      "grad_norm": 1.08656221527542,
      "learning_rate": 1.2804096276526312e-06,
      "loss": 0.2572918891906738,
      "num_input_tokens_seen": 373137752,
      "step": 22785,
      "train_runtime": 43893.2168,
      "train_tokens_per_second": 8501.035
    },
    {
      "epoch": 1.5344734715863182,
      "grad_norm": 1.3186767863405588,
      "learning_rate": 1.2786431848517184e-06,
      "loss": 0.25913515090942385,
      "num_input_tokens_seen": 373219192,
      "step": 22790,
      "train_runtime": 43903.5958,
      "train_tokens_per_second": 8500.88
    },
    {
      "epoch": 1.5348101265822784,
      "grad_norm": 1.17155051038691,
      "learning_rate": 1.2768777827163725e-06,
      "loss": 0.266420841217041,
      "num_input_tokens_seen": 373301112,
      "step": 22795,
      "train_runtime": 43913.0929,
      "train_tokens_per_second": 8500.907
    },
    {
      "epoch": 1.5351467815782387,
      "grad_norm": 1.0434256841369718,
      "learning_rate": 1.2751134217402811e-06,
      "loss": 0.27203760147094724,
      "num_input_tokens_seen": 373383032,
      "step": 22800,
      "train_runtime": 43922.6643,
      "train_tokens_per_second": 8500.919
    },
    {
      "epoch": 1.5354834365741987,
      "grad_norm": 1.1681343885837017,
      "learning_rate": 1.2733501024168465e-06,
      "loss": 0.27003931999206543,
      "num_input_tokens_seen": 373464952,
      "step": 22805,
      "train_runtime": 43932.3153,
      "train_tokens_per_second": 8500.917
    },
    {
      "epoch": 1.5358200915701588,
      "grad_norm": 1.2116845480876237,
      "learning_rate": 1.2715878252391695e-06,
      "loss": 0.30901920795440674,
      "num_input_tokens_seen": 373546872,
      "step": 22810,
      "train_runtime": 43941.8172,
      "train_tokens_per_second": 8500.943
    },
    {
      "epoch": 1.536156746566119,
      "grad_norm": 1.1175893235318966,
      "learning_rate": 1.269826590700069e-06,
      "loss": 0.29273533821105957,
      "num_input_tokens_seen": 373628792,
      "step": 22815,
      "train_runtime": 43951.324,
      "train_tokens_per_second": 8500.968
    },
    {
      "epoch": 1.5364934015620793,
      "grad_norm": 1.0009113678348562,
      "learning_rate": 1.2680663992920684e-06,
      "loss": 0.28387084007263186,
      "num_input_tokens_seen": 373710712,
      "step": 22820,
      "train_runtime": 43960.8334,
      "train_tokens_per_second": 8500.992
    },
    {
      "epoch": 1.5368300565580393,
      "grad_norm": 1.0958340553290673,
      "learning_rate": 1.2663072515073988e-06,
      "loss": 0.29856162071228026,
      "num_input_tokens_seen": 373792632,
      "step": 22825,
      "train_runtime": 43970.3366,
      "train_tokens_per_second": 8501.018
    },
    {
      "epoch": 1.5371667115539993,
      "grad_norm": 1.0713500437527115,
      "learning_rate": 1.2645491478380002e-06,
      "loss": 0.2793197870254517,
      "num_input_tokens_seen": 373874552,
      "step": 22830,
      "train_runtime": 43979.9451,
      "train_tokens_per_second": 8501.024
    },
    {
      "epoch": 1.5375033665499596,
      "grad_norm": 1.2059293947284555,
      "learning_rate": 1.2627920887755212e-06,
      "loss": 0.2713118553161621,
      "num_input_tokens_seen": 373956048,
      "step": 22835,
      "train_runtime": 43989.9558,
      "train_tokens_per_second": 8500.942
    },
    {
      "epoch": 1.5378400215459198,
      "grad_norm": 1.021232908517482,
      "learning_rate": 1.261036074811317e-06,
      "loss": 0.2792724847793579,
      "num_input_tokens_seen": 374037968,
      "step": 22840,
      "train_runtime": 43999.6226,
      "train_tokens_per_second": 8500.936
    },
    {
      "epoch": 1.5381766765418798,
      "grad_norm": 1.2088799042373872,
      "learning_rate": 1.2592811064364524e-06,
      "loss": 0.2774949550628662,
      "num_input_tokens_seen": 374119888,
      "step": 22845,
      "train_runtime": 44009.3268,
      "train_tokens_per_second": 8500.923
    },
    {
      "epoch": 1.5385133315378399,
      "grad_norm": 0.9893715471775155,
      "learning_rate": 1.2575271841416963e-06,
      "loss": 0.29388771057128904,
      "num_input_tokens_seen": 374201528,
      "step": 22850,
      "train_runtime": 44019.401,
      "train_tokens_per_second": 8500.832
    },
    {
      "epoch": 1.5388499865338001,
      "grad_norm": 1.1389076459856766,
      "learning_rate": 1.2557743084175327e-06,
      "loss": 0.2637117862701416,
      "num_input_tokens_seen": 374283448,
      "step": 22855,
      "train_runtime": 44029.0561,
      "train_tokens_per_second": 8500.828
    },
    {
      "epoch": 1.5391866415297604,
      "grad_norm": 1.1271044961787278,
      "learning_rate": 1.254022479754141e-06,
      "loss": 0.2881939172744751,
      "num_input_tokens_seen": 374365368,
      "step": 22860,
      "train_runtime": 44038.5679,
      "train_tokens_per_second": 8500.852
    },
    {
      "epoch": 1.5395232965257204,
      "grad_norm": 1.0944167003452792,
      "learning_rate": 1.252271698641419e-06,
      "loss": 0.27046918869018555,
      "num_input_tokens_seen": 374447288,
      "step": 22865,
      "train_runtime": 44048.1003,
      "train_tokens_per_second": 8500.873
    },
    {
      "epoch": 1.5398599515216804,
      "grad_norm": 1.1653391976781031,
      "learning_rate": 1.2505219655689659e-06,
      "loss": 0.28942179679870605,
      "num_input_tokens_seen": 374529208,
      "step": 22870,
      "train_runtime": 44057.7748,
      "train_tokens_per_second": 8500.865
    },
    {
      "epoch": 1.5401966065176407,
      "grad_norm": 1.0035742962980683,
      "learning_rate": 1.2487732810260877e-06,
      "loss": 0.28796982765197754,
      "num_input_tokens_seen": 374611128,
      "step": 22875,
      "train_runtime": 44067.4269,
      "train_tokens_per_second": 8500.862
    },
    {
      "epoch": 1.540533261513601,
      "grad_norm": 1.0317745352365837,
      "learning_rate": 1.2470256455018027e-06,
      "loss": 0.25850939750671387,
      "num_input_tokens_seen": 374693048,
      "step": 22880,
      "train_runtime": 44076.9367,
      "train_tokens_per_second": 8500.887
    },
    {
      "epoch": 1.540869916509561,
      "grad_norm": 1.0208151023310406,
      "learning_rate": 1.2452790594848257e-06,
      "loss": 0.27174038887023927,
      "num_input_tokens_seen": 374774968,
      "step": 22885,
      "train_runtime": 44086.4402,
      "train_tokens_per_second": 8500.912
    },
    {
      "epoch": 1.541206571505521,
      "grad_norm": 1.2675449647015138,
      "learning_rate": 1.2435335234635892e-06,
      "loss": 0.3196163892745972,
      "num_input_tokens_seen": 374856888,
      "step": 22890,
      "train_runtime": 44095.9486,
      "train_tokens_per_second": 8500.937
    },
    {
      "epoch": 1.5415432265014812,
      "grad_norm": 1.136404306282998,
      "learning_rate": 1.2417890379262221e-06,
      "loss": 0.2770135164260864,
      "num_input_tokens_seen": 374938808,
      "step": 22895,
      "train_runtime": 44105.4525,
      "train_tokens_per_second": 8500.963
    },
    {
      "epoch": 1.5418798814974415,
      "grad_norm": 1.25937417509291,
      "learning_rate": 1.2400456033605672e-06,
      "loss": 0.2759985446929932,
      "num_input_tokens_seen": 375020728,
      "step": 22900,
      "train_runtime": 44114.9597,
      "train_tokens_per_second": 8500.988
    },
    {
      "epoch": 1.5422165364934015,
      "grad_norm": 1.1632556444412572,
      "learning_rate": 1.2383032202541706e-06,
      "loss": 0.29490966796875,
      "num_input_tokens_seen": 375102648,
      "step": 22905,
      "train_runtime": 44124.4726,
      "train_tokens_per_second": 8501.011
    },
    {
      "epoch": 1.5425531914893615,
      "grad_norm": 1.1075506140695648,
      "learning_rate": 1.2365618890942827e-06,
      "loss": 0.2975841283798218,
      "num_input_tokens_seen": 375184568,
      "step": 22910,
      "train_runtime": 44133.9864,
      "train_tokens_per_second": 8501.035
    },
    {
      "epoch": 1.5428898464853218,
      "grad_norm": 1.2057127379262274,
      "learning_rate": 1.2348216103678618e-06,
      "loss": 0.28368778228759767,
      "num_input_tokens_seen": 375266488,
      "step": 22915,
      "train_runtime": 44143.4944,
      "train_tokens_per_second": 8501.06
    },
    {
      "epoch": 1.543226501481282,
      "grad_norm": 1.0782791317865275,
      "learning_rate": 1.2330823845615713e-06,
      "loss": 0.28462269306182864,
      "num_input_tokens_seen": 375348408,
      "step": 22920,
      "train_runtime": 44152.9981,
      "train_tokens_per_second": 8501.085
    },
    {
      "epoch": 1.543563156477242,
      "grad_norm": 1.3132007292217618,
      "learning_rate": 1.2313442121617785e-06,
      "loss": 0.3027054786682129,
      "num_input_tokens_seen": 375430328,
      "step": 22925,
      "train_runtime": 44162.5012,
      "train_tokens_per_second": 8501.111
    },
    {
      "epoch": 1.543899811473202,
      "grad_norm": 3.7192361911790845,
      "learning_rate": 1.229607093654563e-06,
      "loss": 0.28672099113464355,
      "num_input_tokens_seen": 375512248,
      "step": 22930,
      "train_runtime": 44172.0082,
      "train_tokens_per_second": 8501.136
    },
    {
      "epoch": 1.5442364664691623,
      "grad_norm": 0.8627892369728252,
      "learning_rate": 1.2278710295256974e-06,
      "loss": 0.265485954284668,
      "num_input_tokens_seen": 375594168,
      "step": 22935,
      "train_runtime": 44181.5089,
      "train_tokens_per_second": 8501.162
    },
    {
      "epoch": 1.5445731214651226,
      "grad_norm": 0.9975495459341882,
      "learning_rate": 1.2261360202606725e-06,
      "loss": 0.26546463966369627,
      "num_input_tokens_seen": 375676088,
      "step": 22940,
      "train_runtime": 44191.013,
      "train_tokens_per_second": 8501.188
    },
    {
      "epoch": 1.5449097764610826,
      "grad_norm": 1.1137866442427997,
      "learning_rate": 1.2244020663446731e-06,
      "loss": 0.2833407878875732,
      "num_input_tokens_seen": 375758008,
      "step": 22945,
      "train_runtime": 44200.5139,
      "train_tokens_per_second": 8501.214
    },
    {
      "epoch": 1.5452464314570427,
      "grad_norm": 1.2434111023379528,
      "learning_rate": 1.222669168262598e-06,
      "loss": 0.2776118278503418,
      "num_input_tokens_seen": 375839928,
      "step": 22950,
      "train_runtime": 44210.0152,
      "train_tokens_per_second": 8501.24
    },
    {
      "epoch": 1.545583086453003,
      "grad_norm": 1.0071422468950029,
      "learning_rate": 1.2209373264990448e-06,
      "loss": 0.2864145517349243,
      "num_input_tokens_seen": 375921544,
      "step": 22955,
      "train_runtime": 44219.9405,
      "train_tokens_per_second": 8501.177
    },
    {
      "epoch": 1.5459197414489632,
      "grad_norm": 0.9325737307054234,
      "learning_rate": 1.2192065415383176e-06,
      "loss": 0.28579139709472656,
      "num_input_tokens_seen": 376003464,
      "step": 22960,
      "train_runtime": 44229.4472,
      "train_tokens_per_second": 8501.202
    },
    {
      "epoch": 1.5462563964449232,
      "grad_norm": 1.1754521655662016,
      "learning_rate": 1.217476813864425e-06,
      "loss": 0.3436045408248901,
      "num_input_tokens_seen": 376085384,
      "step": 22965,
      "train_runtime": 44238.9501,
      "train_tokens_per_second": 8501.228
    },
    {
      "epoch": 1.5465930514408834,
      "grad_norm": 1.0325952869803092,
      "learning_rate": 1.2157481439610801e-06,
      "loss": 0.24822657108306884,
      "num_input_tokens_seen": 376167304,
      "step": 22970,
      "train_runtime": 44248.5659,
      "train_tokens_per_second": 8501.232
    },
    {
      "epoch": 1.5469297064368437,
      "grad_norm": 1.0811790314565708,
      "learning_rate": 1.2140205323116976e-06,
      "loss": 0.2973485946655273,
      "num_input_tokens_seen": 376249224,
      "step": 22975,
      "train_runtime": 44258.2062,
      "train_tokens_per_second": 8501.231
    },
    {
      "epoch": 1.5472663614328037,
      "grad_norm": 1.0300753140165821,
      "learning_rate": 1.2122939793994043e-06,
      "loss": 0.2734852313995361,
      "num_input_tokens_seen": 376331144,
      "step": 22980,
      "train_runtime": 44267.8211,
      "train_tokens_per_second": 8501.235
    },
    {
      "epoch": 1.5476030164287637,
      "grad_norm": 1.0631282356519725,
      "learning_rate": 1.2105684857070182e-06,
      "loss": 0.27679440975189207,
      "num_input_tokens_seen": 376413064,
      "step": 22985,
      "train_runtime": 44277.435,
      "train_tokens_per_second": 8501.239
    },
    {
      "epoch": 1.547939671424724,
      "grad_norm": 1.369666644732004,
      "learning_rate": 1.2088440517170729e-06,
      "loss": 0.298726487159729,
      "num_input_tokens_seen": 376494984,
      "step": 22990,
      "train_runtime": 44286.9499,
      "train_tokens_per_second": 8501.262
    },
    {
      "epoch": 1.5482763264206842,
      "grad_norm": 1.114467739011022,
      "learning_rate": 1.2071206779117994e-06,
      "loss": 0.27312796115875243,
      "num_input_tokens_seen": 376576904,
      "step": 22995,
      "train_runtime": 44296.5747,
      "train_tokens_per_second": 8501.265
    },
    {
      "epoch": 1.5486129814166443,
      "grad_norm": 1.2661707430513658,
      "learning_rate": 1.2053983647731337e-06,
      "loss": 0.27147536277770995,
      "num_input_tokens_seen": 376658824,
      "step": 23000,
      "train_runtime": 44306.1226,
      "train_tokens_per_second": 8501.282
    },
    {
      "epoch": 1.5489496364126043,
      "grad_norm": 0.9534953652169184,
      "learning_rate": 1.2036771127827152e-06,
      "loss": 0.2871167421340942,
      "num_input_tokens_seen": 376740744,
      "step": 23005,
      "train_runtime": 44315.7036,
      "train_tokens_per_second": 8501.292
    },
    {
      "epoch": 1.5492862914085646,
      "grad_norm": 1.138449529316284,
      "learning_rate": 1.2019569224218869e-06,
      "loss": 0.311629056930542,
      "num_input_tokens_seen": 376822664,
      "step": 23010,
      "train_runtime": 44325.3346,
      "train_tokens_per_second": 8501.293
    },
    {
      "epoch": 1.5496229464045248,
      "grad_norm": 0.9337273884216901,
      "learning_rate": 1.2002377941716936e-06,
      "loss": 0.2629313230514526,
      "num_input_tokens_seen": 376904584,
      "step": 23015,
      "train_runtime": 44334.8432,
      "train_tokens_per_second": 8501.318
    },
    {
      "epoch": 1.5499596014004848,
      "grad_norm": 1.5350784627087084,
      "learning_rate": 1.1985197285128853e-06,
      "loss": 0.2763652324676514,
      "num_input_tokens_seen": 376986504,
      "step": 23020,
      "train_runtime": 44344.4769,
      "train_tokens_per_second": 8501.318
    },
    {
      "epoch": 1.5502962563964449,
      "grad_norm": 1.063946428091197,
      "learning_rate": 1.1968027259259107e-06,
      "loss": 0.29704627990722654,
      "num_input_tokens_seen": 377068424,
      "step": 23025,
      "train_runtime": 44354.0354,
      "train_tokens_per_second": 8501.333
    },
    {
      "epoch": 1.5506329113924051,
      "grad_norm": 0.9774072909421417,
      "learning_rate": 1.1950867868909293e-06,
      "loss": 0.24588255882263182,
      "num_input_tokens_seen": 377150344,
      "step": 23030,
      "train_runtime": 44363.569,
      "train_tokens_per_second": 8501.353
    },
    {
      "epoch": 1.5509695663883654,
      "grad_norm": 1.2979000805194432,
      "learning_rate": 1.1933719118877923e-06,
      "loss": 0.31828866004943845,
      "num_input_tokens_seen": 377232264,
      "step": 23035,
      "train_runtime": 44373.1252,
      "train_tokens_per_second": 8501.368
    },
    {
      "epoch": 1.5513062213843254,
      "grad_norm": 0.9309671180058491,
      "learning_rate": 1.1916581013960632e-06,
      "loss": 0.2587330102920532,
      "num_input_tokens_seen": 377314184,
      "step": 23040,
      "train_runtime": 44382.7666,
      "train_tokens_per_second": 8501.367
    },
    {
      "epoch": 1.5516428763802854,
      "grad_norm": 1.3287340550772653,
      "learning_rate": 1.189945355895002e-06,
      "loss": 0.28154711723327636,
      "num_input_tokens_seen": 377396104,
      "step": 23045,
      "train_runtime": 44392.4469,
      "train_tokens_per_second": 8501.358
    },
    {
      "epoch": 1.5519795313762457,
      "grad_norm": 1.2214646247412557,
      "learning_rate": 1.1882336758635727e-06,
      "loss": 0.24556937217712402,
      "num_input_tokens_seen": 377478024,
      "step": 23050,
      "train_runtime": 44402.1122,
      "train_tokens_per_second": 8501.353
    },
    {
      "epoch": 1.552316186372206,
      "grad_norm": 1.3568557264513776,
      "learning_rate": 1.1865230617804412e-06,
      "loss": 0.33971514701843264,
      "num_input_tokens_seen": 377559864,
      "step": 23055,
      "train_runtime": 44412.3024,
      "train_tokens_per_second": 8501.245
    },
    {
      "epoch": 1.552652841368166,
      "grad_norm": 0.9883343179578433,
      "learning_rate": 1.1848135141239753e-06,
      "loss": 0.27654294967651366,
      "num_input_tokens_seen": 377641784,
      "step": 23060,
      "train_runtime": 44421.8225,
      "train_tokens_per_second": 8501.267
    },
    {
      "epoch": 1.552989496364126,
      "grad_norm": 1.102908351096398,
      "learning_rate": 1.1831050333722438e-06,
      "loss": 0.26790504455566405,
      "num_input_tokens_seen": 377723704,
      "step": 23065,
      "train_runtime": 44431.504,
      "train_tokens_per_second": 8501.259
    },
    {
      "epoch": 1.5533261513600862,
      "grad_norm": 1.0496885957121074,
      "learning_rate": 1.1813976200030191e-06,
      "loss": 0.2597496509552002,
      "num_input_tokens_seen": 377805624,
      "step": 23070,
      "train_runtime": 44441.0127,
      "train_tokens_per_second": 8501.283
    },
    {
      "epoch": 1.5536628063560465,
      "grad_norm": 1.0359861128390822,
      "learning_rate": 1.1796912744937717e-06,
      "loss": 0.28036651611328123,
      "num_input_tokens_seen": 377887544,
      "step": 23075,
      "train_runtime": 44450.5235,
      "train_tokens_per_second": 8501.307
    },
    {
      "epoch": 1.5539994613520065,
      "grad_norm": 1.0475125269945518,
      "learning_rate": 1.1779859973216802e-06,
      "loss": 0.28643574714660647,
      "num_input_tokens_seen": 377969464,
      "step": 23080,
      "train_runtime": 44460.0322,
      "train_tokens_per_second": 8501.331
    },
    {
      "epoch": 1.5543361163479665,
      "grad_norm": 1.1903806557799408,
      "learning_rate": 1.1762817889636142e-06,
      "loss": 0.26949305534362794,
      "num_input_tokens_seen": 378050304,
      "step": 23085,
      "train_runtime": 44469.6903,
      "train_tokens_per_second": 8501.303
    },
    {
      "epoch": 1.5546727713439268,
      "grad_norm": 1.2859324225942714,
      "learning_rate": 1.1745786498961542e-06,
      "loss": 0.2914459228515625,
      "num_input_tokens_seen": 378132224,
      "step": 23090,
      "train_runtime": 44479.1985,
      "train_tokens_per_second": 8501.327
    },
    {
      "epoch": 1.555009426339887,
      "grad_norm": 1.0155764554616848,
      "learning_rate": 1.1728765805955766e-06,
      "loss": 0.25040545463562014,
      "num_input_tokens_seen": 378214144,
      "step": 23095,
      "train_runtime": 44488.7095,
      "train_tokens_per_second": 8501.351
    },
    {
      "epoch": 1.555346081335847,
      "grad_norm": 1.260317019640958,
      "learning_rate": 1.1711755815378595e-06,
      "loss": 0.31939024925231935,
      "num_input_tokens_seen": 378296064,
      "step": 23100,
      "train_runtime": 44498.2356,
      "train_tokens_per_second": 8501.372
    },
    {
      "epoch": 1.555682736331807,
      "grad_norm": 1.1735036523210516,
      "learning_rate": 1.1694756531986818e-06,
      "loss": 0.2918237686157227,
      "num_input_tokens_seen": 378377984,
      "step": 23105,
      "train_runtime": 44507.7479,
      "train_tokens_per_second": 8501.396
    },
    {
      "epoch": 1.5560193913277673,
      "grad_norm": 1.0169598835713525,
      "learning_rate": 1.1677767960534232e-06,
      "loss": 0.28102545738220214,
      "num_input_tokens_seen": 378459400,
      "step": 23110,
      "train_runtime": 44517.2596,
      "train_tokens_per_second": 8501.408
    },
    {
      "epoch": 1.5563560463237276,
      "grad_norm": 1.0655186192657913,
      "learning_rate": 1.166079010577163e-06,
      "loss": 0.2771064281463623,
      "num_input_tokens_seen": 378541320,
      "step": 23115,
      "train_runtime": 44526.7637,
      "train_tokens_per_second": 8501.433
    },
    {
      "epoch": 1.5566927013196876,
      "grad_norm": 1.1358760899601008,
      "learning_rate": 1.1643822972446823e-06,
      "loss": 0.30643367767333984,
      "num_input_tokens_seen": 378623240,
      "step": 23120,
      "train_runtime": 44536.2707,
      "train_tokens_per_second": 8501.458
    },
    {
      "epoch": 1.5570293563156477,
      "grad_norm": 0.9792776532575561,
      "learning_rate": 1.1626866565304594e-06,
      "loss": 0.27242064476013184,
      "num_input_tokens_seen": 378705160,
      "step": 23125,
      "train_runtime": 44545.7812,
      "train_tokens_per_second": 8501.482
    },
    {
      "epoch": 1.557366011311608,
      "grad_norm": 1.0577240567154915,
      "learning_rate": 1.16099208890868e-06,
      "loss": 0.2929515361785889,
      "num_input_tokens_seen": 378787080,
      "step": 23130,
      "train_runtime": 44555.3066,
      "train_tokens_per_second": 8501.503
    },
    {
      "epoch": 1.5577026663075682,
      "grad_norm": 1.2531105559333433,
      "learning_rate": 1.1592985948532187e-06,
      "loss": 0.2719587326049805,
      "num_input_tokens_seen": 378868848,
      "step": 23135,
      "train_runtime": 44565.318,
      "train_tokens_per_second": 8501.428
    },
    {
      "epoch": 1.5580393213035282,
      "grad_norm": 1.209219351397273,
      "learning_rate": 1.1576061748376594e-06,
      "loss": 0.26745309829711916,
      "num_input_tokens_seen": 378950768,
      "step": 23140,
      "train_runtime": 44574.8349,
      "train_tokens_per_second": 8501.451
    },
    {
      "epoch": 1.5583759762994882,
      "grad_norm": 1.1606637556285468,
      "learning_rate": 1.1559148293352805e-06,
      "loss": 0.25669350624084475,
      "num_input_tokens_seen": 379032688,
      "step": 23145,
      "train_runtime": 44584.3393,
      "train_tokens_per_second": 8501.476
    },
    {
      "epoch": 1.5587126312954485,
      "grad_norm": 1.1265609382460489,
      "learning_rate": 1.154224558819062e-06,
      "loss": 0.27748961448669435,
      "num_input_tokens_seen": 379114608,
      "step": 23150,
      "train_runtime": 44593.8489,
      "train_tokens_per_second": 8501.5
    },
    {
      "epoch": 1.5590492862914087,
      "grad_norm": 1.0121288074329722,
      "learning_rate": 1.1525353637616821e-06,
      "loss": 0.26909971237182617,
      "num_input_tokens_seen": 379196528,
      "step": 23155,
      "train_runtime": 44603.3634,
      "train_tokens_per_second": 8501.523
    },
    {
      "epoch": 1.5593859412873687,
      "grad_norm": 1.2626907584010185,
      "learning_rate": 1.1508472446355196e-06,
      "loss": 0.2838627099990845,
      "num_input_tokens_seen": 379278448,
      "step": 23160,
      "train_runtime": 44612.868,
      "train_tokens_per_second": 8501.548
    },
    {
      "epoch": 1.5597225962833288,
      "grad_norm": 1.0823051788389797,
      "learning_rate": 1.1491602019126501e-06,
      "loss": 0.3008756160736084,
      "num_input_tokens_seen": 379360368,
      "step": 23165,
      "train_runtime": 44622.3794,
      "train_tokens_per_second": 8501.572
    },
    {
      "epoch": 1.560059251279289,
      "grad_norm": 1.0911323736213097,
      "learning_rate": 1.1474742360648515e-06,
      "loss": 0.2952786207199097,
      "num_input_tokens_seen": 379442288,
      "step": 23170,
      "train_runtime": 44632.0257,
      "train_tokens_per_second": 8501.57
    },
    {
      "epoch": 1.5603959062752493,
      "grad_norm": 1.0718058167126798,
      "learning_rate": 1.1457893475635968e-06,
      "loss": 0.29765868186950684,
      "num_input_tokens_seen": 379524208,
      "step": 23175,
      "train_runtime": 44641.5849,
      "train_tokens_per_second": 8501.585
    },
    {
      "epoch": 1.5607325612712093,
      "grad_norm": 1.0176436205624828,
      "learning_rate": 1.144105536880063e-06,
      "loss": 0.2726754665374756,
      "num_input_tokens_seen": 379606128,
      "step": 23180,
      "train_runtime": 44651.1939,
      "train_tokens_per_second": 8501.59
    },
    {
      "epoch": 1.5610692162671693,
      "grad_norm": 1.112253510446376,
      "learning_rate": 1.1424228044851176e-06,
      "loss": 0.2530678272247314,
      "num_input_tokens_seen": 379688048,
      "step": 23185,
      "train_runtime": 44660.7836,
      "train_tokens_per_second": 8501.598
    },
    {
      "epoch": 1.5614058712631296,
      "grad_norm": 0.9607953335097564,
      "learning_rate": 1.1407411508493355e-06,
      "loss": 0.2942652225494385,
      "num_input_tokens_seen": 379769968,
      "step": 23190,
      "train_runtime": 44670.3798,
      "train_tokens_per_second": 8501.606
    },
    {
      "epoch": 1.5617425262590898,
      "grad_norm": 1.164308578906528,
      "learning_rate": 1.1390605764429846e-06,
      "loss": 0.2689234733581543,
      "num_input_tokens_seen": 379851888,
      "step": 23195,
      "train_runtime": 44679.8911,
      "train_tokens_per_second": 8501.63
    },
    {
      "epoch": 1.5620791812550499,
      "grad_norm": 1.1503410796169578,
      "learning_rate": 1.1373810817360314e-06,
      "loss": 0.27048044204711913,
      "num_input_tokens_seen": 379933808,
      "step": 23200,
      "train_runtime": 44689.4137,
      "train_tokens_per_second": 8501.651
    },
    {
      "epoch": 1.5624158362510099,
      "grad_norm": 1.131117332204639,
      "learning_rate": 1.135702667198142e-06,
      "loss": 0.28995776176452637,
      "num_input_tokens_seen": 380015728,
      "step": 23205,
      "train_runtime": 44699.0561,
      "train_tokens_per_second": 8501.65
    },
    {
      "epoch": 1.5627524912469701,
      "grad_norm": 1.2194889826462125,
      "learning_rate": 1.134025333298679e-06,
      "loss": 0.2583333730697632,
      "num_input_tokens_seen": 380097648,
      "step": 23210,
      "train_runtime": 44708.7075,
      "train_tokens_per_second": 8501.647
    },
    {
      "epoch": 1.5630891462429304,
      "grad_norm": 1.100293490166426,
      "learning_rate": 1.1323490805067045e-06,
      "loss": 0.27681469917297363,
      "num_input_tokens_seen": 380179568,
      "step": 23215,
      "train_runtime": 44718.2928,
      "train_tokens_per_second": 8501.657
    },
    {
      "epoch": 1.5634258012388904,
      "grad_norm": 1.1040323537632113,
      "learning_rate": 1.1306739092909757e-06,
      "loss": 0.25775976181030275,
      "num_input_tokens_seen": 380261488,
      "step": 23220,
      "train_runtime": 44727.8096,
      "train_tokens_per_second": 8501.679
    },
    {
      "epoch": 1.5637624562348504,
      "grad_norm": 1.1947817089840984,
      "learning_rate": 1.1289998201199493e-06,
      "loss": 0.2692866802215576,
      "num_input_tokens_seen": 380343408,
      "step": 23225,
      "train_runtime": 44737.3317,
      "train_tokens_per_second": 8501.701
    },
    {
      "epoch": 1.5640991112308107,
      "grad_norm": 1.0280722072671844,
      "learning_rate": 1.1273268134617816e-06,
      "loss": 0.3001739740371704,
      "num_input_tokens_seen": 380425328,
      "step": 23230,
      "train_runtime": 44747.0265,
      "train_tokens_per_second": 8501.69
    },
    {
      "epoch": 1.564435766226771,
      "grad_norm": 1.2080031594709986,
      "learning_rate": 1.1256548897843189e-06,
      "loss": 0.29253787994384767,
      "num_input_tokens_seen": 380507248,
      "step": 23235,
      "train_runtime": 44756.6668,
      "train_tokens_per_second": 8501.689
    },
    {
      "epoch": 1.564772421222731,
      "grad_norm": 1.1408973398613607,
      "learning_rate": 1.1239840495551136e-06,
      "loss": 0.2642754077911377,
      "num_input_tokens_seen": 380589168,
      "step": 23240,
      "train_runtime": 44766.3416,
      "train_tokens_per_second": 8501.681
    },
    {
      "epoch": 1.565109076218691,
      "grad_norm": 1.3154052369570242,
      "learning_rate": 1.1223142932414087e-06,
      "loss": 0.34589567184448244,
      "num_input_tokens_seen": 380671088,
      "step": 23245,
      "train_runtime": 44775.9958,
      "train_tokens_per_second": 8501.678
    },
    {
      "epoch": 1.5654457312146512,
      "grad_norm": 0.9282638718461919,
      "learning_rate": 1.120645621310147e-06,
      "loss": 0.28694982528686525,
      "num_input_tokens_seen": 380753008,
      "step": 23250,
      "train_runtime": 44785.7074,
      "train_tokens_per_second": 8501.663
    },
    {
      "epoch": 1.5657823862106115,
      "grad_norm": 1.4467659132559805,
      "learning_rate": 1.1189780342279666e-06,
      "loss": 0.28979203701019285,
      "num_input_tokens_seen": 380833544,
      "step": 23255,
      "train_runtime": 44795.4441,
      "train_tokens_per_second": 8501.613
    },
    {
      "epoch": 1.5661190412065715,
      "grad_norm": 1.6328339876245608,
      "learning_rate": 1.1173115324612033e-06,
      "loss": 0.3159745693206787,
      "num_input_tokens_seen": 380915464,
      "step": 23260,
      "train_runtime": 44805.2123,
      "train_tokens_per_second": 8501.588
    },
    {
      "epoch": 1.5664556962025316,
      "grad_norm": 1.137618583278796,
      "learning_rate": 1.1156461164758891e-06,
      "loss": 0.2859102010726929,
      "num_input_tokens_seen": 380997384,
      "step": 23265,
      "train_runtime": 44814.889,
      "train_tokens_per_second": 8501.58
    },
    {
      "epoch": 1.5667923511984918,
      "grad_norm": 1.1256800575933534,
      "learning_rate": 1.1139817867377522e-06,
      "loss": 0.23657603263854982,
      "num_input_tokens_seen": 381079304,
      "step": 23270,
      "train_runtime": 44824.4703,
      "train_tokens_per_second": 8501.591
    },
    {
      "epoch": 1.567129006194452,
      "grad_norm": 1.1453243359709482,
      "learning_rate": 1.1123185437122153e-06,
      "loss": 0.2838107109069824,
      "num_input_tokens_seen": 381161224,
      "step": 23275,
      "train_runtime": 44834.1372,
      "train_tokens_per_second": 8501.585
    },
    {
      "epoch": 1.567465661190412,
      "grad_norm": 1.0941583793502954,
      "learning_rate": 1.1106563878644038e-06,
      "loss": 0.2697242259979248,
      "num_input_tokens_seen": 381243144,
      "step": 23280,
      "train_runtime": 44843.8752,
      "train_tokens_per_second": 8501.566
    },
    {
      "epoch": 1.5678023161863721,
      "grad_norm": 1.1900440281698494,
      "learning_rate": 1.1089953196591286e-06,
      "loss": 0.2903543472290039,
      "num_input_tokens_seen": 381325064,
      "step": 23285,
      "train_runtime": 44853.5878,
      "train_tokens_per_second": 8501.551
    },
    {
      "epoch": 1.5681389711823324,
      "grad_norm": 1.1722096807463875,
      "learning_rate": 1.107335339560906e-06,
      "loss": 0.3063366889953613,
      "num_input_tokens_seen": 381406984,
      "step": 23290,
      "train_runtime": 44863.2876,
      "train_tokens_per_second": 8501.539
    },
    {
      "epoch": 1.5684756261782926,
      "grad_norm": 1.319935199685491,
      "learning_rate": 1.1056764480339427e-06,
      "loss": 0.29465622901916505,
      "num_input_tokens_seen": 381488904,
      "step": 23295,
      "train_runtime": 44872.982,
      "train_tokens_per_second": 8501.528
    },
    {
      "epoch": 1.5688122811742526,
      "grad_norm": 1.279128283804008,
      "learning_rate": 1.1040186455421425e-06,
      "loss": 0.32643096446990966,
      "num_input_tokens_seen": 381570824,
      "step": 23300,
      "train_runtime": 44882.8266,
      "train_tokens_per_second": 8501.488
    },
    {
      "epoch": 1.5691489361702127,
      "grad_norm": 1.056495428211009,
      "learning_rate": 1.1023619325491052e-06,
      "loss": 0.27594969272613523,
      "num_input_tokens_seen": 381652744,
      "step": 23305,
      "train_runtime": 44892.366,
      "train_tokens_per_second": 8501.507
    },
    {
      "epoch": 1.569485591166173,
      "grad_norm": 0.9491857822857843,
      "learning_rate": 1.1007063095181248e-06,
      "loss": 0.26450135707855227,
      "num_input_tokens_seen": 381734664,
      "step": 23310,
      "train_runtime": 44901.8832,
      "train_tokens_per_second": 8501.529
    },
    {
      "epoch": 1.5698222461621332,
      "grad_norm": 1.0496794055606344,
      "learning_rate": 1.0990517769121905e-06,
      "loss": 0.2983846187591553,
      "num_input_tokens_seen": 381816320,
      "step": 23315,
      "train_runtime": 44911.7958,
      "train_tokens_per_second": 8501.471
    },
    {
      "epoch": 1.5701589011580932,
      "grad_norm": 1.3165785828393062,
      "learning_rate": 1.0973983351939876e-06,
      "loss": 0.29865097999572754,
      "num_input_tokens_seen": 381898240,
      "step": 23320,
      "train_runtime": 44921.4185,
      "train_tokens_per_second": 8501.473
    },
    {
      "epoch": 1.5704955561540532,
      "grad_norm": 1.2240043770053701,
      "learning_rate": 1.095745984825895e-06,
      "loss": 0.2940565347671509,
      "num_input_tokens_seen": 381980160,
      "step": 23325,
      "train_runtime": 44930.9344,
      "train_tokens_per_second": 8501.496
    },
    {
      "epoch": 1.5708322111500135,
      "grad_norm": 1.0810996882716464,
      "learning_rate": 1.0940947262699902e-06,
      "loss": 0.2859734058380127,
      "num_input_tokens_seen": 382062080,
      "step": 23330,
      "train_runtime": 44940.956,
      "train_tokens_per_second": 8501.423
    },
    {
      "epoch": 1.5711688661459737,
      "grad_norm": 1.0738392282954217,
      "learning_rate": 1.092444559988038e-06,
      "loss": 0.2721406936645508,
      "num_input_tokens_seen": 382144000,
      "step": 23335,
      "train_runtime": 44950.5876,
      "train_tokens_per_second": 8501.424
    },
    {
      "epoch": 1.5715055211419338,
      "grad_norm": 1.0803525336637725,
      "learning_rate": 1.0907954864415048e-06,
      "loss": 0.26363511085510255,
      "num_input_tokens_seen": 382225920,
      "step": 23340,
      "train_runtime": 44960.1106,
      "train_tokens_per_second": 8501.445
    },
    {
      "epoch": 1.5718421761378938,
      "grad_norm": 1.2091557081135333,
      "learning_rate": 1.089147506091549e-06,
      "loss": 0.2760855436325073,
      "num_input_tokens_seen": 382307840,
      "step": 23345,
      "train_runtime": 44969.7284,
      "train_tokens_per_second": 8501.449
    },
    {
      "epoch": 1.572178831133854,
      "grad_norm": 1.3524522208206915,
      "learning_rate": 1.0875006193990217e-06,
      "loss": 0.2993258714675903,
      "num_input_tokens_seen": 382389760,
      "step": 23350,
      "train_runtime": 44979.2305,
      "train_tokens_per_second": 8501.474
    },
    {
      "epoch": 1.5725154861298143,
      "grad_norm": 1.2204989999729008,
      "learning_rate": 1.0858548268244706e-06,
      "loss": 0.2509489059448242,
      "num_input_tokens_seen": 382471680,
      "step": 23355,
      "train_runtime": 44988.8636,
      "train_tokens_per_second": 8501.475
    },
    {
      "epoch": 1.5728521411257743,
      "grad_norm": 1.015153471580095,
      "learning_rate": 1.0842101288281359e-06,
      "loss": 0.264686131477356,
      "num_input_tokens_seen": 382553600,
      "step": 23360,
      "train_runtime": 44998.5089,
      "train_tokens_per_second": 8501.473
    },
    {
      "epoch": 1.5731887961217343,
      "grad_norm": 1.2117687796950949,
      "learning_rate": 1.082566525869952e-06,
      "loss": 0.2751550436019897,
      "num_input_tokens_seen": 382635520,
      "step": 23365,
      "train_runtime": 45008.0163,
      "train_tokens_per_second": 8501.497
    },
    {
      "epoch": 1.5735254511176946,
      "grad_norm": 1.1164465864299866,
      "learning_rate": 1.0809240184095476e-06,
      "loss": 0.2911424160003662,
      "num_input_tokens_seen": 382717440,
      "step": 23370,
      "train_runtime": 45017.5157,
      "train_tokens_per_second": 8501.523
    },
    {
      "epoch": 1.5738621061136548,
      "grad_norm": 1.1012015370822812,
      "learning_rate": 1.0792826069062429e-06,
      "loss": 0.2723315954208374,
      "num_input_tokens_seen": 382799360,
      "step": 23375,
      "train_runtime": 45027.0233,
      "train_tokens_per_second": 8501.547
    },
    {
      "epoch": 1.5741987611096149,
      "grad_norm": 1.0131687842913972,
      "learning_rate": 1.0776422918190576e-06,
      "loss": 0.2979328155517578,
      "num_input_tokens_seen": 382881248,
      "step": 23380,
      "train_runtime": 45037.2942,
      "train_tokens_per_second": 8501.427
    },
    {
      "epoch": 1.574535416105575,
      "grad_norm": 1.2420289203082344,
      "learning_rate": 1.0760030736066952e-06,
      "loss": 0.2966400146484375,
      "num_input_tokens_seen": 382963168,
      "step": 23385,
      "train_runtime": 45046.8008,
      "train_tokens_per_second": 8501.451
    },
    {
      "epoch": 1.5748720711015352,
      "grad_norm": 1.0561267785511395,
      "learning_rate": 1.0743649527275619e-06,
      "loss": 0.2602482795715332,
      "num_input_tokens_seen": 383045088,
      "step": 23390,
      "train_runtime": 45056.3032,
      "train_tokens_per_second": 8501.476
    },
    {
      "epoch": 1.5752087260974954,
      "grad_norm": 1.180651864985584,
      "learning_rate": 1.0727279296397514e-06,
      "loss": 0.2928131580352783,
      "num_input_tokens_seen": 383127008,
      "step": 23395,
      "train_runtime": 45065.86,
      "train_tokens_per_second": 8501.491
    },
    {
      "epoch": 1.5755453810934554,
      "grad_norm": 1.168512013960508,
      "learning_rate": 1.0710920048010526e-06,
      "loss": 0.29563264846801757,
      "num_input_tokens_seen": 383208928,
      "step": 23400,
      "train_runtime": 45075.453,
      "train_tokens_per_second": 8501.499
    },
    {
      "epoch": 1.5758820360894155,
      "grad_norm": 1.1872117874054737,
      "learning_rate": 1.0694571786689462e-06,
      "loss": 0.29703340530395506,
      "num_input_tokens_seen": 383290848,
      "step": 23405,
      "train_runtime": 45085.2096,
      "train_tokens_per_second": 8501.476
    },
    {
      "epoch": 1.5762186910853757,
      "grad_norm": 1.41846145223946,
      "learning_rate": 1.067823451700606e-06,
      "loss": 0.29056758880615235,
      "num_input_tokens_seen": 383372768,
      "step": 23410,
      "train_runtime": 45095.0252,
      "train_tokens_per_second": 8501.443
    },
    {
      "epoch": 1.576555346081336,
      "grad_norm": 1.2123141729897153,
      "learning_rate": 1.066190824352899e-06,
      "loss": 0.2783070087432861,
      "num_input_tokens_seen": 383454688,
      "step": 23415,
      "train_runtime": 45104.5258,
      "train_tokens_per_second": 8501.468
    },
    {
      "epoch": 1.576892001077296,
      "grad_norm": 1.0436939958074685,
      "learning_rate": 1.0645592970823837e-06,
      "loss": 0.26841566562652586,
      "num_input_tokens_seen": 383536608,
      "step": 23420,
      "train_runtime": 45114.1825,
      "train_tokens_per_second": 8501.464
    },
    {
      "epoch": 1.577228656073256,
      "grad_norm": 1.3224093078505466,
      "learning_rate": 1.0629288703453105e-06,
      "loss": 0.3038398027420044,
      "num_input_tokens_seen": 383618528,
      "step": 23425,
      "train_runtime": 45123.8522,
      "train_tokens_per_second": 8501.458
    },
    {
      "epoch": 1.5775653110692163,
      "grad_norm": 0.9951483804574928,
      "learning_rate": 1.061299544597627e-06,
      "loss": 0.28993375301361085,
      "num_input_tokens_seen": 383700448,
      "step": 23430,
      "train_runtime": 45133.8008,
      "train_tokens_per_second": 8501.399
    },
    {
      "epoch": 1.5779019660651765,
      "grad_norm": 1.0611098082348114,
      "learning_rate": 1.0596713202949637e-06,
      "loss": 0.25359716415405276,
      "num_input_tokens_seen": 383782368,
      "step": 23435,
      "train_runtime": 45143.4844,
      "train_tokens_per_second": 8501.39
    },
    {
      "epoch": 1.5782386210611365,
      "grad_norm": 1.2446448946672217,
      "learning_rate": 1.0580441978926527e-06,
      "loss": 0.2891294002532959,
      "num_input_tokens_seen": 383864288,
      "step": 23440,
      "train_runtime": 45153.0831,
      "train_tokens_per_second": 8501.397
    },
    {
      "epoch": 1.5785752760570966,
      "grad_norm": 1.1209716354948482,
      "learning_rate": 1.056418177845711e-06,
      "loss": 0.2604016065597534,
      "num_input_tokens_seen": 383945040,
      "step": 23445,
      "train_runtime": 45162.7226,
      "train_tokens_per_second": 8501.371
    },
    {
      "epoch": 1.5789119310530568,
      "grad_norm": 0.9532346206298307,
      "learning_rate": 1.0547932606088506e-06,
      "loss": 0.27610054016113283,
      "num_input_tokens_seen": 384026960,
      "step": 23450,
      "train_runtime": 45172.2307,
      "train_tokens_per_second": 8501.395
    },
    {
      "epoch": 1.579248586049017,
      "grad_norm": 1.1167584795842045,
      "learning_rate": 1.053169446636475e-06,
      "loss": 0.2620047092437744,
      "num_input_tokens_seen": 384108880,
      "step": 23455,
      "train_runtime": 45181.9324,
      "train_tokens_per_second": 8501.382
    },
    {
      "epoch": 1.579585241044977,
      "grad_norm": 1.1794144221962575,
      "learning_rate": 1.0515467363826782e-06,
      "loss": 0.25177998542785646,
      "num_input_tokens_seen": 384190800,
      "step": 23460,
      "train_runtime": 45191.5997,
      "train_tokens_per_second": 8501.376
    },
    {
      "epoch": 1.5799218960409371,
      "grad_norm": 1.2846347723265221,
      "learning_rate": 1.0499251303012454e-06,
      "loss": 0.2626774549484253,
      "num_input_tokens_seen": 384272720,
      "step": 23465,
      "train_runtime": 45201.327,
      "train_tokens_per_second": 8501.359
    },
    {
      "epoch": 1.5802585510368974,
      "grad_norm": 1.1389934033173563,
      "learning_rate": 1.0483046288456544e-06,
      "loss": 0.2822685718536377,
      "num_input_tokens_seen": 384354640,
      "step": 23470,
      "train_runtime": 45210.9891,
      "train_tokens_per_second": 8501.354
    },
    {
      "epoch": 1.5805952060328576,
      "grad_norm": 1.4827595396946394,
      "learning_rate": 1.0466852324690719e-06,
      "loss": 0.23718514442443847,
      "num_input_tokens_seen": 384436560,
      "step": 23475,
      "train_runtime": 45220.5044,
      "train_tokens_per_second": 8501.377
    },
    {
      "epoch": 1.5809318610288177,
      "grad_norm": 0.9461499279330142,
      "learning_rate": 1.0450669416243596e-06,
      "loss": 0.2729912519454956,
      "num_input_tokens_seen": 384518480,
      "step": 23480,
      "train_runtime": 45230.2933,
      "train_tokens_per_second": 8501.348
    },
    {
      "epoch": 1.5812685160247777,
      "grad_norm": 1.0192881227952286,
      "learning_rate": 1.0434497567640639e-06,
      "loss": 0.2990957975387573,
      "num_input_tokens_seen": 384600400,
      "step": 23485,
      "train_runtime": 45239.7914,
      "train_tokens_per_second": 8501.374
    },
    {
      "epoch": 1.581605171020738,
      "grad_norm": 1.0836516729175958,
      "learning_rate": 1.0418336783404282e-06,
      "loss": 0.2673347949981689,
      "num_input_tokens_seen": 384682176,
      "step": 23490,
      "train_runtime": 45249.2923,
      "train_tokens_per_second": 8501.397
    },
    {
      "epoch": 1.5819418260166982,
      "grad_norm": 1.0999490275486183,
      "learning_rate": 1.0402187068053825e-06,
      "loss": 0.27590317726135255,
      "num_input_tokens_seen": 384764096,
      "step": 23495,
      "train_runtime": 45258.7997,
      "train_tokens_per_second": 8501.421
    },
    {
      "epoch": 1.5822784810126582,
      "grad_norm": 1.024273695657375,
      "learning_rate": 1.0386048426105495e-06,
      "loss": 0.2607422351837158,
      "num_input_tokens_seen": 384846016,
      "step": 23500,
      "train_runtime": 45268.598,
      "train_tokens_per_second": 8501.39
    },
    {
      "epoch": 1.5826151360086183,
      "grad_norm": 1.0649242074199523,
      "learning_rate": 1.0369920862072396e-06,
      "loss": 0.2565386533737183,
      "num_input_tokens_seen": 384927136,
      "step": 23505,
      "train_runtime": 45278.4058,
      "train_tokens_per_second": 8501.34
    },
    {
      "epoch": 1.5829517910045785,
      "grad_norm": 1.1753966618702818,
      "learning_rate": 1.0353804380464556e-06,
      "loss": 0.2533755302429199,
      "num_input_tokens_seen": 385009056,
      "step": 23510,
      "train_runtime": 45288.1049,
      "train_tokens_per_second": 8501.328
    },
    {
      "epoch": 1.5832884460005388,
      "grad_norm": 0.9908960385583614,
      "learning_rate": 1.0337698985788903e-06,
      "loss": 0.26073219776153567,
      "num_input_tokens_seen": 385090976,
      "step": 23515,
      "train_runtime": 45297.6759,
      "train_tokens_per_second": 8501.341
    },
    {
      "epoch": 1.5836251009964988,
      "grad_norm": 1.3301452557069833,
      "learning_rate": 1.0321604682549247e-06,
      "loss": 0.305767560005188,
      "num_input_tokens_seen": 385172896,
      "step": 23520,
      "train_runtime": 45307.1912,
      "train_tokens_per_second": 8501.363
    },
    {
      "epoch": 1.5839617559924588,
      "grad_norm": 1.1268606099774208,
      "learning_rate": 1.0305521475246311e-06,
      "loss": 0.30964372158050535,
      "num_input_tokens_seen": 385254816,
      "step": 23525,
      "train_runtime": 45316.7023,
      "train_tokens_per_second": 8501.387
    },
    {
      "epoch": 1.584298410988419,
      "grad_norm": 1.0622005873925766,
      "learning_rate": 1.028944936837774e-06,
      "loss": 0.28492140769958496,
      "num_input_tokens_seen": 385336736,
      "step": 23530,
      "train_runtime": 45326.2168,
      "train_tokens_per_second": 8501.41
    },
    {
      "epoch": 1.5846350659843793,
      "grad_norm": 1.018630951788663,
      "learning_rate": 1.0273388366438003e-06,
      "loss": 0.2857835292816162,
      "num_input_tokens_seen": 385418656,
      "step": 23535,
      "train_runtime": 45335.728,
      "train_tokens_per_second": 8501.433
    },
    {
      "epoch": 1.5849717209803393,
      "grad_norm": 1.1243799796970309,
      "learning_rate": 1.0257338473918538e-06,
      "loss": 0.26912431716918944,
      "num_input_tokens_seen": 385500576,
      "step": 23540,
      "train_runtime": 45345.2517,
      "train_tokens_per_second": 8501.454
    },
    {
      "epoch": 1.5853083759762994,
      "grad_norm": 1.11859734746521,
      "learning_rate": 1.0241299695307644e-06,
      "loss": 0.2667213201522827,
      "num_input_tokens_seen": 385582496,
      "step": 23545,
      "train_runtime": 45354.8012,
      "train_tokens_per_second": 8501.47
    },
    {
      "epoch": 1.5856450309722596,
      "grad_norm": 1.082517117703693,
      "learning_rate": 1.0225272035090506e-06,
      "loss": 0.24421443939208984,
      "num_input_tokens_seen": 385664416,
      "step": 23550,
      "train_runtime": 45364.3355,
      "train_tokens_per_second": 8501.489
    },
    {
      "epoch": 1.5859816859682199,
      "grad_norm": 1.162267983691833,
      "learning_rate": 1.0209255497749209e-06,
      "loss": 0.2987729787826538,
      "num_input_tokens_seen": 385746336,
      "step": 23555,
      "train_runtime": 45373.8556,
      "train_tokens_per_second": 8501.511
    },
    {
      "epoch": 1.58631834096418,
      "grad_norm": 1.192305871588599,
      "learning_rate": 1.0193250087762735e-06,
      "loss": 0.29590449333190916,
      "num_input_tokens_seen": 385828256,
      "step": 23560,
      "train_runtime": 45383.3647,
      "train_tokens_per_second": 8501.535
    },
    {
      "epoch": 1.58665499596014,
      "grad_norm": 0.8971649274245843,
      "learning_rate": 1.0177255809606934e-06,
      "loss": 0.286736798286438,
      "num_input_tokens_seen": 385910176,
      "step": 23565,
      "train_runtime": 45392.9938,
      "train_tokens_per_second": 8501.536
    },
    {
      "epoch": 1.5869916509561002,
      "grad_norm": 1.2596097203621754,
      "learning_rate": 1.0161272667754562e-06,
      "loss": 0.2894620180130005,
      "num_input_tokens_seen": 385992096,
      "step": 23570,
      "train_runtime": 45402.8086,
      "train_tokens_per_second": 8501.503
    },
    {
      "epoch": 1.5873283059520604,
      "grad_norm": 0.904303143284443,
      "learning_rate": 1.014530066667524e-06,
      "loss": 0.2777758359909058,
      "num_input_tokens_seen": 386074016,
      "step": 23575,
      "train_runtime": 45412.454,
      "train_tokens_per_second": 8501.501
    },
    {
      "epoch": 1.5876649609480205,
      "grad_norm": 1.259092213878727,
      "learning_rate": 1.0129339810835526e-06,
      "loss": 0.26447505950927735,
      "num_input_tokens_seen": 386155936,
      "step": 23580,
      "train_runtime": 45422.021,
      "train_tokens_per_second": 8501.514
    },
    {
      "epoch": 1.5880016159439805,
      "grad_norm": 1.1020510127322658,
      "learning_rate": 1.0113390104698767e-06,
      "loss": 0.2804602861404419,
      "num_input_tokens_seen": 386237856,
      "step": 23585,
      "train_runtime": 45431.6205,
      "train_tokens_per_second": 8501.521
    },
    {
      "epoch": 1.5883382709399407,
      "grad_norm": 1.1594459951642921,
      "learning_rate": 1.009745155272529e-06,
      "loss": 0.31473143100738527,
      "num_input_tokens_seen": 386319776,
      "step": 23590,
      "train_runtime": 45441.2735,
      "train_tokens_per_second": 8501.517
    },
    {
      "epoch": 1.588674925935901,
      "grad_norm": 0.9843817391696829,
      "learning_rate": 1.0081524159372246e-06,
      "loss": 0.2792079925537109,
      "num_input_tokens_seen": 386401696,
      "step": 23595,
      "train_runtime": 45450.9287,
      "train_tokens_per_second": 8501.514
    },
    {
      "epoch": 1.589011580931861,
      "grad_norm": 1.0306896119085174,
      "learning_rate": 1.0065607929093685e-06,
      "loss": 0.26775689125061036,
      "num_input_tokens_seen": 386483368,
      "step": 23600,
      "train_runtime": 45461.1345,
      "train_tokens_per_second": 8501.402
    },
    {
      "epoch": 1.589348235927821,
      "grad_norm": 1.1776284543381947,
      "learning_rate": 1.0049702866340521e-06,
      "loss": 0.26525781154632566,
      "num_input_tokens_seen": 386565288,
      "step": 23605,
      "train_runtime": 45470.7395,
      "train_tokens_per_second": 8501.408
    },
    {
      "epoch": 1.5896848909237813,
      "grad_norm": 0.9282276605407265,
      "learning_rate": 1.0033808975560556e-06,
      "loss": 0.2592625141143799,
      "num_input_tokens_seen": 386647208,
      "step": 23610,
      "train_runtime": 45480.249,
      "train_tokens_per_second": 8501.431
    },
    {
      "epoch": 1.5900215459197415,
      "grad_norm": 1.0907500389551816,
      "learning_rate": 1.0017926261198473e-06,
      "loss": 0.27492365837097166,
      "num_input_tokens_seen": 386729128,
      "step": 23615,
      "train_runtime": 45489.7855,
      "train_tokens_per_second": 8501.45
    },
    {
      "epoch": 1.5903582009157016,
      "grad_norm": 1.2623080075361162,
      "learning_rate": 1.0002054727695814e-06,
      "loss": 0.2629379749298096,
      "num_input_tokens_seen": 386811048,
      "step": 23620,
      "train_runtime": 45499.677,
      "train_tokens_per_second": 8501.402
    },
    {
      "epoch": 1.5906948559116616,
      "grad_norm": 1.2684126668073208,
      "learning_rate": 9.986194379490993e-07,
      "loss": 0.28175017833709715,
      "num_input_tokens_seen": 386892968,
      "step": 23625,
      "train_runtime": 45509.3867,
      "train_tokens_per_second": 8501.388
    },
    {
      "epoch": 1.5910315109076218,
      "grad_norm": 1.2497806479576872,
      "learning_rate": 9.970345221019345e-07,
      "loss": 0.306184983253479,
      "num_input_tokens_seen": 386974728,
      "step": 23630,
      "train_runtime": 45519.4403,
      "train_tokens_per_second": 8501.307
    },
    {
      "epoch": 1.591368165903582,
      "grad_norm": 1.2756491791261715,
      "learning_rate": 9.954507256712987e-07,
      "loss": 0.27993087768554686,
      "num_input_tokens_seen": 387056440,
      "step": 23635,
      "train_runtime": 45529.457,
      "train_tokens_per_second": 8501.231
    },
    {
      "epoch": 1.5917048208995421,
      "grad_norm": 1.0617856364319453,
      "learning_rate": 9.938680491000991e-07,
      "loss": 0.26629910469055174,
      "num_input_tokens_seen": 387138360,
      "step": 23640,
      "train_runtime": 45538.9609,
      "train_tokens_per_second": 8501.256
    },
    {
      "epoch": 1.5920414758955022,
      "grad_norm": 0.9640215167996303,
      "learning_rate": 9.92286492830925e-07,
      "loss": 0.26482083797454836,
      "num_input_tokens_seen": 387220280,
      "step": 23645,
      "train_runtime": 45548.6673,
      "train_tokens_per_second": 8501.243
    },
    {
      "epoch": 1.5923781308914624,
      "grad_norm": 1.163247619664731,
      "learning_rate": 9.907060573060535e-07,
      "loss": 0.30429229736328123,
      "num_input_tokens_seen": 387302200,
      "step": 23650,
      "train_runtime": 45558.1749,
      "train_tokens_per_second": 8501.267
    },
    {
      "epoch": 1.5927147858874227,
      "grad_norm": 1.0425179919077014,
      "learning_rate": 9.891267429674484e-07,
      "loss": 0.2897973537445068,
      "num_input_tokens_seen": 387384120,
      "step": 23655,
      "train_runtime": 45567.6915,
      "train_tokens_per_second": 8501.289
    },
    {
      "epoch": 1.5930514408833827,
      "grad_norm": 1.0482392682550108,
      "learning_rate": 9.8754855025676e-07,
      "loss": 0.2829550266265869,
      "num_input_tokens_seen": 387466040,
      "step": 23660,
      "train_runtime": 45577.2318,
      "train_tokens_per_second": 8501.307
    },
    {
      "epoch": 1.5933880958793427,
      "grad_norm": 0.9755493396865007,
      "learning_rate": 9.85971479615324e-07,
      "loss": 0.26168217658996584,
      "num_input_tokens_seen": 387547960,
      "step": 23665,
      "train_runtime": 45586.7399,
      "train_tokens_per_second": 8501.331
    },
    {
      "epoch": 1.593724750875303,
      "grad_norm": 0.9777966597598119,
      "learning_rate": 9.843955314841647e-07,
      "loss": 0.2483842372894287,
      "num_input_tokens_seen": 387629880,
      "step": 23670,
      "train_runtime": 45596.2478,
      "train_tokens_per_second": 8501.355
    },
    {
      "epoch": 1.5940614058712632,
      "grad_norm": 1.0579678192635693,
      "learning_rate": 9.82820706303989e-07,
      "loss": 0.27978200912475587,
      "num_input_tokens_seen": 387711800,
      "step": 23675,
      "train_runtime": 45605.7429,
      "train_tokens_per_second": 8501.381
    },
    {
      "epoch": 1.5943980608672232,
      "grad_norm": 1.096933457113597,
      "learning_rate": 9.812470045151952e-07,
      "loss": 0.30229945182800294,
      "num_input_tokens_seen": 387793720,
      "step": 23680,
      "train_runtime": 45615.2401,
      "train_tokens_per_second": 8501.407
    },
    {
      "epoch": 1.5947347158631833,
      "grad_norm": 0.992445109746781,
      "learning_rate": 9.79674426557859e-07,
      "loss": 0.2599113702774048,
      "num_input_tokens_seen": 387875432,
      "step": 23685,
      "train_runtime": 45625.2298,
      "train_tokens_per_second": 8501.337
    },
    {
      "epoch": 1.5950713708591435,
      "grad_norm": 1.0004577265793275,
      "learning_rate": 9.781029728717513e-07,
      "loss": 0.26121954917907714,
      "num_input_tokens_seen": 387957352,
      "step": 23690,
      "train_runtime": 45634.9344,
      "train_tokens_per_second": 8501.324
    },
    {
      "epoch": 1.5954080258551038,
      "grad_norm": 1.02629526680162,
      "learning_rate": 9.765326438963218e-07,
      "loss": 0.260677695274353,
      "num_input_tokens_seen": 388039272,
      "step": 23695,
      "train_runtime": 45644.4435,
      "train_tokens_per_second": 8501.347
    },
    {
      "epoch": 1.5957446808510638,
      "grad_norm": 0.9634048523076317,
      "learning_rate": 9.749634400707087e-07,
      "loss": 0.2573985576629639,
      "num_input_tokens_seen": 388121192,
      "step": 23700,
      "train_runtime": 45653.9449,
      "train_tokens_per_second": 8501.373
    },
    {
      "epoch": 1.5960813358470238,
      "grad_norm": 0.9395173271591507,
      "learning_rate": 9.733953618337344e-07,
      "loss": 0.29003100395202636,
      "num_input_tokens_seen": 388203112,
      "step": 23705,
      "train_runtime": 45663.4474,
      "train_tokens_per_second": 8501.397
    },
    {
      "epoch": 1.596417990842984,
      "grad_norm": 1.1303475222783468,
      "learning_rate": 9.718284096239077e-07,
      "loss": 0.3202070236206055,
      "num_input_tokens_seen": 388285032,
      "step": 23710,
      "train_runtime": 45672.9456,
      "train_tokens_per_second": 8501.423
    },
    {
      "epoch": 1.5967546458389443,
      "grad_norm": 1.012800034073632,
      "learning_rate": 9.702625838794215e-07,
      "loss": 0.25941555500030516,
      "num_input_tokens_seen": 388366952,
      "step": 23715,
      "train_runtime": 45682.4449,
      "train_tokens_per_second": 8501.448
    },
    {
      "epoch": 1.5970913008349044,
      "grad_norm": 1.095618077522545,
      "learning_rate": 9.686978850381535e-07,
      "loss": 0.27076637744903564,
      "num_input_tokens_seen": 388448872,
      "step": 23720,
      "train_runtime": 45691.9526,
      "train_tokens_per_second": 8501.472
    },
    {
      "epoch": 1.5974279558308644,
      "grad_norm": 1.1427906247017923,
      "learning_rate": 9.67134313537666e-07,
      "loss": 0.26061382293701174,
      "num_input_tokens_seen": 388530792,
      "step": 23725,
      "train_runtime": 45701.4549,
      "train_tokens_per_second": 8501.497
    },
    {
      "epoch": 1.5977646108268246,
      "grad_norm": 1.2667696322572501,
      "learning_rate": 9.6557186981521e-07,
      "loss": 0.2552548170089722,
      "num_input_tokens_seen": 388612712,
      "step": 23730,
      "train_runtime": 45710.9758,
      "train_tokens_per_second": 8501.519
    },
    {
      "epoch": 1.5981012658227849,
      "grad_norm": 1.1508230426455588,
      "learning_rate": 9.640105543077133e-07,
      "loss": 0.28630962371826174,
      "num_input_tokens_seen": 388694632,
      "step": 23735,
      "train_runtime": 45720.852,
      "train_tokens_per_second": 8501.474
    },
    {
      "epoch": 1.598437920818745,
      "grad_norm": 1.2452409376050773,
      "learning_rate": 9.624503674517972e-07,
      "loss": 0.3057712078094482,
      "num_input_tokens_seen": 388776552,
      "step": 23740,
      "train_runtime": 45730.3949,
      "train_tokens_per_second": 8501.491
    },
    {
      "epoch": 1.598774575814705,
      "grad_norm": 1.168383290700704,
      "learning_rate": 9.608913096837603e-07,
      "loss": 0.25673842430114746,
      "num_input_tokens_seen": 388858472,
      "step": 23745,
      "train_runtime": 45739.9756,
      "train_tokens_per_second": 8501.502
    },
    {
      "epoch": 1.5991112308106652,
      "grad_norm": 0.9822866417618372,
      "learning_rate": 9.593333814395889e-07,
      "loss": 0.2994598388671875,
      "num_input_tokens_seen": 388940392,
      "step": 23750,
      "train_runtime": 45749.5291,
      "train_tokens_per_second": 8501.517
    },
    {
      "epoch": 1.5994478858066254,
      "grad_norm": 1.049658603675908,
      "learning_rate": 9.577765831549529e-07,
      "loss": 0.2662473678588867,
      "num_input_tokens_seen": 389022312,
      "step": 23755,
      "train_runtime": 45759.1205,
      "train_tokens_per_second": 8501.525
    },
    {
      "epoch": 1.5997845408025855,
      "grad_norm": 1.0479294560711612,
      "learning_rate": 9.562209152652058e-07,
      "loss": 0.26864352226257326,
      "num_input_tokens_seen": 389104232,
      "step": 23760,
      "train_runtime": 45768.7427,
      "train_tokens_per_second": 8501.528
    },
    {
      "epoch": 1.6001211957985455,
      "grad_norm": 1.076620246067862,
      "learning_rate": 9.54666378205385e-07,
      "loss": 0.2800644874572754,
      "num_input_tokens_seen": 389186152,
      "step": 23765,
      "train_runtime": 45778.3606,
      "train_tokens_per_second": 8501.531
    },
    {
      "epoch": 1.6004578507945058,
      "grad_norm": 1.1080793451047952,
      "learning_rate": 9.531129724102117e-07,
      "loss": 0.27475264072418215,
      "num_input_tokens_seen": 389268072,
      "step": 23770,
      "train_runtime": 45788.1208,
      "train_tokens_per_second": 8501.508
    },
    {
      "epoch": 1.600794505790466,
      "grad_norm": 1.1056555999432969,
      "learning_rate": 9.515606983140896e-07,
      "loss": 0.2894002914428711,
      "num_input_tokens_seen": 389349992,
      "step": 23775,
      "train_runtime": 45797.7948,
      "train_tokens_per_second": 8501.501
    },
    {
      "epoch": 1.601131160786426,
      "grad_norm": 1.120563161264481,
      "learning_rate": 9.500095563511119e-07,
      "loss": 0.2947906255722046,
      "num_input_tokens_seen": 389431912,
      "step": 23780,
      "train_runtime": 45807.3003,
      "train_tokens_per_second": 8501.525
    },
    {
      "epoch": 1.601467815782386,
      "grad_norm": 1.121408698515219,
      "learning_rate": 9.484595469550445e-07,
      "loss": 0.26262125968933103,
      "num_input_tokens_seen": 389513832,
      "step": 23785,
      "train_runtime": 45816.7996,
      "train_tokens_per_second": 8501.55
    },
    {
      "epoch": 1.6018044707783463,
      "grad_norm": 1.0378770664565797,
      "learning_rate": 9.469106705593462e-07,
      "loss": 0.26172149181365967,
      "num_input_tokens_seen": 389595752,
      "step": 23790,
      "train_runtime": 45826.3861,
      "train_tokens_per_second": 8501.56
    },
    {
      "epoch": 1.6021411257743066,
      "grad_norm": 1.186669262842113,
      "learning_rate": 9.453629275971549e-07,
      "loss": 0.2591130256652832,
      "num_input_tokens_seen": 389677672,
      "step": 23795,
      "train_runtime": 45836.0151,
      "train_tokens_per_second": 8501.561
    },
    {
      "epoch": 1.6024777807702666,
      "grad_norm": 1.136884051021456,
      "learning_rate": 9.438163185012916e-07,
      "loss": 0.29005908966064453,
      "num_input_tokens_seen": 389759416,
      "step": 23800,
      "train_runtime": 45845.7125,
      "train_tokens_per_second": 8501.546
    },
    {
      "epoch": 1.6028144357662266,
      "grad_norm": 1.1963032906784843,
      "learning_rate": 9.422708437042604e-07,
      "loss": 0.2919311046600342,
      "num_input_tokens_seen": 389841336,
      "step": 23805,
      "train_runtime": 45855.2385,
      "train_tokens_per_second": 8501.566
    },
    {
      "epoch": 1.6031510907621869,
      "grad_norm": 1.0412761240384847,
      "learning_rate": 9.40726503638249e-07,
      "loss": 0.3115460157394409,
      "num_input_tokens_seen": 389923256,
      "step": 23810,
      "train_runtime": 45865.0943,
      "train_tokens_per_second": 8501.525
    },
    {
      "epoch": 1.6034877457581471,
      "grad_norm": 1.1810443821757604,
      "learning_rate": 9.391832987351268e-07,
      "loss": 0.28829245567321776,
      "num_input_tokens_seen": 390005176,
      "step": 23815,
      "train_runtime": 45874.6006,
      "train_tokens_per_second": 8501.549
    },
    {
      "epoch": 1.6038244007541071,
      "grad_norm": 1.3522221937119945,
      "learning_rate": 9.376412294264458e-07,
      "loss": 0.30164685249328616,
      "num_input_tokens_seen": 390087096,
      "step": 23820,
      "train_runtime": 45884.11,
      "train_tokens_per_second": 8501.573
    },
    {
      "epoch": 1.6041610557500672,
      "grad_norm": 1.0528367605769076,
      "learning_rate": 9.361002961434401e-07,
      "loss": 0.25253901481628416,
      "num_input_tokens_seen": 390169016,
      "step": 23825,
      "train_runtime": 45893.6156,
      "train_tokens_per_second": 8501.597
    },
    {
      "epoch": 1.6044977107460274,
      "grad_norm": 1.7159660217466728,
      "learning_rate": 9.345604993170299e-07,
      "loss": 0.31148738861083985,
      "num_input_tokens_seen": 390250936,
      "step": 23830,
      "train_runtime": 45903.1155,
      "train_tokens_per_second": 8501.622
    },
    {
      "epoch": 1.6048343657419877,
      "grad_norm": 1.204118614987814,
      "learning_rate": 9.330218393778101e-07,
      "loss": 0.28018531799316404,
      "num_input_tokens_seen": 390332536,
      "step": 23835,
      "train_runtime": 45913.0383,
      "train_tokens_per_second": 8501.562
    },
    {
      "epoch": 1.6051710207379477,
      "grad_norm": 1.1142911503215878,
      "learning_rate": 9.314843167560656e-07,
      "loss": 0.27665252685546876,
      "num_input_tokens_seen": 390414144,
      "step": 23840,
      "train_runtime": 45922.9601,
      "train_tokens_per_second": 8501.502
    },
    {
      "epoch": 1.6055076757339077,
      "grad_norm": 1.1318913319481865,
      "learning_rate": 9.299479318817578e-07,
      "loss": 0.2566752672195435,
      "num_input_tokens_seen": 390496064,
      "step": 23845,
      "train_runtime": 45932.8083,
      "train_tokens_per_second": 8501.463
    },
    {
      "epoch": 1.605844330729868,
      "grad_norm": 1.0997611044838975,
      "learning_rate": 9.284126851845316e-07,
      "loss": 0.2614766597747803,
      "num_input_tokens_seen": 390577984,
      "step": 23850,
      "train_runtime": 45942.3609,
      "train_tokens_per_second": 8501.478
    },
    {
      "epoch": 1.6061809857258282,
      "grad_norm": 1.0921548993939751,
      "learning_rate": 9.268785770937172e-07,
      "loss": 0.27283554077148436,
      "num_input_tokens_seen": 390659904,
      "step": 23855,
      "train_runtime": 45952.0606,
      "train_tokens_per_second": 8501.467
    },
    {
      "epoch": 1.6065176407217883,
      "grad_norm": 1.3593164756079348,
      "learning_rate": 9.253456080383178e-07,
      "loss": 0.295823335647583,
      "num_input_tokens_seen": 390741824,
      "step": 23860,
      "train_runtime": 45961.7292,
      "train_tokens_per_second": 8501.46
    },
    {
      "epoch": 1.6068542957177483,
      "grad_norm": 1.0791465397576894,
      "learning_rate": 9.238137784470286e-07,
      "loss": 0.31561057567596434,
      "num_input_tokens_seen": 390823744,
      "step": 23865,
      "train_runtime": 45971.2234,
      "train_tokens_per_second": 8501.487
    },
    {
      "epoch": 1.6071909507137085,
      "grad_norm": 1.2373216063766275,
      "learning_rate": 9.222830887482153e-07,
      "loss": 0.2606209278106689,
      "num_input_tokens_seen": 390905664,
      "step": 23870,
      "train_runtime": 45980.7207,
      "train_tokens_per_second": 8501.512
    },
    {
      "epoch": 1.6075276057096688,
      "grad_norm": 1.1691891332750186,
      "learning_rate": 9.207535393699351e-07,
      "loss": 0.2863821268081665,
      "num_input_tokens_seen": 390987584,
      "step": 23875,
      "train_runtime": 45990.2231,
      "train_tokens_per_second": 8501.537
    },
    {
      "epoch": 1.6078642607056288,
      "grad_norm": 1.3769750298531938,
      "learning_rate": 9.192251307399197e-07,
      "loss": 0.27773423194885255,
      "num_input_tokens_seen": 391069504,
      "step": 23880,
      "train_runtime": 45999.7437,
      "train_tokens_per_second": 8501.558
    },
    {
      "epoch": 1.6082009157015889,
      "grad_norm": 1.1723336101137132,
      "learning_rate": 9.176978632855842e-07,
      "loss": 0.2686755657196045,
      "num_input_tokens_seen": 391151424,
      "step": 23885,
      "train_runtime": 46009.2674,
      "train_tokens_per_second": 8501.579
    },
    {
      "epoch": 1.608537570697549,
      "grad_norm": 1.237574955734388,
      "learning_rate": 9.161717374340235e-07,
      "loss": 0.28773415088653564,
      "num_input_tokens_seen": 391233344,
      "step": 23890,
      "train_runtime": 46018.9636,
      "train_tokens_per_second": 8501.568
    },
    {
      "epoch": 1.6088742256935094,
      "grad_norm": 1.2702820670073791,
      "learning_rate": 9.14646753612014e-07,
      "loss": 0.2947193145751953,
      "num_input_tokens_seen": 391315264,
      "step": 23895,
      "train_runtime": 46028.5992,
      "train_tokens_per_second": 8501.568
    },
    {
      "epoch": 1.6092108806894694,
      "grad_norm": 1.0112269300715733,
      "learning_rate": 9.131229122460112e-07,
      "loss": 0.28203320503234863,
      "num_input_tokens_seen": 391397184,
      "step": 23900,
      "train_runtime": 46038.2364,
      "train_tokens_per_second": 8501.568
    },
    {
      "epoch": 1.6095475356854294,
      "grad_norm": 0.9580316832072985,
      "learning_rate": 9.116002137621566e-07,
      "loss": 0.27213234901428224,
      "num_input_tokens_seen": 391479104,
      "step": 23905,
      "train_runtime": 46047.7417,
      "train_tokens_per_second": 8501.592
    },
    {
      "epoch": 1.6098841906813897,
      "grad_norm": 1.1186580396211034,
      "learning_rate": 9.100786585862626e-07,
      "loss": 0.2986690282821655,
      "num_input_tokens_seen": 391561024,
      "step": 23910,
      "train_runtime": 46057.2493,
      "train_tokens_per_second": 8501.615
    },
    {
      "epoch": 1.61022084567735,
      "grad_norm": 1.2646761406496094,
      "learning_rate": 9.085582471438326e-07,
      "loss": 0.29596710205078125,
      "num_input_tokens_seen": 391642944,
      "step": 23915,
      "train_runtime": 46066.7528,
      "train_tokens_per_second": 8501.64
    },
    {
      "epoch": 1.61055750067331,
      "grad_norm": 0.9868681788769119,
      "learning_rate": 9.070389798600394e-07,
      "loss": 0.2938655376434326,
      "num_input_tokens_seen": 391724864,
      "step": 23920,
      "train_runtime": 46076.26,
      "train_tokens_per_second": 8501.664
    },
    {
      "epoch": 1.61089415566927,
      "grad_norm": 1.159943717372718,
      "learning_rate": 9.055208571597451e-07,
      "loss": 0.26012635231018066,
      "num_input_tokens_seen": 391806784,
      "step": 23925,
      "train_runtime": 46085.7611,
      "train_tokens_per_second": 8501.688
    },
    {
      "epoch": 1.6112308106652302,
      "grad_norm": 1.2044984697641412,
      "learning_rate": 9.040038794674872e-07,
      "loss": 0.28842649459838865,
      "num_input_tokens_seen": 391888704,
      "step": 23930,
      "train_runtime": 46095.3389,
      "train_tokens_per_second": 8501.699
    },
    {
      "epoch": 1.6115674656611905,
      "grad_norm": 1.106716630068702,
      "learning_rate": 9.024880472074831e-07,
      "loss": 0.25046167373657224,
      "num_input_tokens_seen": 391970624,
      "step": 23935,
      "train_runtime": 46104.9514,
      "train_tokens_per_second": 8501.703
    },
    {
      "epoch": 1.6119041206571505,
      "grad_norm": 1.177320113113113,
      "learning_rate": 9.009733608036308e-07,
      "loss": 0.26928622722625734,
      "num_input_tokens_seen": 392052544,
      "step": 23940,
      "train_runtime": 46114.6723,
      "train_tokens_per_second": 8501.688
    },
    {
      "epoch": 1.6122407756531107,
      "grad_norm": 1.1505663569388083,
      "learning_rate": 8.994598206795068e-07,
      "loss": 0.26603877544403076,
      "num_input_tokens_seen": 392134464,
      "step": 23945,
      "train_runtime": 46124.3287,
      "train_tokens_per_second": 8501.684
    },
    {
      "epoch": 1.612577430649071,
      "grad_norm": 1.0639053813863266,
      "learning_rate": 8.979474272583672e-07,
      "loss": 0.2724013328552246,
      "num_input_tokens_seen": 392216104,
      "step": 23950,
      "train_runtime": 46134.407,
      "train_tokens_per_second": 8501.596
    },
    {
      "epoch": 1.612914085645031,
      "grad_norm": 1.1159743482827227,
      "learning_rate": 8.964361809631517e-07,
      "loss": 0.2910635471343994,
      "num_input_tokens_seen": 392298024,
      "step": 23955,
      "train_runtime": 46143.9993,
      "train_tokens_per_second": 8501.604
    },
    {
      "epoch": 1.613250740640991,
      "grad_norm": 1.2161412900006374,
      "learning_rate": 8.949260822164707e-07,
      "loss": 0.30412001609802247,
      "num_input_tokens_seen": 392379944,
      "step": 23960,
      "train_runtime": 46153.521,
      "train_tokens_per_second": 8501.625
    },
    {
      "epoch": 1.6135873956369513,
      "grad_norm": 1.016566086142413,
      "learning_rate": 8.934171314406226e-07,
      "loss": 0.28904085159301757,
      "num_input_tokens_seen": 392461864,
      "step": 23965,
      "train_runtime": 46163.2347,
      "train_tokens_per_second": 8501.611
    },
    {
      "epoch": 1.6139240506329116,
      "grad_norm": 1.0931336859102745,
      "learning_rate": 8.919093290575765e-07,
      "loss": 0.2746469020843506,
      "num_input_tokens_seen": 392543600,
      "step": 23970,
      "train_runtime": 46173.2361,
      "train_tokens_per_second": 8501.54
    },
    {
      "epoch": 1.6142607056288716,
      "grad_norm": 1.1835007648480886,
      "learning_rate": 8.904026754889877e-07,
      "loss": 0.2753277063369751,
      "num_input_tokens_seen": 392625520,
      "step": 23975,
      "train_runtime": 46182.7462,
      "train_tokens_per_second": 8501.563
    },
    {
      "epoch": 1.6145973606248316,
      "grad_norm": 1.1882224374674288,
      "learning_rate": 8.888971711561867e-07,
      "loss": 0.29573779106140136,
      "num_input_tokens_seen": 392707440,
      "step": 23980,
      "train_runtime": 46192.2562,
      "train_tokens_per_second": 8501.586
    },
    {
      "epoch": 1.6149340156207919,
      "grad_norm": 1.218055879689815,
      "learning_rate": 8.873928164801821e-07,
      "loss": 0.32491240501403806,
      "num_input_tokens_seen": 392789360,
      "step": 23985,
      "train_runtime": 46201.7775,
      "train_tokens_per_second": 8501.607
    },
    {
      "epoch": 1.6152706706167521,
      "grad_norm": 0.9847269112383216,
      "learning_rate": 8.858896118816624e-07,
      "loss": 0.2841338634490967,
      "num_input_tokens_seen": 392870952,
      "step": 23990,
      "train_runtime": 46211.7018,
      "train_tokens_per_second": 8501.547
    },
    {
      "epoch": 1.6156073256127121,
      "grad_norm": 1.1018900832967424,
      "learning_rate": 8.84387557780994e-07,
      "loss": 0.2804899215698242,
      "num_input_tokens_seen": 392952872,
      "step": 23995,
      "train_runtime": 46221.1994,
      "train_tokens_per_second": 8501.572
    },
    {
      "epoch": 1.6159439806086722,
      "grad_norm": 1.3031057871940823,
      "learning_rate": 8.828866545982196e-07,
      "loss": 0.2811701774597168,
      "num_input_tokens_seen": 393034792,
      "step": 24000,
      "train_runtime": 46230.7067,
      "train_tokens_per_second": 8501.596
    },
    {
      "epoch": 1.6162806356046324,
      "grad_norm": 1.0073100484824367,
      "learning_rate": 8.813869027530664e-07,
      "loss": 0.26002798080444334,
      "num_input_tokens_seen": 393116712,
      "step": 24005,
      "train_runtime": 46240.209,
      "train_tokens_per_second": 8501.621
    },
    {
      "epoch": 1.6166172906005927,
      "grad_norm": 1.2662526615476828,
      "learning_rate": 8.798883026649302e-07,
      "loss": 0.3085346221923828,
      "num_input_tokens_seen": 393198632,
      "step": 24010,
      "train_runtime": 46249.9603,
      "train_tokens_per_second": 8501.599
    },
    {
      "epoch": 1.6169539455965527,
      "grad_norm": 0.9373238859719877,
      "learning_rate": 8.783908547528947e-07,
      "loss": 0.2788339138031006,
      "num_input_tokens_seen": 393280552,
      "step": 24015,
      "train_runtime": 46259.5934,
      "train_tokens_per_second": 8501.6
    },
    {
      "epoch": 1.6172906005925127,
      "grad_norm": 1.1947809716067117,
      "learning_rate": 8.768945594357109e-07,
      "loss": 0.26144990921020506,
      "num_input_tokens_seen": 393362472,
      "step": 24020,
      "train_runtime": 46269.1468,
      "train_tokens_per_second": 8501.615
    },
    {
      "epoch": 1.617627255588473,
      "grad_norm": 1.1879143293838612,
      "learning_rate": 8.753994171318175e-07,
      "loss": 0.2642103672027588,
      "num_input_tokens_seen": 393443736,
      "step": 24025,
      "train_runtime": 46279.1467,
      "train_tokens_per_second": 8501.534
    },
    {
      "epoch": 1.6179639105844332,
      "grad_norm": 1.1024991793029981,
      "learning_rate": 8.739054282593245e-07,
      "loss": 0.3180799722671509,
      "num_input_tokens_seen": 393525656,
      "step": 24030,
      "train_runtime": 46288.6511,
      "train_tokens_per_second": 8501.558
    },
    {
      "epoch": 1.6183005655803933,
      "grad_norm": 1.2397370448930891,
      "learning_rate": 8.724125932360217e-07,
      "loss": 0.2929280996322632,
      "num_input_tokens_seen": 393607576,
      "step": 24035,
      "train_runtime": 46298.1434,
      "train_tokens_per_second": 8501.584
    },
    {
      "epoch": 1.6186372205763533,
      "grad_norm": 1.1713941510368664,
      "learning_rate": 8.70920912479375e-07,
      "loss": 0.30363457202911376,
      "num_input_tokens_seen": 393688904,
      "step": 24040,
      "train_runtime": 46307.9552,
      "train_tokens_per_second": 8501.539
    },
    {
      "epoch": 1.6189738755723135,
      "grad_norm": 1.2195932230183413,
      "learning_rate": 8.694303864065285e-07,
      "loss": 0.2618419885635376,
      "num_input_tokens_seen": 393770824,
      "step": 24045,
      "train_runtime": 46317.4648,
      "train_tokens_per_second": 8501.563
    },
    {
      "epoch": 1.6193105305682738,
      "grad_norm": 1.1295747542737005,
      "learning_rate": 8.679410154343015e-07,
      "loss": 0.2851057529449463,
      "num_input_tokens_seen": 393852744,
      "step": 24050,
      "train_runtime": 46326.9721,
      "train_tokens_per_second": 8501.586
    },
    {
      "epoch": 1.6196471855642338,
      "grad_norm": 1.3579132762537596,
      "learning_rate": 8.664527999791961e-07,
      "loss": 0.3187694072723389,
      "num_input_tokens_seen": 393934664,
      "step": 24055,
      "train_runtime": 46336.4722,
      "train_tokens_per_second": 8501.611
    },
    {
      "epoch": 1.6199838405601938,
      "grad_norm": 1.2695503173205263,
      "learning_rate": 8.649657404573819e-07,
      "loss": 0.28104820251464846,
      "num_input_tokens_seen": 394016584,
      "step": 24060,
      "train_runtime": 46345.9711,
      "train_tokens_per_second": 8501.636
    },
    {
      "epoch": 1.620320495556154,
      "grad_norm": 1.3176199899421015,
      "learning_rate": 8.634798372847148e-07,
      "loss": 0.27482309341430666,
      "num_input_tokens_seen": 394098504,
      "step": 24065,
      "train_runtime": 46355.4745,
      "train_tokens_per_second": 8501.66
    },
    {
      "epoch": 1.6206571505521143,
      "grad_norm": 1.0677024625753018,
      "learning_rate": 8.619950908767189e-07,
      "loss": 0.2554584741592407,
      "num_input_tokens_seen": 394180424,
      "step": 24070,
      "train_runtime": 46364.9677,
      "train_tokens_per_second": 8501.687
    },
    {
      "epoch": 1.6209938055480744,
      "grad_norm": 1.2682667655750384,
      "learning_rate": 8.605115016486016e-07,
      "loss": 0.265307354927063,
      "num_input_tokens_seen": 394262344,
      "step": 24075,
      "train_runtime": 46374.4671,
      "train_tokens_per_second": 8501.712
    },
    {
      "epoch": 1.6213304605440344,
      "grad_norm": 1.2407134718548054,
      "learning_rate": 8.590290700152426e-07,
      "loss": 0.26495649814605715,
      "num_input_tokens_seen": 394344264,
      "step": 24080,
      "train_runtime": 46383.9619,
      "train_tokens_per_second": 8501.737
    },
    {
      "epoch": 1.6216671155399947,
      "grad_norm": 1.1305393721296064,
      "learning_rate": 8.575477963912005e-07,
      "loss": 0.26687600612640383,
      "num_input_tokens_seen": 394426184,
      "step": 24085,
      "train_runtime": 46393.4648,
      "train_tokens_per_second": 8501.762
    },
    {
      "epoch": 1.622003770535955,
      "grad_norm": 1.1552058320990095,
      "learning_rate": 8.560676811907071e-07,
      "loss": 0.2989794731140137,
      "num_input_tokens_seen": 394508104,
      "step": 24090,
      "train_runtime": 46402.9631,
      "train_tokens_per_second": 8501.787
    },
    {
      "epoch": 1.622340425531915,
      "grad_norm": 1.253168905998303,
      "learning_rate": 8.54588724827673e-07,
      "loss": 0.25719547271728516,
      "num_input_tokens_seen": 394590024,
      "step": 24095,
      "train_runtime": 46412.4572,
      "train_tokens_per_second": 8501.813
    },
    {
      "epoch": 1.622677080527875,
      "grad_norm": 1.1971060152985042,
      "learning_rate": 8.531109277156818e-07,
      "loss": 0.2666452646255493,
      "num_input_tokens_seen": 394671944,
      "step": 24100,
      "train_runtime": 46421.9898,
      "train_tokens_per_second": 8501.832
    },
    {
      "epoch": 1.6230137355238352,
      "grad_norm": 1.1491642236499602,
      "learning_rate": 8.516342902679986e-07,
      "loss": 0.2695929765701294,
      "num_input_tokens_seen": 394753864,
      "step": 24105,
      "train_runtime": 46431.7266,
      "train_tokens_per_second": 8501.813
    },
    {
      "epoch": 1.6233503905197955,
      "grad_norm": 1.0177930806613964,
      "learning_rate": 8.501588128975557e-07,
      "loss": 0.28086285591125487,
      "num_input_tokens_seen": 394835784,
      "step": 24110,
      "train_runtime": 46441.3908,
      "train_tokens_per_second": 8501.808
    },
    {
      "epoch": 1.6236870455157555,
      "grad_norm": 1.0321979572268305,
      "learning_rate": 8.486844960169704e-07,
      "loss": 0.23481175899505616,
      "num_input_tokens_seen": 394917704,
      "step": 24115,
      "train_runtime": 46450.9478,
      "train_tokens_per_second": 8501.822
    },
    {
      "epoch": 1.6240237005117155,
      "grad_norm": 1.2337669230081176,
      "learning_rate": 8.472113400385257e-07,
      "loss": 0.2584872245788574,
      "num_input_tokens_seen": 394999624,
      "step": 24120,
      "train_runtime": 46460.5319,
      "train_tokens_per_second": 8501.832
    },
    {
      "epoch": 1.6243603555076758,
      "grad_norm": 1.953044773764776,
      "learning_rate": 8.457393453741886e-07,
      "loss": 0.3044259071350098,
      "num_input_tokens_seen": 395081544,
      "step": 24125,
      "train_runtime": 46470.2147,
      "train_tokens_per_second": 8501.823
    },
    {
      "epoch": 1.624697010503636,
      "grad_norm": 1.04778204634603,
      "learning_rate": 8.442685124355965e-07,
      "loss": 0.2719284534454346,
      "num_input_tokens_seen": 395163464,
      "step": 24130,
      "train_runtime": 46479.8873,
      "train_tokens_per_second": 8501.816
    },
    {
      "epoch": 1.625033665499596,
      "grad_norm": 1.0901794618145126,
      "learning_rate": 8.427988416340633e-07,
      "loss": 0.2915895223617554,
      "num_input_tokens_seen": 395245384,
      "step": 24135,
      "train_runtime": 46489.3871,
      "train_tokens_per_second": 8501.841
    },
    {
      "epoch": 1.625370320495556,
      "grad_norm": 1.2762760955702925,
      "learning_rate": 8.413303333805777e-07,
      "loss": 0.27487645149230955,
      "num_input_tokens_seen": 395327304,
      "step": 24140,
      "train_runtime": 46498.9014,
      "train_tokens_per_second": 8501.863
    },
    {
      "epoch": 1.6257069754915163,
      "grad_norm": 1.255377141614067,
      "learning_rate": 8.398629880858034e-07,
      "loss": 0.28480732440948486,
      "num_input_tokens_seen": 395409224,
      "step": 24145,
      "train_runtime": 46508.4063,
      "train_tokens_per_second": 8501.887
    },
    {
      "epoch": 1.6260436304874766,
      "grad_norm": 1.0634718588483143,
      "learning_rate": 8.383968061600778e-07,
      "loss": 0.2613415002822876,
      "num_input_tokens_seen": 395491144,
      "step": 24150,
      "train_runtime": 46517.9153,
      "train_tokens_per_second": 8501.91
    },
    {
      "epoch": 1.6263802854834366,
      "grad_norm": 1.0748003280828715,
      "learning_rate": 8.36931788013417e-07,
      "loss": 0.26024181842803956,
      "num_input_tokens_seen": 395573064,
      "step": 24155,
      "train_runtime": 46527.4246,
      "train_tokens_per_second": 8501.933
    },
    {
      "epoch": 1.6267169404793966,
      "grad_norm": 0.9559994668707951,
      "learning_rate": 8.354679340555044e-07,
      "loss": 0.27259278297424316,
      "num_input_tokens_seen": 395654984,
      "step": 24160,
      "train_runtime": 46536.9278,
      "train_tokens_per_second": 8501.958
    },
    {
      "epoch": 1.6270535954753569,
      "grad_norm": 1.049330326700021,
      "learning_rate": 8.34005244695707e-07,
      "loss": 0.2819664716720581,
      "num_input_tokens_seen": 395736904,
      "step": 24165,
      "train_runtime": 46546.432,
      "train_tokens_per_second": 8501.982
    },
    {
      "epoch": 1.6273902504713171,
      "grad_norm": 1.0842159892636287,
      "learning_rate": 8.325437203430558e-07,
      "loss": 0.26789886951446534,
      "num_input_tokens_seen": 395818824,
      "step": 24170,
      "train_runtime": 46555.933,
      "train_tokens_per_second": 8502.006
    },
    {
      "epoch": 1.6277269054672772,
      "grad_norm": 1.0058965344705637,
      "learning_rate": 8.310833614062652e-07,
      "loss": 0.25288238525390627,
      "num_input_tokens_seen": 395900744,
      "step": 24175,
      "train_runtime": 46565.4373,
      "train_tokens_per_second": 8502.03
    },
    {
      "epoch": 1.6280635604632372,
      "grad_norm": 1.2262165123461424,
      "learning_rate": 8.296241682937189e-07,
      "loss": 0.2956819534301758,
      "num_input_tokens_seen": 395982664,
      "step": 24180,
      "train_runtime": 46575.0679,
      "train_tokens_per_second": 8502.031
    },
    {
      "epoch": 1.6284002154591974,
      "grad_norm": 1.1925158807652507,
      "learning_rate": 8.281661414134761e-07,
      "loss": 0.28151884078979494,
      "num_input_tokens_seen": 396064584,
      "step": 24185,
      "train_runtime": 46584.6673,
      "train_tokens_per_second": 8502.037
    },
    {
      "epoch": 1.6287368704551577,
      "grad_norm": 1.1742745260481497,
      "learning_rate": 8.267092811732686e-07,
      "loss": 0.28625390529632566,
      "num_input_tokens_seen": 396146504,
      "step": 24190,
      "train_runtime": 46594.2204,
      "train_tokens_per_second": 8502.052
    },
    {
      "epoch": 1.6290735254511177,
      "grad_norm": 1.0129690825115192,
      "learning_rate": 8.252535879805029e-07,
      "loss": 0.28331952095031737,
      "num_input_tokens_seen": 396228424,
      "step": 24195,
      "train_runtime": 46603.865,
      "train_tokens_per_second": 8502.051
    },
    {
      "epoch": 1.6294101804470777,
      "grad_norm": 1.2158273733916944,
      "learning_rate": 8.237990622422576e-07,
      "loss": 0.2688784599304199,
      "num_input_tokens_seen": 396310344,
      "step": 24200,
      "train_runtime": 46613.3673,
      "train_tokens_per_second": 8502.075
    },
    {
      "epoch": 1.629746835443038,
      "grad_norm": 1.1269932361622588,
      "learning_rate": 8.223457043652905e-07,
      "loss": 0.2969967365264893,
      "num_input_tokens_seen": 396392264,
      "step": 24205,
      "train_runtime": 46623.2161,
      "train_tokens_per_second": 8502.036
    },
    {
      "epoch": 1.6300834904389983,
      "grad_norm": 1.0565784442166568,
      "learning_rate": 8.208935147560227e-07,
      "loss": 0.2873749971389771,
      "num_input_tokens_seen": 396473976,
      "step": 24210,
      "train_runtime": 46633.2073,
      "train_tokens_per_second": 8501.967
    },
    {
      "epoch": 1.6304201454349583,
      "grad_norm": 1.1053413697447367,
      "learning_rate": 8.194424938205597e-07,
      "loss": 0.3091753959655762,
      "num_input_tokens_seen": 396555896,
      "step": 24215,
      "train_runtime": 46642.8491,
      "train_tokens_per_second": 8501.966
    },
    {
      "epoch": 1.6307568004309183,
      "grad_norm": 1.1718031582104866,
      "learning_rate": 8.179926419646705e-07,
      "loss": 0.2880044460296631,
      "num_input_tokens_seen": 396637816,
      "step": 24220,
      "train_runtime": 46652.4753,
      "train_tokens_per_second": 8501.967
    },
    {
      "epoch": 1.6310934554268786,
      "grad_norm": 0.9267382922766199,
      "learning_rate": 8.165439595938047e-07,
      "loss": 0.281280517578125,
      "num_input_tokens_seen": 396719736,
      "step": 24225,
      "train_runtime": 46661.9791,
      "train_tokens_per_second": 8501.991
    },
    {
      "epoch": 1.6314301104228388,
      "grad_norm": 1.117461345555827,
      "learning_rate": 8.150964471130801e-07,
      "loss": 0.28657021522521975,
      "num_input_tokens_seen": 396801656,
      "step": 24230,
      "train_runtime": 46671.4727,
      "train_tokens_per_second": 8502.017
    },
    {
      "epoch": 1.6317667654187988,
      "grad_norm": 1.2656411121670033,
      "learning_rate": 8.136501049272899e-07,
      "loss": 0.28889074325561526,
      "num_input_tokens_seen": 396883576,
      "step": 24235,
      "train_runtime": 46680.9685,
      "train_tokens_per_second": 8502.042
    },
    {
      "epoch": 1.6321034204147589,
      "grad_norm": 1.1708576024436517,
      "learning_rate": 8.122049334408983e-07,
      "loss": 0.29955306053161623,
      "num_input_tokens_seen": 396965496,
      "step": 24240,
      "train_runtime": 46690.4737,
      "train_tokens_per_second": 8502.066
    },
    {
      "epoch": 1.6324400754107191,
      "grad_norm": 1.0298549998201432,
      "learning_rate": 8.107609330580429e-07,
      "loss": 0.28322629928588866,
      "num_input_tokens_seen": 397047416,
      "step": 24245,
      "train_runtime": 46699.9718,
      "train_tokens_per_second": 8502.091
    },
    {
      "epoch": 1.6327767304066794,
      "grad_norm": 0.9516891941436798,
      "learning_rate": 8.093181041825332e-07,
      "loss": 0.27047750949859617,
      "num_input_tokens_seen": 397129336,
      "step": 24250,
      "train_runtime": 46709.4765,
      "train_tokens_per_second": 8502.115
    },
    {
      "epoch": 1.6331133854026394,
      "grad_norm": 1.181020596750115,
      "learning_rate": 8.078764472178552e-07,
      "loss": 0.2839754581451416,
      "num_input_tokens_seen": 397211256,
      "step": 24255,
      "train_runtime": 46718.9757,
      "train_tokens_per_second": 8502.14
    },
    {
      "epoch": 1.6334500403985994,
      "grad_norm": 1.0566863316668378,
      "learning_rate": 8.06435962567158e-07,
      "loss": 0.2886344432830811,
      "num_input_tokens_seen": 397292928,
      "step": 24260,
      "train_runtime": 46729.5222,
      "train_tokens_per_second": 8501.969
    },
    {
      "epoch": 1.6337866953945597,
      "grad_norm": 1.2156005468571531,
      "learning_rate": 8.049966506332746e-07,
      "loss": 0.25585269927978516,
      "num_input_tokens_seen": 397374848,
      "step": 24265,
      "train_runtime": 46739.0235,
      "train_tokens_per_second": 8501.993
    },
    {
      "epoch": 1.63412335039052,
      "grad_norm": 0.9920667196440159,
      "learning_rate": 8.035585118186984e-07,
      "loss": 0.2630089044570923,
      "num_input_tokens_seen": 397456768,
      "step": 24270,
      "train_runtime": 46748.5418,
      "train_tokens_per_second": 8502.014
    },
    {
      "epoch": 1.63446000538648,
      "grad_norm": 1.1719010434784256,
      "learning_rate": 8.021215465256038e-07,
      "loss": 0.24803309440612792,
      "num_input_tokens_seen": 397538688,
      "step": 24275,
      "train_runtime": 46758.2309,
      "train_tokens_per_second": 8502.004
    },
    {
      "epoch": 1.63479666038244,
      "grad_norm": 1.1020044408710774,
      "learning_rate": 8.006857551558328e-07,
      "loss": 0.26781392097473145,
      "num_input_tokens_seen": 397620608,
      "step": 24280,
      "train_runtime": 46767.7339,
      "train_tokens_per_second": 8502.029
    },
    {
      "epoch": 1.6351333153784002,
      "grad_norm": 1.1075918211832296,
      "learning_rate": 7.992511381108997e-07,
      "loss": 0.2911205530166626,
      "num_input_tokens_seen": 397702528,
      "step": 24285,
      "train_runtime": 46777.3675,
      "train_tokens_per_second": 8502.029
    },
    {
      "epoch": 1.6354699703743605,
      "grad_norm": 1.1284116911094433,
      "learning_rate": 7.978176957919909e-07,
      "loss": 0.27969090938568114,
      "num_input_tokens_seen": 397784448,
      "step": 24290,
      "train_runtime": 46786.9921,
      "train_tokens_per_second": 8502.031
    },
    {
      "epoch": 1.6358066253703205,
      "grad_norm": 0.9580646877904402,
      "learning_rate": 7.963854285999634e-07,
      "loss": 0.2409038543701172,
      "num_input_tokens_seen": 397866368,
      "step": 24295,
      "train_runtime": 46796.6537,
      "train_tokens_per_second": 8502.026
    },
    {
      "epoch": 1.6361432803662805,
      "grad_norm": 1.0709496991523402,
      "learning_rate": 7.949543369353452e-07,
      "loss": 0.2485736608505249,
      "num_input_tokens_seen": 397948288,
      "step": 24300,
      "train_runtime": 46806.3397,
      "train_tokens_per_second": 8502.017
    },
    {
      "epoch": 1.6364799353622408,
      "grad_norm": 1.0270373864014344,
      "learning_rate": 7.9352442119834e-07,
      "loss": 0.2980198860168457,
      "num_input_tokens_seen": 398030208,
      "step": 24305,
      "train_runtime": 46816.0439,
      "train_tokens_per_second": 8502.004
    },
    {
      "epoch": 1.636816590358201,
      "grad_norm": 1.263311226649055,
      "learning_rate": 7.920956817888148e-07,
      "loss": 0.29578685760498047,
      "num_input_tokens_seen": 398112128,
      "step": 24310,
      "train_runtime": 46825.5533,
      "train_tokens_per_second": 8502.027
    },
    {
      "epoch": 1.637153245354161,
      "grad_norm": 1.1808290441652078,
      "learning_rate": 7.906681191063165e-07,
      "loss": 0.2596719264984131,
      "num_input_tokens_seen": 398194048,
      "step": 24315,
      "train_runtime": 46835.0665,
      "train_tokens_per_second": 8502.049
    },
    {
      "epoch": 1.637489900350121,
      "grad_norm": 1.275726151022546,
      "learning_rate": 7.892417335500541e-07,
      "loss": 0.2646049737930298,
      "num_input_tokens_seen": 398275968,
      "step": 24320,
      "train_runtime": 46844.5707,
      "train_tokens_per_second": 8502.073
    },
    {
      "epoch": 1.6378265553460813,
      "grad_norm": 1.1199365640802112,
      "learning_rate": 7.878165255189146e-07,
      "loss": 0.28508212566375735,
      "num_input_tokens_seen": 398357888,
      "step": 24325,
      "train_runtime": 46854.0725,
      "train_tokens_per_second": 8502.097
    },
    {
      "epoch": 1.6381632103420416,
      "grad_norm": 1.1467485148767047,
      "learning_rate": 7.863924954114522e-07,
      "loss": 0.2552634716033936,
      "num_input_tokens_seen": 398439464,
      "step": 24330,
      "train_runtime": 46863.9707,
      "train_tokens_per_second": 8502.042
    },
    {
      "epoch": 1.6384998653380016,
      "grad_norm": 0.9780829785681736,
      "learning_rate": 7.849696436258919e-07,
      "loss": 0.2759368658065796,
      "num_input_tokens_seen": 398521384,
      "step": 24335,
      "train_runtime": 46873.4797,
      "train_tokens_per_second": 8502.065
    },
    {
      "epoch": 1.6388365203339617,
      "grad_norm": 1.1672568544335098,
      "learning_rate": 7.835479705601307e-07,
      "loss": 0.2924124002456665,
      "num_input_tokens_seen": 398603304,
      "step": 24340,
      "train_runtime": 46882.9885,
      "train_tokens_per_second": 8502.088
    },
    {
      "epoch": 1.639173175329922,
      "grad_norm": 1.056109663613813,
      "learning_rate": 7.821274766117337e-07,
      "loss": 0.28876123428344724,
      "num_input_tokens_seen": 398685224,
      "step": 24345,
      "train_runtime": 46892.4966,
      "train_tokens_per_second": 8502.111
    },
    {
      "epoch": 1.6395098303258822,
      "grad_norm": 1.1984198832588404,
      "learning_rate": 7.807081621779367e-07,
      "loss": 0.29942405223846436,
      "num_input_tokens_seen": 398767144,
      "step": 24350,
      "train_runtime": 46902.0093,
      "train_tokens_per_second": 8502.133
    },
    {
      "epoch": 1.6398464853218422,
      "grad_norm": 1.1625323009352857,
      "learning_rate": 7.792900276556509e-07,
      "loss": 0.3176168918609619,
      "num_input_tokens_seen": 398849064,
      "step": 24355,
      "train_runtime": 46911.5186,
      "train_tokens_per_second": 8502.156
    },
    {
      "epoch": 1.6401831403178022,
      "grad_norm": 1.2195488899691103,
      "learning_rate": 7.778730734414469e-07,
      "loss": 0.2993926048278809,
      "num_input_tokens_seen": 398930984,
      "step": 24360,
      "train_runtime": 46921.0298,
      "train_tokens_per_second": 8502.179
    },
    {
      "epoch": 1.6405197953137625,
      "grad_norm": 1.0457659104020751,
      "learning_rate": 7.764572999315772e-07,
      "loss": 0.27877163887023926,
      "num_input_tokens_seen": 399012904,
      "step": 24365,
      "train_runtime": 46930.5431,
      "train_tokens_per_second": 8502.201
    },
    {
      "epoch": 1.6408564503097227,
      "grad_norm": 1.3172772946393705,
      "learning_rate": 7.750427075219536e-07,
      "loss": 0.30132236480712893,
      "num_input_tokens_seen": 399094824,
      "step": 24370,
      "train_runtime": 46940.2836,
      "train_tokens_per_second": 8502.182
    },
    {
      "epoch": 1.6411931053056827,
      "grad_norm": 1.169454608534556,
      "learning_rate": 7.736292966081655e-07,
      "loss": 0.25836422443389895,
      "num_input_tokens_seen": 399176744,
      "step": 24375,
      "train_runtime": 46950.0023,
      "train_tokens_per_second": 8502.167
    },
    {
      "epoch": 1.6415297603016428,
      "grad_norm": 1.0096928449206783,
      "learning_rate": 7.722170675854673e-07,
      "loss": 0.2715602874755859,
      "num_input_tokens_seen": 399258664,
      "step": 24380,
      "train_runtime": 46959.6937,
      "train_tokens_per_second": 8502.156
    },
    {
      "epoch": 1.641866415297603,
      "grad_norm": 1.069997159644791,
      "learning_rate": 7.70806020848785e-07,
      "loss": 0.26581408977508547,
      "num_input_tokens_seen": 399340584,
      "step": 24385,
      "train_runtime": 46969.1917,
      "train_tokens_per_second": 8502.181
    },
    {
      "epoch": 1.6422030702935633,
      "grad_norm": 1.0268392924935175,
      "learning_rate": 7.693961567927133e-07,
      "loss": 0.2814692497253418,
      "num_input_tokens_seen": 399422504,
      "step": 24390,
      "train_runtime": 46978.6977,
      "train_tokens_per_second": 8502.205
    },
    {
      "epoch": 1.6425397252895233,
      "grad_norm": 1.1788796621070379,
      "learning_rate": 7.679874758115153e-07,
      "loss": 0.26958043575286866,
      "num_input_tokens_seen": 399504424,
      "step": 24395,
      "train_runtime": 46988.1888,
      "train_tokens_per_second": 8502.231
    },
    {
      "epoch": 1.6428763802854833,
      "grad_norm": 1.0704055115208653,
      "learning_rate": 7.66579978299124e-07,
      "loss": 0.29738826751708985,
      "num_input_tokens_seen": 399586344,
      "step": 24400,
      "train_runtime": 46997.6846,
      "train_tokens_per_second": 8502.256
    },
    {
      "epoch": 1.6432130352814436,
      "grad_norm": 1.181155723685576,
      "learning_rate": 7.651736646491447e-07,
      "loss": 0.27281038761138915,
      "num_input_tokens_seen": 399668264,
      "step": 24405,
      "train_runtime": 47007.181,
      "train_tokens_per_second": 8502.281
    },
    {
      "epoch": 1.6435496902774038,
      "grad_norm": 0.9804798435113213,
      "learning_rate": 7.63768535254844e-07,
      "loss": 0.24521489143371583,
      "num_input_tokens_seen": 399750184,
      "step": 24410,
      "train_runtime": 47016.676,
      "train_tokens_per_second": 8502.306
    },
    {
      "epoch": 1.6438863452733639,
      "grad_norm": 1.0471485527803879,
      "learning_rate": 7.623645905091664e-07,
      "loss": 0.297992467880249,
      "num_input_tokens_seen": 399832104,
      "step": 24415,
      "train_runtime": 47026.1752,
      "train_tokens_per_second": 8502.331
    },
    {
      "epoch": 1.6442230002693239,
      "grad_norm": 1.220921160721647,
      "learning_rate": 7.609618308047156e-07,
      "loss": 0.2969163179397583,
      "num_input_tokens_seen": 399914024,
      "step": 24420,
      "train_runtime": 47035.6746,
      "train_tokens_per_second": 8502.355
    },
    {
      "epoch": 1.6445596552652841,
      "grad_norm": 0.96411883355588,
      "learning_rate": 7.595602565337729e-07,
      "loss": 0.25785677433013915,
      "num_input_tokens_seen": 399995944,
      "step": 24425,
      "train_runtime": 47045.1701,
      "train_tokens_per_second": 8502.381
    },
    {
      "epoch": 1.6448963102612444,
      "grad_norm": 1.246757352049753,
      "learning_rate": 7.581598680882829e-07,
      "loss": 0.308620023727417,
      "num_input_tokens_seen": 400077864,
      "step": 24430,
      "train_runtime": 47054.683,
      "train_tokens_per_second": 8502.403
    },
    {
      "epoch": 1.6452329652572044,
      "grad_norm": 1.1358474636567377,
      "learning_rate": 7.567606658598592e-07,
      "loss": 0.2763088226318359,
      "num_input_tokens_seen": 400159784,
      "step": 24435,
      "train_runtime": 47064.2037,
      "train_tokens_per_second": 8502.423
    },
    {
      "epoch": 1.6455696202531644,
      "grad_norm": 1.2047042614479342,
      "learning_rate": 7.553626502397848e-07,
      "loss": 0.2842827796936035,
      "num_input_tokens_seen": 400241704,
      "step": 24440,
      "train_runtime": 47073.9263,
      "train_tokens_per_second": 8502.408
    },
    {
      "epoch": 1.6459062752491247,
      "grad_norm": 1.1689694009381453,
      "learning_rate": 7.539658216190099e-07,
      "loss": 0.2663823127746582,
      "num_input_tokens_seen": 400323624,
      "step": 24445,
      "train_runtime": 47083.6246,
      "train_tokens_per_second": 8502.396
    },
    {
      "epoch": 1.646242930245085,
      "grad_norm": 1.148116822486856,
      "learning_rate": 7.525701803881525e-07,
      "loss": 0.27954413890838625,
      "num_input_tokens_seen": 400405544,
      "step": 24450,
      "train_runtime": 47093.1494,
      "train_tokens_per_second": 8502.416
    },
    {
      "epoch": 1.646579585241045,
      "grad_norm": 1.0886112461639637,
      "learning_rate": 7.511757269375019e-07,
      "loss": 0.2625134944915771,
      "num_input_tokens_seen": 400487464,
      "step": 24455,
      "train_runtime": 47102.7776,
      "train_tokens_per_second": 8502.417
    },
    {
      "epoch": 1.646916240237005,
      "grad_norm": 0.9914417669587412,
      "learning_rate": 7.497824616570087e-07,
      "loss": 0.29954941272735597,
      "num_input_tokens_seen": 400569384,
      "step": 24460,
      "train_runtime": 47112.4965,
      "train_tokens_per_second": 8502.402
    },
    {
      "epoch": 1.6472528952329653,
      "grad_norm": 1.0861596082365812,
      "learning_rate": 7.483903849362995e-07,
      "loss": 0.3166023254394531,
      "num_input_tokens_seen": 400651304,
      "step": 24465,
      "train_runtime": 47122.1317,
      "train_tokens_per_second": 8502.402
    },
    {
      "epoch": 1.6475895502289255,
      "grad_norm": 0.9248319138314769,
      "learning_rate": 7.469994971646594e-07,
      "loss": 0.26682281494140625,
      "num_input_tokens_seen": 400733224,
      "step": 24470,
      "train_runtime": 47131.8536,
      "train_tokens_per_second": 8502.386
    },
    {
      "epoch": 1.6479262052248855,
      "grad_norm": 1.1570571088738366,
      "learning_rate": 7.456097987310495e-07,
      "loss": 0.2714073181152344,
      "num_input_tokens_seen": 400815144,
      "step": 24475,
      "train_runtime": 47141.485,
      "train_tokens_per_second": 8502.387
    },
    {
      "epoch": 1.6482628602208456,
      "grad_norm": 1.079452738546984,
      "learning_rate": 7.442212900240931e-07,
      "loss": 0.305905818939209,
      "num_input_tokens_seen": 400897064,
      "step": 24480,
      "train_runtime": 47150.9944,
      "train_tokens_per_second": 8502.41
    },
    {
      "epoch": 1.6485995152168058,
      "grad_norm": 1.1114037101114744,
      "learning_rate": 7.42833971432082e-07,
      "loss": 0.2689499855041504,
      "num_input_tokens_seen": 400978984,
      "step": 24485,
      "train_runtime": 47160.4997,
      "train_tokens_per_second": 8502.433
    },
    {
      "epoch": 1.648936170212766,
      "grad_norm": 1.1346402030490934,
      "learning_rate": 7.414478433429756e-07,
      "loss": 0.2959397792816162,
      "num_input_tokens_seen": 401060904,
      "step": 24490,
      "train_runtime": 47170.0008,
      "train_tokens_per_second": 8502.457
    },
    {
      "epoch": 1.649272825208726,
      "grad_norm": 0.9879105488390796,
      "learning_rate": 7.400629061444003e-07,
      "loss": 0.24684202671051025,
      "num_input_tokens_seen": 401142824,
      "step": 24495,
      "train_runtime": 47179.4997,
      "train_tokens_per_second": 8502.482
    },
    {
      "epoch": 1.6496094802046861,
      "grad_norm": 0.9793928134874397,
      "learning_rate": 7.386791602236482e-07,
      "loss": 0.275003719329834,
      "num_input_tokens_seen": 401224744,
      "step": 24500,
      "train_runtime": 47188.9961,
      "train_tokens_per_second": 8502.506
    },
    {
      "epoch": 1.6499461352006464,
      "grad_norm": 1.014385444401237,
      "learning_rate": 7.372966059676834e-07,
      "loss": 0.30480117797851564,
      "num_input_tokens_seen": 401306664,
      "step": 24505,
      "train_runtime": 47198.495,
      "train_tokens_per_second": 8502.531
    },
    {
      "epoch": 1.6502827901966066,
      "grad_norm": 1.1542212551172317,
      "learning_rate": 7.359152437631273e-07,
      "loss": 0.30479605197906495,
      "num_input_tokens_seen": 401388584,
      "step": 24510,
      "train_runtime": 47208.0002,
      "train_tokens_per_second": 8502.554
    },
    {
      "epoch": 1.6506194451925666,
      "grad_norm": 1.0190226615252114,
      "learning_rate": 7.345350739962781e-07,
      "loss": 0.26066811084747316,
      "num_input_tokens_seen": 401470504,
      "step": 24515,
      "train_runtime": 47217.5078,
      "train_tokens_per_second": 8502.577
    },
    {
      "epoch": 1.6509561001885267,
      "grad_norm": 1.0495381166944646,
      "learning_rate": 7.331560970530921e-07,
      "loss": 0.2749703645706177,
      "num_input_tokens_seen": 401552424,
      "step": 24520,
      "train_runtime": 47227.2371,
      "train_tokens_per_second": 8502.56
    },
    {
      "epoch": 1.651292755184487,
      "grad_norm": 1.0814499772567854,
      "learning_rate": 7.317783133191986e-07,
      "loss": 0.26789488792419436,
      "num_input_tokens_seen": 401634344,
      "step": 24525,
      "train_runtime": 47236.8046,
      "train_tokens_per_second": 8502.572
    },
    {
      "epoch": 1.6516294101804472,
      "grad_norm": 1.1959743621636945,
      "learning_rate": 7.304017231798893e-07,
      "loss": 0.28021559715270994,
      "num_input_tokens_seen": 401716264,
      "step": 24530,
      "train_runtime": 47246.5778,
      "train_tokens_per_second": 8502.547
    },
    {
      "epoch": 1.6519660651764072,
      "grad_norm": 1.1408995015088645,
      "learning_rate": 7.290263270201231e-07,
      "loss": 0.27730655670166016,
      "num_input_tokens_seen": 401798184,
      "step": 24535,
      "train_runtime": 47256.2791,
      "train_tokens_per_second": 8502.535
    },
    {
      "epoch": 1.6523027201723672,
      "grad_norm": 0.9363443215907664,
      "learning_rate": 7.276521252245261e-07,
      "loss": 0.2732395172119141,
      "num_input_tokens_seen": 401880104,
      "step": 24540,
      "train_runtime": 47266.0864,
      "train_tokens_per_second": 8502.504
    },
    {
      "epoch": 1.6526393751683275,
      "grad_norm": 1.297171327253613,
      "learning_rate": 7.262791181773882e-07,
      "loss": 0.2667700290679932,
      "num_input_tokens_seen": 401962024,
      "step": 24545,
      "train_runtime": 47275.972,
      "train_tokens_per_second": 8502.459
    },
    {
      "epoch": 1.6529760301642877,
      "grad_norm": 1.140657765531615,
      "learning_rate": 7.24907306262666e-07,
      "loss": 0.30754752159118653,
      "num_input_tokens_seen": 402043944,
      "step": 24550,
      "train_runtime": 47285.5878,
      "train_tokens_per_second": 8502.463
    },
    {
      "epoch": 1.6533126851602478,
      "grad_norm": 1.2007909123937532,
      "learning_rate": 7.235366898639856e-07,
      "loss": 0.2977156639099121,
      "num_input_tokens_seen": 402125864,
      "step": 24555,
      "train_runtime": 47295.2582,
      "train_tokens_per_second": 8502.456
    },
    {
      "epoch": 1.6536493401562078,
      "grad_norm": 1.0671050668542992,
      "learning_rate": 7.221672693646309e-07,
      "loss": 0.27275633811950684,
      "num_input_tokens_seen": 402207784,
      "step": 24560,
      "train_runtime": 47304.9249,
      "train_tokens_per_second": 8502.451
    },
    {
      "epoch": 1.653985995152168,
      "grad_norm": 1.3389463689796566,
      "learning_rate": 7.207990451475594e-07,
      "loss": 0.3187251567840576,
      "num_input_tokens_seen": 402289488,
      "step": 24565,
      "train_runtime": 47314.6762,
      "train_tokens_per_second": 8502.425
    },
    {
      "epoch": 1.6543226501481283,
      "grad_norm": 1.2303520474932685,
      "learning_rate": 7.194320175953901e-07,
      "loss": 0.27974677085876465,
      "num_input_tokens_seen": 402371360,
      "step": 24570,
      "train_runtime": 47325.0507,
      "train_tokens_per_second": 8502.291
    },
    {
      "epoch": 1.6546593051440883,
      "grad_norm": 1.1413263535825202,
      "learning_rate": 7.180661870904066e-07,
      "loss": 0.2822734832763672,
      "num_input_tokens_seen": 402453280,
      "step": 24575,
      "train_runtime": 47334.6341,
      "train_tokens_per_second": 8502.3
    },
    {
      "epoch": 1.6549959601400483,
      "grad_norm": 1.0374871698175212,
      "learning_rate": 7.167015540145599e-07,
      "loss": 0.25928311347961425,
      "num_input_tokens_seen": 402535200,
      "step": 24580,
      "train_runtime": 47344.3141,
      "train_tokens_per_second": 8502.292
    },
    {
      "epoch": 1.6553326151360086,
      "grad_norm": 1.0454766355450045,
      "learning_rate": 7.153381187494657e-07,
      "loss": 0.2801440000534058,
      "num_input_tokens_seen": 402617120,
      "step": 24585,
      "train_runtime": 47353.9809,
      "train_tokens_per_second": 8502.287
    },
    {
      "epoch": 1.6556692701319689,
      "grad_norm": 1.0340534814862876,
      "learning_rate": 7.139758816764036e-07,
      "loss": 0.2610173225402832,
      "num_input_tokens_seen": 402699040,
      "step": 24590,
      "train_runtime": 47363.5128,
      "train_tokens_per_second": 8502.305
    },
    {
      "epoch": 1.6560059251279289,
      "grad_norm": 1.3263093042985648,
      "learning_rate": 7.126148431763191e-07,
      "loss": 0.2728844165802002,
      "num_input_tokens_seen": 402780960,
      "step": 24595,
      "train_runtime": 47373.2393,
      "train_tokens_per_second": 8502.289
    },
    {
      "epoch": 1.656342580123889,
      "grad_norm": 0.97924254390705,
      "learning_rate": 7.11255003629821e-07,
      "loss": 0.29476346969604494,
      "num_input_tokens_seen": 402862880,
      "step": 24600,
      "train_runtime": 47382.8729,
      "train_tokens_per_second": 8502.289
    },
    {
      "epoch": 1.6566792351198492,
      "grad_norm": 1.1347634532412565,
      "learning_rate": 7.098963634171874e-07,
      "loss": 0.28133330345153806,
      "num_input_tokens_seen": 402944800,
      "step": 24605,
      "train_runtime": 47392.5702,
      "train_tokens_per_second": 8502.278
    },
    {
      "epoch": 1.6570158901158094,
      "grad_norm": 1.2992060065605862,
      "learning_rate": 7.085389229183537e-07,
      "loss": 0.29391767978668215,
      "num_input_tokens_seen": 403026720,
      "step": 24610,
      "train_runtime": 47402.1943,
      "train_tokens_per_second": 8502.28
    },
    {
      "epoch": 1.6573525451117694,
      "grad_norm": 1.0716938768641804,
      "learning_rate": 7.071826825129264e-07,
      "loss": 0.27213344573974607,
      "num_input_tokens_seen": 403108640,
      "step": 24615,
      "train_runtime": 47411.8098,
      "train_tokens_per_second": 8502.283
    },
    {
      "epoch": 1.6576892001077295,
      "grad_norm": 1.1589804569059228,
      "learning_rate": 7.058276425801735e-07,
      "loss": 0.2852905511856079,
      "num_input_tokens_seen": 403190560,
      "step": 24620,
      "train_runtime": 47421.5563,
      "train_tokens_per_second": 8502.263
    },
    {
      "epoch": 1.6580258551036897,
      "grad_norm": 1.2392981086312407,
      "learning_rate": 7.044738034990273e-07,
      "loss": 0.2987481355667114,
      "num_input_tokens_seen": 403272480,
      "step": 24625,
      "train_runtime": 47431.405,
      "train_tokens_per_second": 8502.225
    },
    {
      "epoch": 1.65836251009965,
      "grad_norm": 1.2705088518732797,
      "learning_rate": 7.031211656480841e-07,
      "loss": 0.28595986366271975,
      "num_input_tokens_seen": 403354400,
      "step": 24630,
      "train_runtime": 47441.341,
      "train_tokens_per_second": 8502.171
    },
    {
      "epoch": 1.65869916509561,
      "grad_norm": 1.222617693833914,
      "learning_rate": 7.017697294056058e-07,
      "loss": 0.27545087337493895,
      "num_input_tokens_seen": 403436320,
      "step": 24635,
      "train_runtime": 47451.4135,
      "train_tokens_per_second": 8502.093
    },
    {
      "epoch": 1.65903582009157,
      "grad_norm": 1.260895371165626,
      "learning_rate": 7.004194951495163e-07,
      "loss": 0.28240270614624025,
      "num_input_tokens_seen": 403518240,
      "step": 24640,
      "train_runtime": 47461.1976,
      "train_tokens_per_second": 8502.066
    },
    {
      "epoch": 1.6593724750875303,
      "grad_norm": 1.1532338922643088,
      "learning_rate": 6.990704632574046e-07,
      "loss": 0.2820900440216064,
      "num_input_tokens_seen": 403600160,
      "step": 24645,
      "train_runtime": 47470.9523,
      "train_tokens_per_second": 8502.045
    },
    {
      "epoch": 1.6597091300834905,
      "grad_norm": 1.2432930710381302,
      "learning_rate": 6.977226341065219e-07,
      "loss": 0.28559260368347167,
      "num_input_tokens_seen": 403682080,
      "step": 24650,
      "train_runtime": 47480.8916,
      "train_tokens_per_second": 8501.99
    },
    {
      "epoch": 1.6600457850794506,
      "grad_norm": 0.9948496558174752,
      "learning_rate": 6.96376008073788e-07,
      "loss": 0.2679098129272461,
      "num_input_tokens_seen": 403764000,
      "step": 24655,
      "train_runtime": 47490.6087,
      "train_tokens_per_second": 8501.976
    },
    {
      "epoch": 1.6603824400754106,
      "grad_norm": 1.2625496759382853,
      "learning_rate": 6.950305855357781e-07,
      "loss": 0.27260556221008303,
      "num_input_tokens_seen": 403845920,
      "step": 24660,
      "train_runtime": 47500.3828,
      "train_tokens_per_second": 8501.951
    },
    {
      "epoch": 1.6607190950713708,
      "grad_norm": 1.196716250636794,
      "learning_rate": 6.936863668687383e-07,
      "loss": 0.2981389284133911,
      "num_input_tokens_seen": 403927840,
      "step": 24665,
      "train_runtime": 47510.0781,
      "train_tokens_per_second": 8501.94
    },
    {
      "epoch": 1.661055750067331,
      "grad_norm": 1.3263410387813586,
      "learning_rate": 6.923433524485745e-07,
      "loss": 0.3217895030975342,
      "num_input_tokens_seen": 404009760,
      "step": 24670,
      "train_runtime": 47519.9026,
      "train_tokens_per_second": 8501.906
    },
    {
      "epoch": 1.6613924050632911,
      "grad_norm": 1.136989905345013,
      "learning_rate": 6.910015426508554e-07,
      "loss": 0.2792982578277588,
      "num_input_tokens_seen": 404091680,
      "step": 24675,
      "train_runtime": 47529.665,
      "train_tokens_per_second": 8501.884
    },
    {
      "epoch": 1.6617290600592511,
      "grad_norm": 1.2649384779808879,
      "learning_rate": 6.896609378508152e-07,
      "loss": 0.27901790142059324,
      "num_input_tokens_seen": 404173600,
      "step": 24680,
      "train_runtime": 47539.514,
      "train_tokens_per_second": 8501.845
    },
    {
      "epoch": 1.6620657150552114,
      "grad_norm": 1.0046343558389743,
      "learning_rate": 6.883215384233488e-07,
      "loss": 0.298006534576416,
      "num_input_tokens_seen": 404255520,
      "step": 24685,
      "train_runtime": 47549.3643,
      "train_tokens_per_second": 8501.807
    },
    {
      "epoch": 1.6624023700511716,
      "grad_norm": 1.0970508836431059,
      "learning_rate": 6.869833447430152e-07,
      "loss": 0.27531554698944094,
      "num_input_tokens_seen": 404337440,
      "step": 24690,
      "train_runtime": 47559.1463,
      "train_tokens_per_second": 8501.781
    },
    {
      "epoch": 1.6627390250471317,
      "grad_norm": 1.1142308133531351,
      "learning_rate": 6.856463571840361e-07,
      "loss": 0.25077157020568847,
      "num_input_tokens_seen": 404419360,
      "step": 24695,
      "train_runtime": 47568.8079,
      "train_tokens_per_second": 8501.776
    },
    {
      "epoch": 1.6630756800430917,
      "grad_norm": 1.1649681634411728,
      "learning_rate": 6.843105761202945e-07,
      "loss": 0.31010823249816893,
      "num_input_tokens_seen": 404501280,
      "step": 24700,
      "train_runtime": 47578.4455,
      "train_tokens_per_second": 8501.776
    },
    {
      "epoch": 1.663412335039052,
      "grad_norm": 1.1845258159259593,
      "learning_rate": 6.829760019253412e-07,
      "loss": 0.3039370536804199,
      "num_input_tokens_seen": 404583200,
      "step": 24705,
      "train_runtime": 47588.1577,
      "train_tokens_per_second": 8501.762
    },
    {
      "epoch": 1.6637489900350122,
      "grad_norm": 1.0489009862958931,
      "learning_rate": 6.816426349723804e-07,
      "loss": 0.288358736038208,
      "num_input_tokens_seen": 404665120,
      "step": 24710,
      "train_runtime": 47597.7251,
      "train_tokens_per_second": 8501.774
    },
    {
      "epoch": 1.6640856450309722,
      "grad_norm": 1.1367308378032814,
      "learning_rate": 6.803104756342877e-07,
      "loss": 0.2841032028198242,
      "num_input_tokens_seen": 404747040,
      "step": 24715,
      "train_runtime": 47607.6098,
      "train_tokens_per_second": 8501.73
    },
    {
      "epoch": 1.6644223000269323,
      "grad_norm": 1.0969786754020758,
      "learning_rate": 6.789795242835962e-07,
      "loss": 0.26358304023742674,
      "num_input_tokens_seen": 404828960,
      "step": 24720,
      "train_runtime": 47617.1249,
      "train_tokens_per_second": 8501.751
    },
    {
      "epoch": 1.6647589550228925,
      "grad_norm": 1.0151325208426847,
      "learning_rate": 6.776497812925015e-07,
      "loss": 0.25214352607727053,
      "num_input_tokens_seen": 404910880,
      "step": 24725,
      "train_runtime": 47626.835,
      "train_tokens_per_second": 8501.738
    },
    {
      "epoch": 1.6650956100188528,
      "grad_norm": 1.0070227059569317,
      "learning_rate": 6.763212470328628e-07,
      "loss": 0.29368374347686765,
      "num_input_tokens_seen": 404992800,
      "step": 24730,
      "train_runtime": 47636.4024,
      "train_tokens_per_second": 8501.75
    },
    {
      "epoch": 1.6654322650148128,
      "grad_norm": 1.0962813193345962,
      "learning_rate": 6.749939218762003e-07,
      "loss": 0.27081894874572754,
      "num_input_tokens_seen": 405074720,
      "step": 24735,
      "train_runtime": 47646.1182,
      "train_tokens_per_second": 8501.736
    },
    {
      "epoch": 1.6657689200107728,
      "grad_norm": 1.1731615504520345,
      "learning_rate": 6.736678061936958e-07,
      "loss": 0.29559624195098877,
      "num_input_tokens_seen": 405156640,
      "step": 24740,
      "train_runtime": 47655.7423,
      "train_tokens_per_second": 8501.738
    },
    {
      "epoch": 1.666105575006733,
      "grad_norm": 1.477310333781094,
      "learning_rate": 6.723429003561937e-07,
      "loss": 0.27633512020111084,
      "num_input_tokens_seen": 405238560,
      "step": 24745,
      "train_runtime": 47665.3699,
      "train_tokens_per_second": 8501.74
    },
    {
      "epoch": 1.6664422300026933,
      "grad_norm": 1.179156382320855,
      "learning_rate": 6.710192047341984e-07,
      "loss": 0.28730823993682864,
      "num_input_tokens_seen": 405320480,
      "step": 24750,
      "train_runtime": 47675.1388,
      "train_tokens_per_second": 8501.716
    },
    {
      "epoch": 1.6667788849986533,
      "grad_norm": 1.1417426803609307,
      "learning_rate": 6.69696719697881e-07,
      "loss": 0.26996359825134275,
      "num_input_tokens_seen": 405402400,
      "step": 24755,
      "train_runtime": 47684.766,
      "train_tokens_per_second": 8501.717
    },
    {
      "epoch": 1.6671155399946134,
      "grad_norm": 1.1652718163021434,
      "learning_rate": 6.683754456170655e-07,
      "loss": 0.29858424663543703,
      "num_input_tokens_seen": 405484320,
      "step": 24760,
      "train_runtime": 47694.6335,
      "train_tokens_per_second": 8501.676
    },
    {
      "epoch": 1.6674521949905736,
      "grad_norm": 1.0791888081255885,
      "learning_rate": 6.670553828612448e-07,
      "loss": 0.25715298652648927,
      "num_input_tokens_seen": 405566000,
      "step": 24765,
      "train_runtime": 47704.636,
      "train_tokens_per_second": 8501.606
    },
    {
      "epoch": 1.6677888499865339,
      "grad_norm": 1.0097146471489553,
      "learning_rate": 6.657365317995696e-07,
      "loss": 0.25123958587646483,
      "num_input_tokens_seen": 405647920,
      "step": 24770,
      "train_runtime": 47714.2546,
      "train_tokens_per_second": 8501.609
    },
    {
      "epoch": 1.668125504982494,
      "grad_norm": 1.2327357319652184,
      "learning_rate": 6.644188928008521e-07,
      "loss": 0.31403188705444335,
      "num_input_tokens_seen": 405729840,
      "step": 24775,
      "train_runtime": 47723.9299,
      "train_tokens_per_second": 8501.602
    },
    {
      "epoch": 1.668462159978454,
      "grad_norm": 1.0908529594542204,
      "learning_rate": 6.631024662335667e-07,
      "loss": 0.2792530536651611,
      "num_input_tokens_seen": 405811760,
      "step": 24780,
      "train_runtime": 47733.5142,
      "train_tokens_per_second": 8501.611
    },
    {
      "epoch": 1.6687988149744142,
      "grad_norm": 1.132316502928142,
      "learning_rate": 6.617872524658469e-07,
      "loss": 0.2599059581756592,
      "num_input_tokens_seen": 405893680,
      "step": 24785,
      "train_runtime": 47743.2281,
      "train_tokens_per_second": 8501.597
    },
    {
      "epoch": 1.6691354699703744,
      "grad_norm": 1.3253300673908184,
      "learning_rate": 6.604732518654882e-07,
      "loss": 0.29565930366516113,
      "num_input_tokens_seen": 405975600,
      "step": 24790,
      "train_runtime": 47752.9999,
      "train_tokens_per_second": 8501.573
    },
    {
      "epoch": 1.6694721249663345,
      "grad_norm": 1.3557179658908498,
      "learning_rate": 6.591604647999473e-07,
      "loss": 0.31192169189453123,
      "num_input_tokens_seen": 406057520,
      "step": 24795,
      "train_runtime": 47762.7738,
      "train_tokens_per_second": 8501.548
    },
    {
      "epoch": 1.6698087799622945,
      "grad_norm": 1.2184593943833735,
      "learning_rate": 6.57848891636339e-07,
      "loss": 0.29067816734313967,
      "num_input_tokens_seen": 406139440,
      "step": 24800,
      "train_runtime": 47772.6017,
      "train_tokens_per_second": 8501.514
    },
    {
      "epoch": 1.6701454349582547,
      "grad_norm": 1.377613107826096,
      "learning_rate": 6.565385327414441e-07,
      "loss": 0.3044891834259033,
      "num_input_tokens_seen": 406221360,
      "step": 24805,
      "train_runtime": 47782.2217,
      "train_tokens_per_second": 8501.517
    },
    {
      "epoch": 1.670482089954215,
      "grad_norm": 1.2090034964985112,
      "learning_rate": 6.552293884816968e-07,
      "loss": 0.2571259021759033,
      "num_input_tokens_seen": 406303280,
      "step": 24810,
      "train_runtime": 47791.8043,
      "train_tokens_per_second": 8501.526
    },
    {
      "epoch": 1.670818744950175,
      "grad_norm": 1.3151318212778396,
      "learning_rate": 6.539214592231968e-07,
      "loss": 0.29944624900817873,
      "num_input_tokens_seen": 406385200,
      "step": 24815,
      "train_runtime": 47801.4233,
      "train_tokens_per_second": 8501.529
    },
    {
      "epoch": 1.671155399946135,
      "grad_norm": 1.0091171938678642,
      "learning_rate": 6.526147453317028e-07,
      "loss": 0.26236777305603026,
      "num_input_tokens_seen": 406467120,
      "step": 24820,
      "train_runtime": 47811.2537,
      "train_tokens_per_second": 8501.495
    },
    {
      "epoch": 1.6714920549420953,
      "grad_norm": 1.1992398239601143,
      "learning_rate": 6.513092471726307e-07,
      "loss": 0.27171261310577394,
      "num_input_tokens_seen": 406549040,
      "step": 24825,
      "train_runtime": 47820.9861,
      "train_tokens_per_second": 8501.478
    },
    {
      "epoch": 1.6718287099380555,
      "grad_norm": 1.0563973749166327,
      "learning_rate": 6.500049651110635e-07,
      "loss": 0.25964038372039794,
      "num_input_tokens_seen": 406630960,
      "step": 24830,
      "train_runtime": 47830.6443,
      "train_tokens_per_second": 8501.474
    },
    {
      "epoch": 1.6721653649340156,
      "grad_norm": 1.138264582438342,
      "learning_rate": 6.487018995117339e-07,
      "loss": 0.2637629508972168,
      "num_input_tokens_seen": 406712880,
      "step": 24835,
      "train_runtime": 47840.2574,
      "train_tokens_per_second": 8501.478
    },
    {
      "epoch": 1.6725020199299756,
      "grad_norm": 1.0027053444330052,
      "learning_rate": 6.474000507390455e-07,
      "loss": 0.28280115127563477,
      "num_input_tokens_seen": 406794800,
      "step": 24840,
      "train_runtime": 47850.0645,
      "train_tokens_per_second": 8501.447
    },
    {
      "epoch": 1.6728386749259359,
      "grad_norm": 1.2510491428164192,
      "learning_rate": 6.460994191570513e-07,
      "loss": 0.29479689598083497,
      "num_input_tokens_seen": 406876720,
      "step": 24845,
      "train_runtime": 47859.7912,
      "train_tokens_per_second": 8501.431
    },
    {
      "epoch": 1.673175329921896,
      "grad_norm": 1.104624415474013,
      "learning_rate": 6.448000051294717e-07,
      "loss": 0.2599815607070923,
      "num_input_tokens_seen": 406958640,
      "step": 24850,
      "train_runtime": 47869.6701,
      "train_tokens_per_second": 8501.388
    },
    {
      "epoch": 1.6735119849178561,
      "grad_norm": 1.1634192280391278,
      "learning_rate": 6.435018090196832e-07,
      "loss": 0.2720421314239502,
      "num_input_tokens_seen": 407040560,
      "step": 24855,
      "train_runtime": 47879.5452,
      "train_tokens_per_second": 8501.346
    },
    {
      "epoch": 1.6738486399138162,
      "grad_norm": 1.1533087482805797,
      "learning_rate": 6.42204831190722e-07,
      "loss": 0.27932515144348147,
      "num_input_tokens_seen": 407122480,
      "step": 24860,
      "train_runtime": 47889.2649,
      "train_tokens_per_second": 8501.331
    },
    {
      "epoch": 1.6741852949097764,
      "grad_norm": 1.1616784058617122,
      "learning_rate": 6.409090720052835e-07,
      "loss": 0.28957037925720214,
      "num_input_tokens_seen": 407204400,
      "step": 24865,
      "train_runtime": 47898.9445,
      "train_tokens_per_second": 8501.323
    },
    {
      "epoch": 1.6745219499057367,
      "grad_norm": 1.0498710947644505,
      "learning_rate": 6.396145318257229e-07,
      "loss": 0.24988236427307128,
      "num_input_tokens_seen": 407286320,
      "step": 24870,
      "train_runtime": 47908.7021,
      "train_tokens_per_second": 8501.301
    },
    {
      "epoch": 1.6748586049016967,
      "grad_norm": 1.148118569590767,
      "learning_rate": 6.383212110140535e-07,
      "loss": 0.3053235292434692,
      "num_input_tokens_seen": 407368240,
      "step": 24875,
      "train_runtime": 47918.4336,
      "train_tokens_per_second": 8501.285
    },
    {
      "epoch": 1.6751952598976567,
      "grad_norm": 1.243116960930438,
      "learning_rate": 6.370291099319509e-07,
      "loss": 0.3063422679901123,
      "num_input_tokens_seen": 407450160,
      "step": 24880,
      "train_runtime": 47928.1718,
      "train_tokens_per_second": 8501.266
    },
    {
      "epoch": 1.675531914893617,
      "grad_norm": 1.2196417916179645,
      "learning_rate": 6.357382289407433e-07,
      "loss": 0.28368048667907714,
      "num_input_tokens_seen": 407532080,
      "step": 24885,
      "train_runtime": 47938.005,
      "train_tokens_per_second": 8501.232
    },
    {
      "epoch": 1.6758685698895772,
      "grad_norm": 1.028632874482127,
      "learning_rate": 6.344485684014251e-07,
      "loss": 0.2825812816619873,
      "num_input_tokens_seen": 407614000,
      "step": 24890,
      "train_runtime": 47947.757,
      "train_tokens_per_second": 8501.211
    },
    {
      "epoch": 1.6762052248855372,
      "grad_norm": 1.2563218674044683,
      "learning_rate": 6.33160128674642e-07,
      "loss": 0.3047327041625977,
      "num_input_tokens_seen": 407695088,
      "step": 24895,
      "train_runtime": 47957.7547,
      "train_tokens_per_second": 8501.13
    },
    {
      "epoch": 1.6765418798814973,
      "grad_norm": 1.1454967224486758,
      "learning_rate": 6.318729101207055e-07,
      "loss": 0.26055312156677246,
      "num_input_tokens_seen": 407777008,
      "step": 24900,
      "train_runtime": 47967.3726,
      "train_tokens_per_second": 8501.133
    },
    {
      "epoch": 1.6768785348774575,
      "grad_norm": 1.0309431596162082,
      "learning_rate": 6.305869130995807e-07,
      "loss": 0.26894540786743165,
      "num_input_tokens_seen": 407858928,
      "step": 24905,
      "train_runtime": 47977.1518,
      "train_tokens_per_second": 8501.108
    },
    {
      "epoch": 1.6772151898734178,
      "grad_norm": 1.0234889499288067,
      "learning_rate": 6.293021379708924e-07,
      "loss": 0.2789770126342773,
      "num_input_tokens_seen": 407940848,
      "step": 24910,
      "train_runtime": 47986.8609,
      "train_tokens_per_second": 8501.095
    },
    {
      "epoch": 1.6775518448693778,
      "grad_norm": 1.2587700633346566,
      "learning_rate": 6.280185850939241e-07,
      "loss": 0.3077085494995117,
      "num_input_tokens_seen": 408022768,
      "step": 24915,
      "train_runtime": 47996.5638,
      "train_tokens_per_second": 8501.083
    },
    {
      "epoch": 1.677888499865338,
      "grad_norm": 1.2167907636640043,
      "learning_rate": 6.267362548276173e-07,
      "loss": 0.28091721534729003,
      "num_input_tokens_seen": 408104688,
      "step": 24920,
      "train_runtime": 48006.3518,
      "train_tokens_per_second": 8501.056
    },
    {
      "epoch": 1.6782251548612983,
      "grad_norm": 1.2244081984277497,
      "learning_rate": 6.254551475305703e-07,
      "loss": 0.2809918880462646,
      "num_input_tokens_seen": 408186608,
      "step": 24925,
      "train_runtime": 48015.9269,
      "train_tokens_per_second": 8501.067
    },
    {
      "epoch": 1.6785618098572583,
      "grad_norm": 1.0829687647531803,
      "learning_rate": 6.241752635610443e-07,
      "loss": 0.30669422149658204,
      "num_input_tokens_seen": 408268528,
      "step": 24930,
      "train_runtime": 48025.5352,
      "train_tokens_per_second": 8501.072
    },
    {
      "epoch": 1.6788984648532184,
      "grad_norm": 1.2843138759068096,
      "learning_rate": 6.228966032769496e-07,
      "loss": 0.30317833423614504,
      "num_input_tokens_seen": 408350448,
      "step": 24935,
      "train_runtime": 48035.3145,
      "train_tokens_per_second": 8501.047
    },
    {
      "epoch": 1.6792351198491786,
      "grad_norm": 0.946969038143989,
      "learning_rate": 6.216191670358651e-07,
      "loss": 0.25751047134399413,
      "num_input_tokens_seen": 408432368,
      "step": 24940,
      "train_runtime": 48044.8562,
      "train_tokens_per_second": 8501.063
    },
    {
      "epoch": 1.6795717748451389,
      "grad_norm": 1.2457552386016968,
      "learning_rate": 6.203429551950163e-07,
      "loss": 0.31105265617370603,
      "num_input_tokens_seen": 408514288,
      "step": 24945,
      "train_runtime": 48054.5227,
      "train_tokens_per_second": 8501.058
    },
    {
      "epoch": 1.679908429841099,
      "grad_norm": 1.0611648499444877,
      "learning_rate": 6.19067968111296e-07,
      "loss": 0.2825040817260742,
      "num_input_tokens_seen": 408596208,
      "step": 24950,
      "train_runtime": 48064.2617,
      "train_tokens_per_second": 8501.04
    },
    {
      "epoch": 1.680245084837059,
      "grad_norm": 0.9744952796190492,
      "learning_rate": 6.177942061412478e-07,
      "loss": 0.26905758380889894,
      "num_input_tokens_seen": 408678128,
      "step": 24955,
      "train_runtime": 48073.8923,
      "train_tokens_per_second": 8501.041
    },
    {
      "epoch": 1.6805817398330192,
      "grad_norm": 1.2241356414787272,
      "learning_rate": 6.16521669641077e-07,
      "loss": 0.28973124027252195,
      "num_input_tokens_seen": 408760048,
      "step": 24960,
      "train_runtime": 48083.5348,
      "train_tokens_per_second": 8501.04
    },
    {
      "epoch": 1.6809183948289794,
      "grad_norm": 0.9366653337888823,
      "learning_rate": 6.152503589666426e-07,
      "loss": 0.2843919277191162,
      "num_input_tokens_seen": 408841968,
      "step": 24965,
      "train_runtime": 48093.4385,
      "train_tokens_per_second": 8500.993
    },
    {
      "epoch": 1.6812550498249395,
      "grad_norm": 1.0461900686218881,
      "learning_rate": 6.139802744734636e-07,
      "loss": 0.2599933624267578,
      "num_input_tokens_seen": 408923888,
      "step": 24970,
      "train_runtime": 48103.2607,
      "train_tokens_per_second": 8500.96
    },
    {
      "epoch": 1.6815917048208995,
      "grad_norm": 1.0034984213797373,
      "learning_rate": 6.127114165167142e-07,
      "loss": 0.24028096199035645,
      "num_input_tokens_seen": 409005808,
      "step": 24975,
      "train_runtime": 48113.1247,
      "train_tokens_per_second": 8500.92
    },
    {
      "epoch": 1.6819283598168597,
      "grad_norm": 1.23619110593004,
      "learning_rate": 6.11443785451229e-07,
      "loss": 0.2692131996154785,
      "num_input_tokens_seen": 409087728,
      "step": 24980,
      "train_runtime": 48122.9927,
      "train_tokens_per_second": 8500.879
    },
    {
      "epoch": 1.68226501481282,
      "grad_norm": 1.0890393711567834,
      "learning_rate": 6.101773816314926e-07,
      "loss": 0.2728546142578125,
      "num_input_tokens_seen": 409169648,
      "step": 24985,
      "train_runtime": 48132.8014,
      "train_tokens_per_second": 8500.848
    },
    {
      "epoch": 1.68260166980878,
      "grad_norm": 1.159103566960516,
      "learning_rate": 6.089122054116558e-07,
      "loss": 0.3043032646179199,
      "num_input_tokens_seen": 409251568,
      "step": 24990,
      "train_runtime": 48142.5282,
      "train_tokens_per_second": 8500.832
    },
    {
      "epoch": 1.68293832480474,
      "grad_norm": 1.2620965221354437,
      "learning_rate": 6.076482571455161e-07,
      "loss": 0.2986126899719238,
      "num_input_tokens_seen": 409333488,
      "step": 24995,
      "train_runtime": 48152.0617,
      "train_tokens_per_second": 8500.851
    },
    {
      "epoch": 1.6832749798007003,
      "grad_norm": 0.9964405162342121,
      "learning_rate": 6.063855371865369e-07,
      "loss": 0.273673415184021,
      "num_input_tokens_seen": 409414648,
      "step": 25000,
      "train_runtime": 48161.8228,
      "train_tokens_per_second": 8500.813
    },
    {
      "epoch": 1.6836116347966605,
      "grad_norm": 1.3103681822425535,
      "learning_rate": 6.051240458878316e-07,
      "loss": 0.32117066383361814,
      "num_input_tokens_seen": 409496568,
      "step": 25005,
      "train_runtime": 48171.3994,
      "train_tokens_per_second": 8500.824
    },
    {
      "epoch": 1.6839482897926206,
      "grad_norm": 1.1560927163861734,
      "learning_rate": 6.038637836021721e-07,
      "loss": 0.2693908452987671,
      "num_input_tokens_seen": 409578488,
      "step": 25010,
      "train_runtime": 48180.9956,
      "train_tokens_per_second": 8500.831
    },
    {
      "epoch": 1.6842849447885806,
      "grad_norm": 0.9918907695212308,
      "learning_rate": 6.026047506819882e-07,
      "loss": 0.27679221630096434,
      "num_input_tokens_seen": 409660408,
      "step": 25015,
      "train_runtime": 48190.5807,
      "train_tokens_per_second": 8500.84
    },
    {
      "epoch": 1.6846215997845408,
      "grad_norm": 1.257327352254784,
      "learning_rate": 6.013469474793632e-07,
      "loss": 0.2850696086883545,
      "num_input_tokens_seen": 409742328,
      "step": 25020,
      "train_runtime": 48200.1599,
      "train_tokens_per_second": 8500.85
    },
    {
      "epoch": 1.684958254780501,
      "grad_norm": 1.0521739416183906,
      "learning_rate": 6.000903743460368e-07,
      "loss": 0.2664146423339844,
      "num_input_tokens_seen": 409824248,
      "step": 25025,
      "train_runtime": 48209.6783,
      "train_tokens_per_second": 8500.871
    },
    {
      "epoch": 1.6852949097764611,
      "grad_norm": 1.054683942534278,
      "learning_rate": 5.98835031633409e-07,
      "loss": 0.2909480810165405,
      "num_input_tokens_seen": 409906168,
      "step": 25030,
      "train_runtime": 48219.2389,
      "train_tokens_per_second": 8500.884
    },
    {
      "epoch": 1.6856315647724212,
      "grad_norm": 1.4907418740285396,
      "learning_rate": 5.97580919692528e-07,
      "loss": 0.3016523838043213,
      "num_input_tokens_seen": 409988088,
      "step": 25035,
      "train_runtime": 48228.8719,
      "train_tokens_per_second": 8500.885
    },
    {
      "epoch": 1.6859682197683814,
      "grad_norm": 1.2162055286105131,
      "learning_rate": 5.963280388741072e-07,
      "loss": 0.2821801662445068,
      "num_input_tokens_seen": 410070008,
      "step": 25040,
      "train_runtime": 48238.5283,
      "train_tokens_per_second": 8500.881
    },
    {
      "epoch": 1.6863048747643417,
      "grad_norm": 1.1915474053627502,
      "learning_rate": 5.950763895285055e-07,
      "loss": 0.2702339172363281,
      "num_input_tokens_seen": 410151928,
      "step": 25045,
      "train_runtime": 48248.0723,
      "train_tokens_per_second": 8500.898
    },
    {
      "epoch": 1.6866415297603017,
      "grad_norm": 0.9983390842029751,
      "learning_rate": 5.938259720057465e-07,
      "loss": 0.2696484088897705,
      "num_input_tokens_seen": 410233848,
      "step": 25050,
      "train_runtime": 48257.6299,
      "train_tokens_per_second": 8500.912
    },
    {
      "epoch": 1.6869781847562617,
      "grad_norm": 0.9985214353645371,
      "learning_rate": 5.925767866555043e-07,
      "loss": 0.24139606952667236,
      "num_input_tokens_seen": 410315768,
      "step": 25055,
      "train_runtime": 48267.2188,
      "train_tokens_per_second": 8500.92
    },
    {
      "epoch": 1.687314839752222,
      "grad_norm": 1.4441847982767055,
      "learning_rate": 5.913288338271095e-07,
      "loss": 0.315148138999939,
      "num_input_tokens_seen": 410397688,
      "step": 25060,
      "train_runtime": 48276.9304,
      "train_tokens_per_second": 8500.907
    },
    {
      "epoch": 1.6876514947481822,
      "grad_norm": 1.3622247499033107,
      "learning_rate": 5.900821138695478e-07,
      "loss": 0.29043102264404297,
      "num_input_tokens_seen": 410479608,
      "step": 25065,
      "train_runtime": 48286.4771,
      "train_tokens_per_second": 8500.923
    },
    {
      "epoch": 1.6879881497441422,
      "grad_norm": 1.2590224438190958,
      "learning_rate": 5.888366271314605e-07,
      "loss": 0.27917678356170655,
      "num_input_tokens_seen": 410561528,
      "step": 25070,
      "train_runtime": 48296.2022,
      "train_tokens_per_second": 8500.907
    },
    {
      "epoch": 1.6883248047401023,
      "grad_norm": 1.1296527037753228,
      "learning_rate": 5.875923739611439e-07,
      "loss": 0.3160854339599609,
      "num_input_tokens_seen": 410643448,
      "step": 25075,
      "train_runtime": 48305.9206,
      "train_tokens_per_second": 8500.893
    },
    {
      "epoch": 1.6886614597360625,
      "grad_norm": 0.9156674959952994,
      "learning_rate": 5.863493547065513e-07,
      "loss": 0.2711193561553955,
      "num_input_tokens_seen": 410725368,
      "step": 25080,
      "train_runtime": 48315.6868,
      "train_tokens_per_second": 8500.87
    },
    {
      "epoch": 1.6889981147320228,
      "grad_norm": 1.0347286535418907,
      "learning_rate": 5.851075697152858e-07,
      "loss": 0.27056231498718264,
      "num_input_tokens_seen": 410807288,
      "step": 25085,
      "train_runtime": 48325.4664,
      "train_tokens_per_second": 8500.845
    },
    {
      "epoch": 1.6893347697279828,
      "grad_norm": 1.181004370284583,
      "learning_rate": 5.838670193346118e-07,
      "loss": 0.2994255304336548,
      "num_input_tokens_seen": 410889208,
      "step": 25090,
      "train_runtime": 48335.1071,
      "train_tokens_per_second": 8500.844
    },
    {
      "epoch": 1.6896714247239428,
      "grad_norm": 1.4755867106782332,
      "learning_rate": 5.826277039114426e-07,
      "loss": 0.28229260444641113,
      "num_input_tokens_seen": 410971128,
      "step": 25095,
      "train_runtime": 48344.7219,
      "train_tokens_per_second": 8500.848
    },
    {
      "epoch": 1.690008079719903,
      "grad_norm": 1.2801600798972295,
      "learning_rate": 5.813896237923505e-07,
      "loss": 0.31733059883117676,
      "num_input_tokens_seen": 411053048,
      "step": 25100,
      "train_runtime": 48354.2309,
      "train_tokens_per_second": 8500.87
    },
    {
      "epoch": 1.6903447347158633,
      "grad_norm": 1.1704870911033332,
      "learning_rate": 5.801527793235606e-07,
      "loss": 0.29116058349609375,
      "num_input_tokens_seen": 411134968,
      "step": 25105,
      "train_runtime": 48363.9893,
      "train_tokens_per_second": 8500.849
    },
    {
      "epoch": 1.6906813897118234,
      "grad_norm": 0.922861402460667,
      "learning_rate": 5.789171708509516e-07,
      "loss": 0.24949290752410888,
      "num_input_tokens_seen": 411216888,
      "step": 25110,
      "train_runtime": 48373.5709,
      "train_tokens_per_second": 8500.859
    },
    {
      "epoch": 1.6910180447077834,
      "grad_norm": 1.0079342756828846,
      "learning_rate": 5.776827987200584e-07,
      "loss": 0.2932230710983276,
      "num_input_tokens_seen": 411298808,
      "step": 25115,
      "train_runtime": 48383.1361,
      "train_tokens_per_second": 8500.871
    },
    {
      "epoch": 1.6913546997037436,
      "grad_norm": 1.165143633985543,
      "learning_rate": 5.764496632760691e-07,
      "loss": 0.28583967685699463,
      "num_input_tokens_seen": 411380728,
      "step": 25120,
      "train_runtime": 48392.7562,
      "train_tokens_per_second": 8500.874
    },
    {
      "epoch": 1.6916913546997039,
      "grad_norm": 1.2500927708701017,
      "learning_rate": 5.752177648638241e-07,
      "loss": 0.2712721586227417,
      "num_input_tokens_seen": 411462648,
      "step": 25125,
      "train_runtime": 48402.2728,
      "train_tokens_per_second": 8500.895
    },
    {
      "epoch": 1.692028009695664,
      "grad_norm": 1.2985137547681735,
      "learning_rate": 5.739871038278239e-07,
      "loss": 0.2946120262145996,
      "num_input_tokens_seen": 411544160,
      "step": 25130,
      "train_runtime": 48412.3592,
      "train_tokens_per_second": 8500.808
    },
    {
      "epoch": 1.692364664691624,
      "grad_norm": 1.1847293044682736,
      "learning_rate": 5.727576805122142e-07,
      "loss": 0.27293665409088136,
      "num_input_tokens_seen": 411626080,
      "step": 25135,
      "train_runtime": 48421.9955,
      "train_tokens_per_second": 8500.808
    },
    {
      "epoch": 1.6927013196875842,
      "grad_norm": 1.082563716176351,
      "learning_rate": 5.715294952608041e-07,
      "loss": 0.2960194110870361,
      "num_input_tokens_seen": 411708000,
      "step": 25140,
      "train_runtime": 48431.5752,
      "train_tokens_per_second": 8500.818
    },
    {
      "epoch": 1.6930379746835444,
      "grad_norm": 1.2510730312188938,
      "learning_rate": 5.703025484170471e-07,
      "loss": 0.29387893676757815,
      "num_input_tokens_seen": 411789920,
      "step": 25145,
      "train_runtime": 48441.1986,
      "train_tokens_per_second": 8500.82
    },
    {
      "epoch": 1.6933746296795045,
      "grad_norm": 1.2103201561580141,
      "learning_rate": 5.690768403240587e-07,
      "loss": 0.29504263401031494,
      "num_input_tokens_seen": 411871840,
      "step": 25150,
      "train_runtime": 48450.7998,
      "train_tokens_per_second": 8500.826
    },
    {
      "epoch": 1.6937112846754645,
      "grad_norm": 1.2652309648546782,
      "learning_rate": 5.67852371324602e-07,
      "loss": 0.3176659107208252,
      "num_input_tokens_seen": 411953760,
      "step": 25155,
      "train_runtime": 48460.5042,
      "train_tokens_per_second": 8500.815
    },
    {
      "epoch": 1.6940479396714248,
      "grad_norm": 1.0728515702143875,
      "learning_rate": 5.66629141761097e-07,
      "loss": 0.2623980283737183,
      "num_input_tokens_seen": 412035680,
      "step": 25160,
      "train_runtime": 48470.0621,
      "train_tokens_per_second": 8500.828
    },
    {
      "epoch": 1.694384594667385,
      "grad_norm": 1.0513266699553547,
      "learning_rate": 5.654071519756154e-07,
      "loss": 0.30491807460784914,
      "num_input_tokens_seen": 412117600,
      "step": 25165,
      "train_runtime": 48479.7082,
      "train_tokens_per_second": 8500.827
    },
    {
      "epoch": 1.694721249663345,
      "grad_norm": 1.0963077860233235,
      "learning_rate": 5.641864023098831e-07,
      "loss": 0.3098304271697998,
      "num_input_tokens_seen": 412199520,
      "step": 25170,
      "train_runtime": 48489.2895,
      "train_tokens_per_second": 8500.836
    },
    {
      "epoch": 1.695057904659305,
      "grad_norm": 1.0372275663974408,
      "learning_rate": 5.629668931052773e-07,
      "loss": 0.27609102725982665,
      "num_input_tokens_seen": 412281440,
      "step": 25175,
      "train_runtime": 48498.947,
      "train_tokens_per_second": 8500.833
    },
    {
      "epoch": 1.6953945596552653,
      "grad_norm": 1.2056095143135315,
      "learning_rate": 5.61748624702833e-07,
      "loss": 0.3181142807006836,
      "num_input_tokens_seen": 412363360,
      "step": 25180,
      "train_runtime": 48508.5471,
      "train_tokens_per_second": 8500.839
    },
    {
      "epoch": 1.6957312146512256,
      "grad_norm": 1.1778415261870134,
      "learning_rate": 5.605315974432313e-07,
      "loss": 0.28058459758758547,
      "num_input_tokens_seen": 412445280,
      "step": 25185,
      "train_runtime": 48518.2336,
      "train_tokens_per_second": 8500.83
    },
    {
      "epoch": 1.6960678696471856,
      "grad_norm": 1.1068916033060086,
      "learning_rate": 5.593158116668146e-07,
      "loss": 0.2748985767364502,
      "num_input_tokens_seen": 412527200,
      "step": 25190,
      "train_runtime": 48527.7808,
      "train_tokens_per_second": 8500.846
    },
    {
      "epoch": 1.6964045246431456,
      "grad_norm": 1.0318889826525561,
      "learning_rate": 5.581012677135683e-07,
      "loss": 0.2703504800796509,
      "num_input_tokens_seen": 412609120,
      "step": 25195,
      "train_runtime": 48537.4558,
      "train_tokens_per_second": 8500.839
    },
    {
      "epoch": 1.6967411796391059,
      "grad_norm": 1.114444304051273,
      "learning_rate": 5.56887965923139e-07,
      "loss": 0.29882102012634276,
      "num_input_tokens_seen": 412691040,
      "step": 25200,
      "train_runtime": 48547.2023,
      "train_tokens_per_second": 8500.82
    },
    {
      "epoch": 1.6970778346350661,
      "grad_norm": 1.201633326397433,
      "learning_rate": 5.556759066348227e-07,
      "loss": 0.2848924160003662,
      "num_input_tokens_seen": 412772960,
      "step": 25205,
      "train_runtime": 48556.9533,
      "train_tokens_per_second": 8500.8
    },
    {
      "epoch": 1.6974144896310261,
      "grad_norm": 1.1329922166343387,
      "learning_rate": 5.54465090187567e-07,
      "loss": 0.3183952569961548,
      "num_input_tokens_seen": 412854880,
      "step": 25210,
      "train_runtime": 48566.796,
      "train_tokens_per_second": 8500.764
    },
    {
      "epoch": 1.6977511446269862,
      "grad_norm": 1.1282246778575975,
      "learning_rate": 5.532555169199727e-07,
      "loss": 0.25163631439208983,
      "num_input_tokens_seen": 412936512,
      "step": 25215,
      "train_runtime": 48576.835,
      "train_tokens_per_second": 8500.688
    },
    {
      "epoch": 1.6980877996229464,
      "grad_norm": 1.1798542736756257,
      "learning_rate": 5.520471871702937e-07,
      "loss": 0.270023775100708,
      "num_input_tokens_seen": 413018432,
      "step": 25220,
      "train_runtime": 48586.4851,
      "train_tokens_per_second": 8500.686
    },
    {
      "epoch": 1.6984244546189067,
      "grad_norm": 1.056698908750375,
      "learning_rate": 5.508401012764341e-07,
      "loss": 0.2860596179962158,
      "num_input_tokens_seen": 413100352,
      "step": 25225,
      "train_runtime": 48596.1591,
      "train_tokens_per_second": 8500.679
    },
    {
      "epoch": 1.6987611096148667,
      "grad_norm": 1.095436670892407,
      "learning_rate": 5.496342595759552e-07,
      "loss": 0.26360230445861815,
      "num_input_tokens_seen": 413182272,
      "step": 25230,
      "train_runtime": 48605.6699,
      "train_tokens_per_second": 8500.701
    },
    {
      "epoch": 1.6990977646108267,
      "grad_norm": 1.1422806694653989,
      "learning_rate": 5.484296624060626e-07,
      "loss": 0.2665090084075928,
      "num_input_tokens_seen": 413264192,
      "step": 25235,
      "train_runtime": 48615.3015,
      "train_tokens_per_second": 8500.702
    },
    {
      "epoch": 1.699434419606787,
      "grad_norm": 1.0613765551765593,
      "learning_rate": 5.472263101036212e-07,
      "loss": 0.28779354095458987,
      "num_input_tokens_seen": 413346112,
      "step": 25240,
      "train_runtime": 48624.8056,
      "train_tokens_per_second": 8500.725
    },
    {
      "epoch": 1.6997710746027472,
      "grad_norm": 1.3959180806464582,
      "learning_rate": 5.460242030051421e-07,
      "loss": 0.29095911979675293,
      "num_input_tokens_seen": 413428032,
      "step": 25245,
      "train_runtime": 48634.4299,
      "train_tokens_per_second": 8500.727
    },
    {
      "epoch": 1.7001077295987073,
      "grad_norm": 1.1062348453691442,
      "learning_rate": 5.448233414467924e-07,
      "loss": 0.310865592956543,
      "num_input_tokens_seen": 413509952,
      "step": 25250,
      "train_runtime": 48643.9334,
      "train_tokens_per_second": 8500.751
    },
    {
      "epoch": 1.7004443845946673,
      "grad_norm": 0.9517185527249682,
      "learning_rate": 5.43623725764389e-07,
      "loss": 0.28531584739685056,
      "num_input_tokens_seen": 413591872,
      "step": 25255,
      "train_runtime": 48653.5389,
      "train_tokens_per_second": 8500.756
    },
    {
      "epoch": 1.7007810395906275,
      "grad_norm": 1.1834592997243951,
      "learning_rate": 5.424253562934001e-07,
      "loss": 0.28069272041320803,
      "num_input_tokens_seen": 413673792,
      "step": 25260,
      "train_runtime": 48663.0975,
      "train_tokens_per_second": 8500.77
    },
    {
      "epoch": 1.7011176945865878,
      "grad_norm": 1.1396719107997837,
      "learning_rate": 5.412282333689461e-07,
      "loss": 0.3211845397949219,
      "num_input_tokens_seen": 413755712,
      "step": 25265,
      "train_runtime": 48672.6609,
      "train_tokens_per_second": 8500.783
    },
    {
      "epoch": 1.7014543495825478,
      "grad_norm": 1.1602423153841437,
      "learning_rate": 5.400323573257987e-07,
      "loss": 0.2746503591537476,
      "num_input_tokens_seen": 413837632,
      "step": 25270,
      "train_runtime": 48682.2404,
      "train_tokens_per_second": 8500.793
    },
    {
      "epoch": 1.7017910045785078,
      "grad_norm": 1.2054853221635602,
      "learning_rate": 5.388377284983803e-07,
      "loss": 0.2878596305847168,
      "num_input_tokens_seen": 413919552,
      "step": 25275,
      "train_runtime": 48691.8503,
      "train_tokens_per_second": 8500.797
    },
    {
      "epoch": 1.702127659574468,
      "grad_norm": 1.241902460060227,
      "learning_rate": 5.37644347220767e-07,
      "loss": 0.2757307767868042,
      "num_input_tokens_seen": 414001472,
      "step": 25280,
      "train_runtime": 48701.5137,
      "train_tokens_per_second": 8500.793
    },
    {
      "epoch": 1.7024643145704283,
      "grad_norm": 1.2239827892309778,
      "learning_rate": 5.364522138266809e-07,
      "loss": 0.28825929164886477,
      "num_input_tokens_seen": 414083392,
      "step": 25285,
      "train_runtime": 48711.1542,
      "train_tokens_per_second": 8500.792
    },
    {
      "epoch": 1.7028009695663884,
      "grad_norm": 1.1368073078788983,
      "learning_rate": 5.352613286495018e-07,
      "loss": 0.28681752681732176,
      "num_input_tokens_seen": 414165256,
      "step": 25290,
      "train_runtime": 48721.2004,
      "train_tokens_per_second": 8500.719
    },
    {
      "epoch": 1.7031376245623484,
      "grad_norm": 0.9979223962433489,
      "learning_rate": 5.340716920222527e-07,
      "loss": 0.2349106788635254,
      "num_input_tokens_seen": 414247176,
      "step": 25295,
      "train_runtime": 48730.8008,
      "train_tokens_per_second": 8500.726
    },
    {
      "epoch": 1.7034742795583087,
      "grad_norm": 1.2996181175328232,
      "learning_rate": 5.32883304277616e-07,
      "loss": 0.29776463508605955,
      "num_input_tokens_seen": 414329096,
      "step": 25300,
      "train_runtime": 48740.4756,
      "train_tokens_per_second": 8500.719
    },
    {
      "epoch": 1.703810934554269,
      "grad_norm": 1.2707249333392494,
      "learning_rate": 5.31696165747918e-07,
      "loss": 0.2748399257659912,
      "num_input_tokens_seen": 414411016,
      "step": 25305,
      "train_runtime": 48750.0621,
      "train_tokens_per_second": 8500.728
    },
    {
      "epoch": 1.704147589550229,
      "grad_norm": 1.0977403026723245,
      "learning_rate": 5.305102767651393e-07,
      "loss": 0.2540059328079224,
      "num_input_tokens_seen": 414492936,
      "step": 25310,
      "train_runtime": 48759.7172,
      "train_tokens_per_second": 8500.725
    },
    {
      "epoch": 1.704484244546189,
      "grad_norm": 1.2777253969391165,
      "learning_rate": 5.293256376609091e-07,
      "loss": 0.30163235664367677,
      "num_input_tokens_seen": 414574856,
      "step": 25315,
      "train_runtime": 48769.4314,
      "train_tokens_per_second": 8500.711
    },
    {
      "epoch": 1.7048208995421492,
      "grad_norm": 1.0804899992363197,
      "learning_rate": 5.281422487665089e-07,
      "loss": 0.2785039901733398,
      "num_input_tokens_seen": 414656776,
      "step": 25320,
      "train_runtime": 48778.9376,
      "train_tokens_per_second": 8500.734
    },
    {
      "epoch": 1.7051575545381095,
      "grad_norm": 0.9893822285289087,
      "learning_rate": 5.269601104128674e-07,
      "loss": 0.2509139537811279,
      "num_input_tokens_seen": 414738696,
      "step": 25325,
      "train_runtime": 48788.4463,
      "train_tokens_per_second": 8500.756
    },
    {
      "epoch": 1.7054942095340695,
      "grad_norm": 1.0625312749099096,
      "learning_rate": 5.257792229305702e-07,
      "loss": 0.2840590953826904,
      "num_input_tokens_seen": 414819704,
      "step": 25330,
      "train_runtime": 48798.1224,
      "train_tokens_per_second": 8500.731
    },
    {
      "epoch": 1.7058308645300295,
      "grad_norm": 1.1324778722936577,
      "learning_rate": 5.245995866498438e-07,
      "loss": 0.25429444313049315,
      "num_input_tokens_seen": 414901624,
      "step": 25335,
      "train_runtime": 48807.6303,
      "train_tokens_per_second": 8500.753
    },
    {
      "epoch": 1.7061675195259898,
      "grad_norm": 1.163226659997658,
      "learning_rate": 5.234212019005741e-07,
      "loss": 0.2939574956893921,
      "num_input_tokens_seen": 414983544,
      "step": 25340,
      "train_runtime": 48817.1258,
      "train_tokens_per_second": 8500.778
    },
    {
      "epoch": 1.70650417452195,
      "grad_norm": 1.1137787854766186,
      "learning_rate": 5.222440690122882e-07,
      "loss": 0.2853169679641724,
      "num_input_tokens_seen": 415065464,
      "step": 25345,
      "train_runtime": 48826.6313,
      "train_tokens_per_second": 8500.801
    },
    {
      "epoch": 1.70684082951791,
      "grad_norm": 1.063568406981526,
      "learning_rate": 5.210681883141716e-07,
      "loss": 0.26641316413879396,
      "num_input_tokens_seen": 415147384,
      "step": 25350,
      "train_runtime": 48836.1281,
      "train_tokens_per_second": 8500.825
    },
    {
      "epoch": 1.70717748451387,
      "grad_norm": 1.105561373561225,
      "learning_rate": 5.19893560135053e-07,
      "loss": 0.2984982967376709,
      "num_input_tokens_seen": 415229304,
      "step": 25355,
      "train_runtime": 48845.6171,
      "train_tokens_per_second": 8500.851
    },
    {
      "epoch": 1.7075141395098303,
      "grad_norm": 1.287864461967834,
      "learning_rate": 5.187201848034146e-07,
      "loss": 0.2932382583618164,
      "num_input_tokens_seen": 415311224,
      "step": 25360,
      "train_runtime": 48855.1308,
      "train_tokens_per_second": 8500.872
    },
    {
      "epoch": 1.7078507945057906,
      "grad_norm": 1.0336033971148297,
      "learning_rate": 5.175480626473873e-07,
      "loss": 0.27900309562683107,
      "num_input_tokens_seen": 415393144,
      "step": 25365,
      "train_runtime": 48864.9252,
      "train_tokens_per_second": 8500.845
    },
    {
      "epoch": 1.7081874495017506,
      "grad_norm": 1.3852411608889588,
      "learning_rate": 5.163771939947504e-07,
      "loss": 0.3110069990158081,
      "num_input_tokens_seen": 415475064,
      "step": 25370,
      "train_runtime": 48874.467,
      "train_tokens_per_second": 8500.861
    },
    {
      "epoch": 1.7085241044977106,
      "grad_norm": 1.1748463772936033,
      "learning_rate": 5.152075791729327e-07,
      "loss": 0.2922385215759277,
      "num_input_tokens_seen": 415556984,
      "step": 25375,
      "train_runtime": 48883.9691,
      "train_tokens_per_second": 8500.885
    },
    {
      "epoch": 1.7088607594936709,
      "grad_norm": 1.079254281863408,
      "learning_rate": 5.140392185090165e-07,
      "loss": 0.2627182722091675,
      "num_input_tokens_seen": 415638432,
      "step": 25380,
      "train_runtime": 48893.9382,
      "train_tokens_per_second": 8500.817
    },
    {
      "epoch": 1.7091974144896311,
      "grad_norm": 1.232715343572229,
      "learning_rate": 5.128721123297265e-07,
      "loss": 0.2719608783721924,
      "num_input_tokens_seen": 415720352,
      "step": 25385,
      "train_runtime": 48903.4739,
      "train_tokens_per_second": 8500.835
    },
    {
      "epoch": 1.7095340694855912,
      "grad_norm": 1.274688794086155,
      "learning_rate": 5.117062609614438e-07,
      "loss": 0.2853534698486328,
      "num_input_tokens_seen": 415802272,
      "step": 25390,
      "train_runtime": 48912.9749,
      "train_tokens_per_second": 8500.858
    },
    {
      "epoch": 1.7098707244815512,
      "grad_norm": 1.2159654971205427,
      "learning_rate": 5.105416647301909e-07,
      "loss": 0.26975398063659667,
      "num_input_tokens_seen": 415884192,
      "step": 25395,
      "train_runtime": 48922.6013,
      "train_tokens_per_second": 8500.86
    },
    {
      "epoch": 1.7102073794775114,
      "grad_norm": 1.1870310022636266,
      "learning_rate": 5.093783239616468e-07,
      "loss": 0.27869083881378176,
      "num_input_tokens_seen": 415965496,
      "step": 25400,
      "train_runtime": 48932.4477,
      "train_tokens_per_second": 8500.811
    },
    {
      "epoch": 1.7105440344734717,
      "grad_norm": 1.0950102441101144,
      "learning_rate": 5.082162389811352e-07,
      "loss": 0.2578935384750366,
      "num_input_tokens_seen": 416047416,
      "step": 25405,
      "train_runtime": 48942.0015,
      "train_tokens_per_second": 8500.826
    },
    {
      "epoch": 1.7108806894694317,
      "grad_norm": 1.2201581270379611,
      "learning_rate": 5.070554101136288e-07,
      "loss": 0.26153993606567383,
      "num_input_tokens_seen": 416129336,
      "step": 25410,
      "train_runtime": 48951.5277,
      "train_tokens_per_second": 8500.845
    },
    {
      "epoch": 1.7112173444653918,
      "grad_norm": 1.1632565004200948,
      "learning_rate": 5.058958376837497e-07,
      "loss": 0.29575481414794924,
      "num_input_tokens_seen": 416211256,
      "step": 25415,
      "train_runtime": 48961.1427,
      "train_tokens_per_second": 8500.848
    },
    {
      "epoch": 1.711553999461352,
      "grad_norm": 1.1208811389796236,
      "learning_rate": 5.047375220157691e-07,
      "loss": 0.274904727935791,
      "num_input_tokens_seen": 416293176,
      "step": 25420,
      "train_runtime": 48970.7115,
      "train_tokens_per_second": 8500.86
    },
    {
      "epoch": 1.7118906544573123,
      "grad_norm": 1.1541090538410592,
      "learning_rate": 5.035804634336056e-07,
      "loss": 0.29239444732666015,
      "num_input_tokens_seen": 416375096,
      "step": 25425,
      "train_runtime": 48986.784,
      "train_tokens_per_second": 8499.743
    },
    {
      "epoch": 1.7122273094532723,
      "grad_norm": 1.055135997261213,
      "learning_rate": 5.024246622608292e-07,
      "loss": 0.27496953010559083,
      "num_input_tokens_seen": 416457016,
      "step": 25430,
      "train_runtime": 48996.3909,
      "train_tokens_per_second": 8499.749
    },
    {
      "epoch": 1.7125639644492323,
      "grad_norm": 1.3346321527211498,
      "learning_rate": 5.01270118820652e-07,
      "loss": 0.30742368698120115,
      "num_input_tokens_seen": 416538936,
      "step": 25435,
      "train_runtime": 49005.8992,
      "train_tokens_per_second": 8499.771
    },
    {
      "epoch": 1.7129006194451926,
      "grad_norm": 1.3475589026055874,
      "learning_rate": 5.001168334359429e-07,
      "loss": 0.34264636039733887,
      "num_input_tokens_seen": 416620856,
      "step": 25440,
      "train_runtime": 49015.4035,
      "train_tokens_per_second": 8499.794
    },
    {
      "epoch": 1.7132372744411528,
      "grad_norm": 1.1709614955535048,
      "learning_rate": 4.989648064292102e-07,
      "loss": 0.2932528257369995,
      "num_input_tokens_seen": 416702776,
      "step": 25445,
      "train_runtime": 49024.9188,
      "train_tokens_per_second": 8499.816
    },
    {
      "epoch": 1.7135739294371128,
      "grad_norm": 1.14707760197385,
      "learning_rate": 4.978140381226177e-07,
      "loss": 0.2727211475372314,
      "num_input_tokens_seen": 416784696,
      "step": 25450,
      "train_runtime": 49034.4332,
      "train_tokens_per_second": 8499.837
    },
    {
      "epoch": 1.7139105844330729,
      "grad_norm": 1.2108346672441248,
      "learning_rate": 4.966645288379729e-07,
      "loss": 0.28725247383117675,
      "num_input_tokens_seen": 416866616,
      "step": 25455,
      "train_runtime": 49043.9428,
      "train_tokens_per_second": 8499.859
    },
    {
      "epoch": 1.7142472394290331,
      "grad_norm": 1.1646592030873442,
      "learning_rate": 4.955162788967327e-07,
      "loss": 0.3043970108032227,
      "num_input_tokens_seen": 416948536,
      "step": 25460,
      "train_runtime": 49053.5799,
      "train_tokens_per_second": 8499.859
    },
    {
      "epoch": 1.7145838944249934,
      "grad_norm": 0.9852795493817371,
      "learning_rate": 4.943692886200013e-07,
      "loss": 0.28295848369598386,
      "num_input_tokens_seen": 417030456,
      "step": 25465,
      "train_runtime": 49063.2007,
      "train_tokens_per_second": 8499.862
    },
    {
      "epoch": 1.7149205494209534,
      "grad_norm": 1.1006668877861567,
      "learning_rate": 4.932235583285305e-07,
      "loss": 0.26774413585662843,
      "num_input_tokens_seen": 417112376,
      "step": 25470,
      "train_runtime": 49072.8291,
      "train_tokens_per_second": 8499.864
    },
    {
      "epoch": 1.7152572044169134,
      "grad_norm": 1.0723059265338064,
      "learning_rate": 4.920790883427201e-07,
      "loss": 0.2687800645828247,
      "num_input_tokens_seen": 417194296,
      "step": 25475,
      "train_runtime": 49082.5182,
      "train_tokens_per_second": 8499.855
    },
    {
      "epoch": 1.7155938594128737,
      "grad_norm": 1.020713820319347,
      "learning_rate": 4.909358789826196e-07,
      "loss": 0.29671525955200195,
      "num_input_tokens_seen": 417276216,
      "step": 25480,
      "train_runtime": 49092.0393,
      "train_tokens_per_second": 8499.875
    },
    {
      "epoch": 1.715930514408834,
      "grad_norm": 1.2936056979479547,
      "learning_rate": 4.897939305679206e-07,
      "loss": 0.3156606912612915,
      "num_input_tokens_seen": 417358136,
      "step": 25485,
      "train_runtime": 49101.7686,
      "train_tokens_per_second": 8499.86
    },
    {
      "epoch": 1.716267169404794,
      "grad_norm": 0.9307932233835466,
      "learning_rate": 4.886532434179686e-07,
      "loss": 0.2885500192642212,
      "num_input_tokens_seen": 417440056,
      "step": 25490,
      "train_runtime": 49111.4833,
      "train_tokens_per_second": 8499.846
    },
    {
      "epoch": 1.716603824400754,
      "grad_norm": 1.0563951903698279,
      "learning_rate": 4.875138178517497e-07,
      "loss": 0.21627120971679686,
      "num_input_tokens_seen": 417521448,
      "step": 25495,
      "train_runtime": 49121.4913,
      "train_tokens_per_second": 8499.771
    },
    {
      "epoch": 1.7169404793967142,
      "grad_norm": 1.2018279873139517,
      "learning_rate": 4.863756541879028e-07,
      "loss": 0.2757680892944336,
      "num_input_tokens_seen": 417603368,
      "step": 25500,
      "train_runtime": 49131.1851,
      "train_tokens_per_second": 8499.762
    },
    {
      "epoch": 1.7172771343926745,
      "grad_norm": 1.0271735527940686,
      "learning_rate": 4.852387527447117e-07,
      "loss": 0.26091513633728025,
      "num_input_tokens_seen": 417685288,
      "step": 25505,
      "train_runtime": 49140.69,
      "train_tokens_per_second": 8499.785
    },
    {
      "epoch": 1.7176137893886345,
      "grad_norm": 1.1065185087664307,
      "learning_rate": 4.84103113840107e-07,
      "loss": 0.294989275932312,
      "num_input_tokens_seen": 417767208,
      "step": 25510,
      "train_runtime": 49150.1915,
      "train_tokens_per_second": 8499.808
    },
    {
      "epoch": 1.7179504443845945,
      "grad_norm": 0.9706322689926283,
      "learning_rate": 4.829687377916664e-07,
      "loss": 0.2625583648681641,
      "num_input_tokens_seen": 417848704,
      "step": 25515,
      "train_runtime": 49160.0281,
      "train_tokens_per_second": 8499.765
    },
    {
      "epoch": 1.7182870993805548,
      "grad_norm": 1.1064621798297813,
      "learning_rate": 4.81835624916614e-07,
      "loss": 0.25977373123168945,
      "num_input_tokens_seen": 417930624,
      "step": 25520,
      "train_runtime": 49169.5661,
      "train_tokens_per_second": 8499.783
    },
    {
      "epoch": 1.718623754376515,
      "grad_norm": 1.269773337870127,
      "learning_rate": 4.807037755318211e-07,
      "loss": 0.2920831680297852,
      "num_input_tokens_seen": 418012544,
      "step": 25525,
      "train_runtime": 49179.3508,
      "train_tokens_per_second": 8499.757
    },
    {
      "epoch": 1.718960409372475,
      "grad_norm": 0.864428592969957,
      "learning_rate": 4.795731899538086e-07,
      "loss": 0.28598287105560305,
      "num_input_tokens_seen": 418094464,
      "step": 25530,
      "train_runtime": 49189.025,
      "train_tokens_per_second": 8499.751
    },
    {
      "epoch": 1.719297064368435,
      "grad_norm": 1.0360742396653662,
      "learning_rate": 4.78443868498737e-07,
      "loss": 0.2793750762939453,
      "num_input_tokens_seen": 418176384,
      "step": 25535,
      "train_runtime": 49198.5305,
      "train_tokens_per_second": 8499.774
    },
    {
      "epoch": 1.7196337193643954,
      "grad_norm": 1.163321827630431,
      "learning_rate": 4.773158114824206e-07,
      "loss": 0.2560555934906006,
      "num_input_tokens_seen": 418258304,
      "step": 25540,
      "train_runtime": 49208.0298,
      "train_tokens_per_second": 8499.798
    },
    {
      "epoch": 1.7199703743603556,
      "grad_norm": 1.303773938740725,
      "learning_rate": 4.7618901922031445e-07,
      "loss": 0.2789867162704468,
      "num_input_tokens_seen": 418340224,
      "step": 25545,
      "train_runtime": 49217.529,
      "train_tokens_per_second": 8499.822
    },
    {
      "epoch": 1.7203070293563156,
      "grad_norm": 1.1519233673199654,
      "learning_rate": 4.750634920275243e-07,
      "loss": 0.2938229084014893,
      "num_input_tokens_seen": 418422144,
      "step": 25550,
      "train_runtime": 49227.0296,
      "train_tokens_per_second": 8499.845
    },
    {
      "epoch": 1.7206436843522757,
      "grad_norm": 1.0806453883355849,
      "learning_rate": 4.739392302188001e-07,
      "loss": 0.28953988552093507,
      "num_input_tokens_seen": 418504064,
      "step": 25555,
      "train_runtime": 49236.5297,
      "train_tokens_per_second": 8499.869
    },
    {
      "epoch": 1.720980339348236,
      "grad_norm": 1.0498128376041842,
      "learning_rate": 4.728162341085368e-07,
      "loss": 0.27934799194335935,
      "num_input_tokens_seen": 418585984,
      "step": 25560,
      "train_runtime": 49246.0353,
      "train_tokens_per_second": 8499.892
    },
    {
      "epoch": 1.7213169943441962,
      "grad_norm": 1.1006091246027188,
      "learning_rate": 4.7169450401077843e-07,
      "loss": 0.26956226825714114,
      "num_input_tokens_seen": 418667904,
      "step": 25565,
      "train_runtime": 49255.5348,
      "train_tokens_per_second": 8499.916
    },
    {
      "epoch": 1.7216536493401562,
      "grad_norm": 1.1150947577527657,
      "learning_rate": 4.7057404023921174e-07,
      "loss": 0.30483641624450686,
      "num_input_tokens_seen": 418749824,
      "step": 25570,
      "train_runtime": 49265.0398,
      "train_tokens_per_second": 8499.939
    },
    {
      "epoch": 1.7219903043361162,
      "grad_norm": 1.202478143251124,
      "learning_rate": 4.694548431071705e-07,
      "loss": 0.2801372051239014,
      "num_input_tokens_seen": 418831744,
      "step": 25575,
      "train_runtime": 49274.7535,
      "train_tokens_per_second": 8499.926
    },
    {
      "epoch": 1.7223269593320765,
      "grad_norm": 1.1586611495099346,
      "learning_rate": 4.683369129276377e-07,
      "loss": 0.27741222381591796,
      "num_input_tokens_seen": 418913664,
      "step": 25580,
      "train_runtime": 49284.451,
      "train_tokens_per_second": 8499.915
    },
    {
      "epoch": 1.7226636143280367,
      "grad_norm": 0.9882294094086101,
      "learning_rate": 4.6722025001323444e-07,
      "loss": 0.22977652549743652,
      "num_input_tokens_seen": 418995584,
      "step": 25585,
      "train_runtime": 49294.0075,
      "train_tokens_per_second": 8499.929
    },
    {
      "epoch": 1.7230002693239967,
      "grad_norm": 1.2197764449632478,
      "learning_rate": 4.6610485467623643e-07,
      "loss": 0.30107898712158204,
      "num_input_tokens_seen": 419077504,
      "step": 25590,
      "train_runtime": 49303.8111,
      "train_tokens_per_second": 8499.901
    },
    {
      "epoch": 1.7233369243199568,
      "grad_norm": 1.2101509940638668,
      "learning_rate": 4.649907272285564e-07,
      "loss": 0.2983065605163574,
      "num_input_tokens_seen": 419159424,
      "step": 25595,
      "train_runtime": 49313.5869,
      "train_tokens_per_second": 8499.877
    },
    {
      "epoch": 1.723673579315917,
      "grad_norm": 1.2530681471619738,
      "learning_rate": 4.638778679817596e-07,
      "loss": 0.2524822235107422,
      "num_input_tokens_seen": 419241344,
      "step": 25600,
      "train_runtime": 49323.2118,
      "train_tokens_per_second": 8499.879
    },
    {
      "epoch": 1.7240102343118773,
      "grad_norm": 1.090365468622243,
      "learning_rate": 4.6276627724705204e-07,
      "loss": 0.2953644752502441,
      "num_input_tokens_seen": 419323264,
      "step": 25605,
      "train_runtime": 49332.841,
      "train_tokens_per_second": 8499.881
    },
    {
      "epoch": 1.7243468893078373,
      "grad_norm": 1.1946984515155183,
      "learning_rate": 4.616559553352873e-07,
      "loss": 0.2589264392852783,
      "num_input_tokens_seen": 419405184,
      "step": 25610,
      "train_runtime": 49342.522,
      "train_tokens_per_second": 8499.873
    },
    {
      "epoch": 1.7246835443037973,
      "grad_norm": 1.3204029945847953,
      "learning_rate": 4.6054690255696256e-07,
      "loss": 0.3149195671081543,
      "num_input_tokens_seen": 419487104,
      "step": 25615,
      "train_runtime": 49352.1856,
      "train_tokens_per_second": 8499.869
    },
    {
      "epoch": 1.7250201992997576,
      "grad_norm": 1.174316054164826,
      "learning_rate": 4.594391192222214e-07,
      "loss": 0.26962804794311523,
      "num_input_tokens_seen": 419568680,
      "step": 25620,
      "train_runtime": 49362.2367,
      "train_tokens_per_second": 8499.791
    },
    {
      "epoch": 1.7253568542957178,
      "grad_norm": 1.0653794991484855,
      "learning_rate": 4.583326056408505e-07,
      "loss": 0.24897661209106445,
      "num_input_tokens_seen": 419650600,
      "step": 25625,
      "train_runtime": 49371.9089,
      "train_tokens_per_second": 8499.785
    },
    {
      "epoch": 1.7256935092916779,
      "grad_norm": 1.2714608343511808,
      "learning_rate": 4.5722736212228613e-07,
      "loss": 0.2818833112716675,
      "num_input_tokens_seen": 419732520,
      "step": 25630,
      "train_runtime": 49381.4395,
      "train_tokens_per_second": 8499.803
    },
    {
      "epoch": 1.726030164287638,
      "grad_norm": 1.2517975056455846,
      "learning_rate": 4.5612338897560227e-07,
      "loss": 0.28359410762786863,
      "num_input_tokens_seen": 419814440,
      "step": 25635,
      "train_runtime": 49390.9618,
      "train_tokens_per_second": 8499.823
    },
    {
      "epoch": 1.7263668192835981,
      "grad_norm": 1.2178186821257153,
      "learning_rate": 4.5502068650952527e-07,
      "loss": 0.3094528913497925,
      "num_input_tokens_seen": 419896360,
      "step": 25640,
      "train_runtime": 49400.6015,
      "train_tokens_per_second": 8499.823
    },
    {
      "epoch": 1.7267034742795584,
      "grad_norm": 1.1163337248802416,
      "learning_rate": 4.5391925503241894e-07,
      "loss": 0.2904594898223877,
      "num_input_tokens_seen": 419978280,
      "step": 25645,
      "train_runtime": 49410.1673,
      "train_tokens_per_second": 8499.835
    },
    {
      "epoch": 1.7270401292755184,
      "grad_norm": 1.3457894787705302,
      "learning_rate": 4.5281909485229704e-07,
      "loss": 0.2813404083251953,
      "num_input_tokens_seen": 420060200,
      "step": 25650,
      "train_runtime": 49419.7586,
      "train_tokens_per_second": 8499.843
    },
    {
      "epoch": 1.7273767842714784,
      "grad_norm": 1.0382801360803462,
      "learning_rate": 4.51720206276815e-07,
      "loss": 0.24936649799346924,
      "num_input_tokens_seen": 420142120,
      "step": 25655,
      "train_runtime": 49429.4158,
      "train_tokens_per_second": 8499.84
    },
    {
      "epoch": 1.7277134392674387,
      "grad_norm": 1.2437477487570727,
      "learning_rate": 4.506225896132743e-07,
      "loss": 0.25764305591583253,
      "num_input_tokens_seen": 420224040,
      "step": 25660,
      "train_runtime": 49439.0529,
      "train_tokens_per_second": 8499.84
    },
    {
      "epoch": 1.728050094263399,
      "grad_norm": 1.3539612709903985,
      "learning_rate": 4.4952624516861944e-07,
      "loss": 0.2956719398498535,
      "num_input_tokens_seen": 420305960,
      "step": 25665,
      "train_runtime": 49448.7712,
      "train_tokens_per_second": 8499.826
    },
    {
      "epoch": 1.728386749259359,
      "grad_norm": 0.8218069892162715,
      "learning_rate": 4.4843117324944005e-07,
      "loss": 0.24618310928344728,
      "num_input_tokens_seen": 420387880,
      "step": 25670,
      "train_runtime": 49458.4576,
      "train_tokens_per_second": 8499.818
    },
    {
      "epoch": 1.728723404255319,
      "grad_norm": 1.1023968980479872,
      "learning_rate": 4.4733737416196745e-07,
      "loss": 0.2703487157821655,
      "num_input_tokens_seen": 420469800,
      "step": 25675,
      "train_runtime": 49467.9559,
      "train_tokens_per_second": 8499.842
    },
    {
      "epoch": 1.7290600592512793,
      "grad_norm": 0.8741078511912121,
      "learning_rate": 4.4624484821208346e-07,
      "loss": 0.2596503734588623,
      "num_input_tokens_seen": 420551720,
      "step": 25680,
      "train_runtime": 49477.5012,
      "train_tokens_per_second": 8499.858
    },
    {
      "epoch": 1.7293967142472395,
      "grad_norm": 0.9504917131025605,
      "learning_rate": 4.451535957053044e-07,
      "loss": 0.2408958911895752,
      "num_input_tokens_seen": 420633640,
      "step": 25685,
      "train_runtime": 49487.1085,
      "train_tokens_per_second": 8499.863
    },
    {
      "epoch": 1.7297333692431995,
      "grad_norm": 1.1404211926365326,
      "learning_rate": 4.440636169467999e-07,
      "loss": 0.27260408401489256,
      "num_input_tokens_seen": 420715560,
      "step": 25690,
      "train_runtime": 49496.6307,
      "train_tokens_per_second": 8499.883
    },
    {
      "epoch": 1.7300700242391596,
      "grad_norm": 1.0394469199238412,
      "learning_rate": 4.429749122413757e-07,
      "loss": 0.28365521430969237,
      "num_input_tokens_seen": 420797480,
      "step": 25695,
      "train_runtime": 49506.2463,
      "train_tokens_per_second": 8499.887
    },
    {
      "epoch": 1.7304066792351198,
      "grad_norm": 1.120076954039208,
      "learning_rate": 4.418874818934865e-07,
      "loss": 0.2464388132095337,
      "num_input_tokens_seen": 420879400,
      "step": 25700,
      "train_runtime": 49515.7411,
      "train_tokens_per_second": 8499.911
    },
    {
      "epoch": 1.73074333423108,
      "grad_norm": 1.0971766264438734,
      "learning_rate": 4.4080132620722806e-07,
      "loss": 0.2982363700866699,
      "num_input_tokens_seen": 420961320,
      "step": 25705,
      "train_runtime": 49525.2525,
      "train_tokens_per_second": 8499.933
    },
    {
      "epoch": 1.73107998922704,
      "grad_norm": 0.9825489040274827,
      "learning_rate": 4.3971644548634116e-07,
      "loss": 0.2774411678314209,
      "num_input_tokens_seen": 421042976,
      "step": 25710,
      "train_runtime": 49534.7471,
      "train_tokens_per_second": 8499.952
    },
    {
      "epoch": 1.7314166442230001,
      "grad_norm": 1.4539698010471387,
      "learning_rate": 4.3863284003420826e-07,
      "loss": 0.2712681770324707,
      "num_input_tokens_seen": 421124896,
      "step": 25715,
      "train_runtime": 49544.2458,
      "train_tokens_per_second": 8499.976
    },
    {
      "epoch": 1.7317532992189604,
      "grad_norm": 1.0488341019788392,
      "learning_rate": 4.375505101538563e-07,
      "loss": 0.2729413270950317,
      "num_input_tokens_seen": 421206816,
      "step": 25720,
      "train_runtime": 49553.897,
      "train_tokens_per_second": 8499.974
    },
    {
      "epoch": 1.7320899542149206,
      "grad_norm": 1.3057119125857248,
      "learning_rate": 4.3646945614795487e-07,
      "loss": 0.2855130910873413,
      "num_input_tokens_seen": 421288736,
      "step": 25725,
      "train_runtime": 49563.408,
      "train_tokens_per_second": 8499.995
    },
    {
      "epoch": 1.7324266092108807,
      "grad_norm": 1.2329842709640708,
      "learning_rate": 4.3538967831881995e-07,
      "loss": 0.29763901233673096,
      "num_input_tokens_seen": 421370656,
      "step": 25730,
      "train_runtime": 49573.0895,
      "train_tokens_per_second": 8499.988
    },
    {
      "epoch": 1.7327632642068407,
      "grad_norm": 1.0608883523090265,
      "learning_rate": 4.343111769684039e-07,
      "loss": 0.27895541191101075,
      "num_input_tokens_seen": 421452576,
      "step": 25735,
      "train_runtime": 49582.6,
      "train_tokens_per_second": 8500.01
    },
    {
      "epoch": 1.733099919202801,
      "grad_norm": 1.1463652029228595,
      "learning_rate": 4.332339523983103e-07,
      "loss": 0.2837217807769775,
      "num_input_tokens_seen": 421534496,
      "step": 25740,
      "train_runtime": 49592.164,
      "train_tokens_per_second": 8500.022
    },
    {
      "epoch": 1.7334365741987612,
      "grad_norm": 1.120958979237641,
      "learning_rate": 4.321580049097773e-07,
      "loss": 0.2743642330169678,
      "num_input_tokens_seen": 421616112,
      "step": 25745,
      "train_runtime": 49602.1153,
      "train_tokens_per_second": 8499.962
    },
    {
      "epoch": 1.7337732291947212,
      "grad_norm": 1.1996334083610125,
      "learning_rate": 4.310833348036941e-07,
      "loss": 0.2682629585266113,
      "num_input_tokens_seen": 421698032,
      "step": 25750,
      "train_runtime": 49611.6258,
      "train_tokens_per_second": 8499.984
    },
    {
      "epoch": 1.7341098841906812,
      "grad_norm": 1.1606448426127902,
      "learning_rate": 4.300099423805865e-07,
      "loss": 0.25842900276184083,
      "num_input_tokens_seen": 421779952,
      "step": 25755,
      "train_runtime": 49621.3286,
      "train_tokens_per_second": 8499.973
    },
    {
      "epoch": 1.7344465391866415,
      "grad_norm": 1.1781562599965418,
      "learning_rate": 4.289378279406264e-07,
      "loss": 0.287480092048645,
      "num_input_tokens_seen": 421861064,
      "step": 25760,
      "train_runtime": 49631.1959,
      "train_tokens_per_second": 8499.917
    },
    {
      "epoch": 1.7347831941826017,
      "grad_norm": 1.1381428976170558,
      "learning_rate": 4.2786699178362647e-07,
      "loss": 0.29442338943481444,
      "num_input_tokens_seen": 421942984,
      "step": 25765,
      "train_runtime": 49640.705,
      "train_tokens_per_second": 8499.939
    },
    {
      "epoch": 1.7351198491785618,
      "grad_norm": 1.410970837330573,
      "learning_rate": 4.2679743420904254e-07,
      "loss": 0.2872114658355713,
      "num_input_tokens_seen": 422024904,
      "step": 25770,
      "train_runtime": 49650.3359,
      "train_tokens_per_second": 8499.941
    },
    {
      "epoch": 1.7354565041745218,
      "grad_norm": 1.0590837655538818,
      "learning_rate": 4.257291555159726e-07,
      "loss": 0.26131129264831543,
      "num_input_tokens_seen": 422106824,
      "step": 25775,
      "train_runtime": 49659.8549,
      "train_tokens_per_second": 8499.961
    },
    {
      "epoch": 1.735793159170482,
      "grad_norm": 1.6049018872629814,
      "learning_rate": 4.246621560031594e-07,
      "loss": 0.27906877994537355,
      "num_input_tokens_seen": 422188744,
      "step": 25780,
      "train_runtime": 49669.4824,
      "train_tokens_per_second": 8499.963
    },
    {
      "epoch": 1.7361298141664423,
      "grad_norm": 1.203827236554391,
      "learning_rate": 4.235964359689837e-07,
      "loss": 0.2719066858291626,
      "num_input_tokens_seen": 422270664,
      "step": 25785,
      "train_runtime": 49679.1089,
      "train_tokens_per_second": 8499.965
    },
    {
      "epoch": 1.7364664691624023,
      "grad_norm": 1.3476745605919551,
      "learning_rate": 4.225319957114726e-07,
      "loss": 0.3073368787765503,
      "num_input_tokens_seen": 422352584,
      "step": 25790,
      "train_runtime": 49688.6181,
      "train_tokens_per_second": 8499.987
    },
    {
      "epoch": 1.7368031241583624,
      "grad_norm": 1.0791309401536417,
      "learning_rate": 4.214688355282909e-07,
      "loss": 0.2616889953613281,
      "num_input_tokens_seen": 422434504,
      "step": 25795,
      "train_runtime": 49698.1194,
      "train_tokens_per_second": 8500.01
    },
    {
      "epoch": 1.7371397791543226,
      "grad_norm": 1.0691219325217696,
      "learning_rate": 4.204069557167506e-07,
      "loss": 0.2711873292922974,
      "num_input_tokens_seen": 422516424,
      "step": 25800,
      "train_runtime": 49707.6324,
      "train_tokens_per_second": 8500.031
    },
    {
      "epoch": 1.7374764341502829,
      "grad_norm": 1.1428903814797684,
      "learning_rate": 4.1934635657380153e-07,
      "loss": 0.3034448862075806,
      "num_input_tokens_seen": 422598344,
      "step": 25805,
      "train_runtime": 49717.1355,
      "train_tokens_per_second": 8500.054
    },
    {
      "epoch": 1.7378130891462429,
      "grad_norm": 1.1598431390992467,
      "learning_rate": 4.1828703839603637e-07,
      "loss": 0.2732102870941162,
      "num_input_tokens_seen": 422680264,
      "step": 25810,
      "train_runtime": 49726.7745,
      "train_tokens_per_second": 8500.054
    },
    {
      "epoch": 1.738149744142203,
      "grad_norm": 1.2812504727917924,
      "learning_rate": 4.172290014796926e-07,
      "loss": 0.2952268123626709,
      "num_input_tokens_seen": 422762184,
      "step": 25815,
      "train_runtime": 49736.3648,
      "train_tokens_per_second": 8500.062
    },
    {
      "epoch": 1.7384863991381632,
      "grad_norm": 1.1444779097002524,
      "learning_rate": 4.1617224612064353e-07,
      "loss": 0.2914626121520996,
      "num_input_tokens_seen": 422844104,
      "step": 25820,
      "train_runtime": 49745.8639,
      "train_tokens_per_second": 8500.086
    },
    {
      "epoch": 1.7388230541341234,
      "grad_norm": 1.229520708011813,
      "learning_rate": 4.151167726144101e-07,
      "loss": 0.277435827255249,
      "num_input_tokens_seen": 422926024,
      "step": 25825,
      "train_runtime": 49755.3774,
      "train_tokens_per_second": 8500.107
    },
    {
      "epoch": 1.7391597091300834,
      "grad_norm": 1.390777335855439,
      "learning_rate": 4.1406258125615106e-07,
      "loss": 0.2713778972625732,
      "num_input_tokens_seen": 423007944,
      "step": 25830,
      "train_runtime": 49765.0718,
      "train_tokens_per_second": 8500.097
    },
    {
      "epoch": 1.7394963641260435,
      "grad_norm": 1.159642883344804,
      "learning_rate": 4.130096723406674e-07,
      "loss": 0.26486155986785886,
      "num_input_tokens_seen": 423089864,
      "step": 25835,
      "train_runtime": 49774.8073,
      "train_tokens_per_second": 8500.08
    },
    {
      "epoch": 1.7398330191220037,
      "grad_norm": 1.0832272145736692,
      "learning_rate": 4.119580461624023e-07,
      "loss": 0.2909510612487793,
      "num_input_tokens_seen": 423171784,
      "step": 25840,
      "train_runtime": 49784.482,
      "train_tokens_per_second": 8500.074
    },
    {
      "epoch": 1.740169674117964,
      "grad_norm": 1.288355952778011,
      "learning_rate": 4.109077030154396e-07,
      "loss": 0.30022611618041994,
      "num_input_tokens_seen": 423253704,
      "step": 25845,
      "train_runtime": 49794.141,
      "train_tokens_per_second": 8500.07
    },
    {
      "epoch": 1.740506329113924,
      "grad_norm": 1.099135532652641,
      "learning_rate": 4.0985864319350366e-07,
      "loss": 0.29097795486450195,
      "num_input_tokens_seen": 423335624,
      "step": 25850,
      "train_runtime": 49803.65,
      "train_tokens_per_second": 8500.092
    },
    {
      "epoch": 1.740842984109884,
      "grad_norm": 1.1444072638526939,
      "learning_rate": 4.0881086698996154e-07,
      "loss": 0.27391667366027833,
      "num_input_tokens_seen": 423417544,
      "step": 25855,
      "train_runtime": 49813.1531,
      "train_tokens_per_second": 8500.115
    },
    {
      "epoch": 1.7411796391058443,
      "grad_norm": 1.1064975828992913,
      "learning_rate": 4.077643746978194e-07,
      "loss": 0.3052505016326904,
      "num_input_tokens_seen": 423499464,
      "step": 25860,
      "train_runtime": 49822.6512,
      "train_tokens_per_second": 8500.139
    },
    {
      "epoch": 1.7415162941018045,
      "grad_norm": 1.0610911294975438,
      "learning_rate": 4.067191666097281e-07,
      "loss": 0.27588233947753904,
      "num_input_tokens_seen": 423581384,
      "step": 25865,
      "train_runtime": 49832.2838,
      "train_tokens_per_second": 8500.14
    },
    {
      "epoch": 1.7418529490977646,
      "grad_norm": 1.0979517654940212,
      "learning_rate": 4.056752430179728e-07,
      "loss": 0.24563975334167482,
      "num_input_tokens_seen": 423663304,
      "step": 25870,
      "train_runtime": 49841.955,
      "train_tokens_per_second": 8500.134
    },
    {
      "epoch": 1.7421896040937246,
      "grad_norm": 1.010244762016623,
      "learning_rate": 4.046326042144866e-07,
      "loss": 0.2597576379776001,
      "num_input_tokens_seen": 423745224,
      "step": 25875,
      "train_runtime": 49851.614,
      "train_tokens_per_second": 8500.13
    },
    {
      "epoch": 1.7425262590896848,
      "grad_norm": 1.0655490513293737,
      "learning_rate": 4.0359125049083956e-07,
      "loss": 0.24855942726135255,
      "num_input_tokens_seen": 423827144,
      "step": 25880,
      "train_runtime": 49861.2638,
      "train_tokens_per_second": 8500.128
    },
    {
      "epoch": 1.742862914085645,
      "grad_norm": 1.1043343329793553,
      "learning_rate": 4.025511821382422e-07,
      "loss": 0.2824984550476074,
      "num_input_tokens_seen": 423909064,
      "step": 25885,
      "train_runtime": 49870.763,
      "train_tokens_per_second": 8500.152
    },
    {
      "epoch": 1.7431995690816051,
      "grad_norm": 0.9986394040445119,
      "learning_rate": 4.015123994475462e-07,
      "loss": 0.2613039016723633,
      "num_input_tokens_seen": 423990984,
      "step": 25890,
      "train_runtime": 49880.2741,
      "train_tokens_per_second": 8500.173
    },
    {
      "epoch": 1.7435362240775654,
      "grad_norm": 1.0511531706166728,
      "learning_rate": 4.0047490270924417e-07,
      "loss": 0.2740230321884155,
      "num_input_tokens_seen": 424072904,
      "step": 25895,
      "train_runtime": 49889.8893,
      "train_tokens_per_second": 8500.177
    },
    {
      "epoch": 1.7438728790735256,
      "grad_norm": 1.1068713750110246,
      "learning_rate": 3.994386922134691e-07,
      "loss": 0.29378037452697753,
      "num_input_tokens_seen": 424154824,
      "step": 25900,
      "train_runtime": 49912.22,
      "train_tokens_per_second": 8498.016
    },
    {
      "epoch": 1.7442095340694856,
      "grad_norm": 1.0346774895825905,
      "learning_rate": 3.984037682499936e-07,
      "loss": 0.27312479019165037,
      "num_input_tokens_seen": 424236744,
      "step": 25905,
      "train_runtime": 49924.343,
      "train_tokens_per_second": 8497.593
    },
    {
      "epoch": 1.7445461890654457,
      "grad_norm": 2.317384495503529,
      "learning_rate": 3.973701311082301e-07,
      "loss": 0.27653326988220217,
      "num_input_tokens_seen": 424318664,
      "step": 25910,
      "train_runtime": 49933.9804,
      "train_tokens_per_second": 8497.593
    },
    {
      "epoch": 1.744882844061406,
      "grad_norm": 1.0180284343842203,
      "learning_rate": 3.9633778107723454e-07,
      "loss": 0.262326717376709,
      "num_input_tokens_seen": 424400584,
      "step": 25915,
      "train_runtime": 49943.4676,
      "train_tokens_per_second": 8497.619
    },
    {
      "epoch": 1.7452194990573662,
      "grad_norm": 1.0661386171784832,
      "learning_rate": 3.953067184456966e-07,
      "loss": 0.2844232082366943,
      "num_input_tokens_seen": 424482504,
      "step": 25920,
      "train_runtime": 49952.9727,
      "train_tokens_per_second": 8497.643
    },
    {
      "epoch": 1.7455561540533262,
      "grad_norm": 1.0830474967552115,
      "learning_rate": 3.9427694350195224e-07,
      "loss": 0.28239855766296384,
      "num_input_tokens_seen": 424564424,
      "step": 25925,
      "train_runtime": 49962.4648,
      "train_tokens_per_second": 8497.668
    },
    {
      "epoch": 1.7458928090492862,
      "grad_norm": 1.0636932798017233,
      "learning_rate": 3.9324845653397446e-07,
      "loss": 0.2972452640533447,
      "num_input_tokens_seen": 424645616,
      "step": 25930,
      "train_runtime": 49972.3206,
      "train_tokens_per_second": 8497.616
    },
    {
      "epoch": 1.7462294640452465,
      "grad_norm": 1.181254965937359,
      "learning_rate": 3.922212578293755e-07,
      "loss": 0.27524452209472655,
      "num_input_tokens_seen": 424727536,
      "step": 25935,
      "train_runtime": 49981.882,
      "train_tokens_per_second": 8497.63
    },
    {
      "epoch": 1.7465661190412067,
      "grad_norm": 1.2298790531681512,
      "learning_rate": 3.911953476754088e-07,
      "loss": 0.27483367919921875,
      "num_input_tokens_seen": 424809456,
      "step": 25940,
      "train_runtime": 49991.5552,
      "train_tokens_per_second": 8497.624
    },
    {
      "epoch": 1.7469027740371668,
      "grad_norm": 1.2240917523583832,
      "learning_rate": 3.9017072635896716e-07,
      "loss": 0.2930711269378662,
      "num_input_tokens_seen": 424891376,
      "step": 25945,
      "train_runtime": 50001.224,
      "train_tokens_per_second": 8497.619
    },
    {
      "epoch": 1.7472394290331268,
      "grad_norm": 1.232526760065908,
      "learning_rate": 3.891473941665819e-07,
      "loss": 0.3145125389099121,
      "num_input_tokens_seen": 424973296,
      "step": 25950,
      "train_runtime": 50010.9078,
      "train_tokens_per_second": 8497.612
    },
    {
      "epoch": 1.747576084029087,
      "grad_norm": 1.1120995911874167,
      "learning_rate": 3.881253513844246e-07,
      "loss": 0.26698150634765627,
      "num_input_tokens_seen": 425054688,
      "step": 25955,
      "train_runtime": 50020.9235,
      "train_tokens_per_second": 8497.538
    },
    {
      "epoch": 1.7479127390250473,
      "grad_norm": 1.1584463919318062,
      "learning_rate": 3.871045982983063e-07,
      "loss": 0.28206896781921387,
      "num_input_tokens_seen": 425136608,
      "step": 25960,
      "train_runtime": 50030.4733,
      "train_tokens_per_second": 8497.553
    },
    {
      "epoch": 1.7482493940210073,
      "grad_norm": 0.9879949281368683,
      "learning_rate": 3.8608513519367897e-07,
      "loss": 0.24982054233551027,
      "num_input_tokens_seen": 425218528,
      "step": 25965,
      "train_runtime": 50040.0577,
      "train_tokens_per_second": 8497.563
    },
    {
      "epoch": 1.7485860490169673,
      "grad_norm": 1.4065601709802265,
      "learning_rate": 3.8506696235562915e-07,
      "loss": 0.26964535713195803,
      "num_input_tokens_seen": 425300448,
      "step": 25970,
      "train_runtime": 50049.7579,
      "train_tokens_per_second": 8497.553
    },
    {
      "epoch": 1.7489227040129276,
      "grad_norm": 1.1597703092424274,
      "learning_rate": 3.840500800688884e-07,
      "loss": 0.2812522888183594,
      "num_input_tokens_seen": 425382368,
      "step": 25975,
      "train_runtime": 50059.4151,
      "train_tokens_per_second": 8497.55
    },
    {
      "epoch": 1.7492593590088878,
      "grad_norm": 1.1206345140626406,
      "learning_rate": 3.8303448861782333e-07,
      "loss": 0.2894416570663452,
      "num_input_tokens_seen": 425464288,
      "step": 25980,
      "train_runtime": 50069.0518,
      "train_tokens_per_second": 8497.55
    },
    {
      "epoch": 1.7495960140048479,
      "grad_norm": 1.133215072481581,
      "learning_rate": 3.8202018828644105e-07,
      "loss": 0.30473804473876953,
      "num_input_tokens_seen": 425546208,
      "step": 25985,
      "train_runtime": 50078.5563,
      "train_tokens_per_second": 8497.573
    },
    {
      "epoch": 1.749932669000808,
      "grad_norm": 1.1981492342964892,
      "learning_rate": 3.8100717935838804e-07,
      "loss": 0.31113734245300295,
      "num_input_tokens_seen": 425628128,
      "step": 25990,
      "train_runtime": 50088.1557,
      "train_tokens_per_second": 8497.58
    },
    {
      "epoch": 1.7502693239967682,
      "grad_norm": 1.070504938386809,
      "learning_rate": 3.799954621169477e-07,
      "loss": 0.24412322044372559,
      "num_input_tokens_seen": 425710048,
      "step": 25995,
      "train_runtime": 50097.7476,
      "train_tokens_per_second": 8497.589
    },
    {
      "epoch": 1.7506059789927284,
      "grad_norm": 1.1750863455920542,
      "learning_rate": 3.7898503684504483e-07,
      "loss": 0.2701544284820557,
      "num_input_tokens_seen": 425791624,
      "step": 26000,
      "train_runtime": 50107.7246,
      "train_tokens_per_second": 8497.525
    },
    {
      "epoch": 1.7509426339886884,
      "grad_norm": 0.9654101096595658,
      "learning_rate": 3.7797590382524066e-07,
      "loss": 0.26930689811706543,
      "num_input_tokens_seen": 425873544,
      "step": 26005,
      "train_runtime": 50117.232,
      "train_tokens_per_second": 8497.547
    },
    {
      "epoch": 1.7512792889846485,
      "grad_norm": 1.0976379538317322,
      "learning_rate": 3.76968063339736e-07,
      "loss": 0.29122223854064944,
      "num_input_tokens_seen": 425955464,
      "step": 26010,
      "train_runtime": 50126.9356,
      "train_tokens_per_second": 8497.536
    },
    {
      "epoch": 1.7516159439806087,
      "grad_norm": 1.2940270555306912,
      "learning_rate": 3.759615156703722e-07,
      "loss": 0.2803593635559082,
      "num_input_tokens_seen": 426037384,
      "step": 26015,
      "train_runtime": 50136.8609,
      "train_tokens_per_second": 8497.488
    },
    {
      "epoch": 1.751952598976569,
      "grad_norm": 1.0288229320100282,
      "learning_rate": 3.749562610986235e-07,
      "loss": 0.28068602085113525,
      "num_input_tokens_seen": 426119304,
      "step": 26020,
      "train_runtime": 50146.5089,
      "train_tokens_per_second": 8497.487
    },
    {
      "epoch": 1.752289253972529,
      "grad_norm": 1.1516807295854103,
      "learning_rate": 3.739522999056094e-07,
      "loss": 0.26851696968078614,
      "num_input_tokens_seen": 426201224,
      "step": 26025,
      "train_runtime": 50156.0058,
      "train_tokens_per_second": 8497.511
    },
    {
      "epoch": 1.752625908968489,
      "grad_norm": 1.1506606994854645,
      "learning_rate": 3.7294963237208326e-07,
      "loss": 0.2817866086959839,
      "num_input_tokens_seen": 426283144,
      "step": 26030,
      "train_runtime": 50165.5061,
      "train_tokens_per_second": 8497.535
    },
    {
      "epoch": 1.7529625639644493,
      "grad_norm": 1.1132553648792778,
      "learning_rate": 3.7194825877843786e-07,
      "loss": 0.2775230646133423,
      "num_input_tokens_seen": 426365064,
      "step": 26035,
      "train_runtime": 50175.123,
      "train_tokens_per_second": 8497.539
    },
    {
      "epoch": 1.7532992189604095,
      "grad_norm": 1.1040907776851887,
      "learning_rate": 3.7094817940470375e-07,
      "loss": 0.2716811180114746,
      "num_input_tokens_seen": 426446984,
      "step": 26040,
      "train_runtime": 50184.8121,
      "train_tokens_per_second": 8497.531
    },
    {
      "epoch": 1.7536358739563696,
      "grad_norm": 1.0722977238334737,
      "learning_rate": 3.6994939453055044e-07,
      "loss": 0.28867387771606445,
      "num_input_tokens_seen": 426528904,
      "step": 26045,
      "train_runtime": 50194.4435,
      "train_tokens_per_second": 8497.532
    },
    {
      "epoch": 1.7539725289523296,
      "grad_norm": 1.2943738026096596,
      "learning_rate": 3.6895190443528515e-07,
      "loss": 0.2684515953063965,
      "num_input_tokens_seen": 426610824,
      "step": 26050,
      "train_runtime": 50203.9413,
      "train_tokens_per_second": 8497.556
    },
    {
      "epoch": 1.7543091839482898,
      "grad_norm": 1.1551080689803608,
      "learning_rate": 3.679557093978525e-07,
      "loss": 0.3072051048278809,
      "num_input_tokens_seen": 426692744,
      "step": 26055,
      "train_runtime": 50213.5635,
      "train_tokens_per_second": 8497.56
    },
    {
      "epoch": 1.75464583894425,
      "grad_norm": 1.1884384691440446,
      "learning_rate": 3.669608096968341e-07,
      "loss": 0.2763687610626221,
      "num_input_tokens_seen": 426774664,
      "step": 26060,
      "train_runtime": 50223.1881,
      "train_tokens_per_second": 8497.562
    },
    {
      "epoch": 1.75498249394021,
      "grad_norm": 1.341374429435881,
      "learning_rate": 3.6596720561045295e-07,
      "loss": 0.2875319480895996,
      "num_input_tokens_seen": 426856584,
      "step": 26065,
      "train_runtime": 50232.7203,
      "train_tokens_per_second": 8497.58
    },
    {
      "epoch": 1.7553191489361701,
      "grad_norm": 1.3323630792868535,
      "learning_rate": 3.6497489741656455e-07,
      "loss": 0.3211801052093506,
      "num_input_tokens_seen": 426938504,
      "step": 26070,
      "train_runtime": 50242.325,
      "train_tokens_per_second": 8497.587
    },
    {
      "epoch": 1.7556558039321304,
      "grad_norm": 1.328917484757415,
      "learning_rate": 3.639838853926669e-07,
      "loss": 0.3048991680145264,
      "num_input_tokens_seen": 427020424,
      "step": 26075,
      "train_runtime": 50252.6,
      "train_tokens_per_second": 8497.479
    },
    {
      "epoch": 1.7559924589280906,
      "grad_norm": 1.2516153309741584,
      "learning_rate": 3.629941698158923e-07,
      "loss": 0.3254580020904541,
      "num_input_tokens_seen": 427102344,
      "step": 26080,
      "train_runtime": 50262.5422,
      "train_tokens_per_second": 8497.428
    },
    {
      "epoch": 1.7563291139240507,
      "grad_norm": 1.0843110768396198,
      "learning_rate": 3.6200575096301207e-07,
      "loss": 0.2440805673599243,
      "num_input_tokens_seen": 427184264,
      "step": 26085,
      "train_runtime": 50272.2229,
      "train_tokens_per_second": 8497.421
    },
    {
      "epoch": 1.7566657689200107,
      "grad_norm": 1.198698218939478,
      "learning_rate": 3.61018629110434e-07,
      "loss": 0.27805390357971194,
      "num_input_tokens_seen": 427266184,
      "step": 26090,
      "train_runtime": 50281.811,
      "train_tokens_per_second": 8497.43
    },
    {
      "epoch": 1.757002423915971,
      "grad_norm": 1.017951968124504,
      "learning_rate": 3.6003280453420386e-07,
      "loss": 0.2577180862426758,
      "num_input_tokens_seen": 427348104,
      "step": 26095,
      "train_runtime": 50291.6066,
      "train_tokens_per_second": 8497.404
    },
    {
      "epoch": 1.7573390789119312,
      "grad_norm": 1.231891007547164,
      "learning_rate": 3.59048277510004e-07,
      "loss": 0.2680124998092651,
      "num_input_tokens_seen": 427430024,
      "step": 26100,
      "train_runtime": 50301.1086,
      "train_tokens_per_second": 8497.428
    },
    {
      "epoch": 1.7576757339078912,
      "grad_norm": 1.235056785453018,
      "learning_rate": 3.5806504831315423e-07,
      "loss": 0.25974788665771487,
      "num_input_tokens_seen": 427511944,
      "step": 26105,
      "train_runtime": 50310.8177,
      "train_tokens_per_second": 8497.416
    },
    {
      "epoch": 1.7580123889038513,
      "grad_norm": 1.1518453730774685,
      "learning_rate": 3.570831172186112e-07,
      "loss": 0.2888121843338013,
      "num_input_tokens_seen": 427593864,
      "step": 26110,
      "train_runtime": 50320.6503,
      "train_tokens_per_second": 8497.384
    },
    {
      "epoch": 1.7583490438998115,
      "grad_norm": 1.1774406622507263,
      "learning_rate": 3.5610248450097084e-07,
      "loss": 0.28595101833343506,
      "num_input_tokens_seen": 427675784,
      "step": 26115,
      "train_runtime": 50330.2077,
      "train_tokens_per_second": 8497.398
    },
    {
      "epoch": 1.7586856988957718,
      "grad_norm": 1.0906566518996208,
      "learning_rate": 3.551231504344604e-07,
      "loss": 0.26587796211242676,
      "num_input_tokens_seen": 427757704,
      "step": 26120,
      "train_runtime": 50339.8141,
      "train_tokens_per_second": 8497.403
    },
    {
      "epoch": 1.7590223538917318,
      "grad_norm": 1.2337060925825791,
      "learning_rate": 3.5414511529295036e-07,
      "loss": 0.3022026062011719,
      "num_input_tokens_seen": 427839624,
      "step": 26125,
      "train_runtime": 50349.3628,
      "train_tokens_per_second": 8497.419
    },
    {
      "epoch": 1.7593590088876918,
      "grad_norm": 1.3819315533722611,
      "learning_rate": 3.5316837934994465e-07,
      "loss": 0.30302987098693845,
      "num_input_tokens_seen": 427921544,
      "step": 26130,
      "train_runtime": 50358.9254,
      "train_tokens_per_second": 8497.432
    },
    {
      "epoch": 1.759695663883652,
      "grad_norm": 1.0540115045325134,
      "learning_rate": 3.521929428785836e-07,
      "loss": 0.28916425704956056,
      "num_input_tokens_seen": 428003464,
      "step": 26135,
      "train_runtime": 50368.5909,
      "train_tokens_per_second": 8497.428
    },
    {
      "epoch": 1.7600323188796123,
      "grad_norm": 1.1176310890703967,
      "learning_rate": 3.5121880615164584e-07,
      "loss": 0.3058877944946289,
      "num_input_tokens_seen": 428085384,
      "step": 26140,
      "train_runtime": 50378.1331,
      "train_tokens_per_second": 8497.444
    },
    {
      "epoch": 1.7603689738755723,
      "grad_norm": 1.1786460583398712,
      "learning_rate": 3.5024596944154445e-07,
      "loss": 0.2892393827438354,
      "num_input_tokens_seen": 428167304,
      "step": 26145,
      "train_runtime": 50387.743,
      "train_tokens_per_second": 8497.45
    },
    {
      "epoch": 1.7607056288715324,
      "grad_norm": 1.2747353479506052,
      "learning_rate": 3.4927443302033127e-07,
      "loss": 0.26247348785400393,
      "num_input_tokens_seen": 428249224,
      "step": 26150,
      "train_runtime": 50397.4006,
      "train_tokens_per_second": 8497.447
    },
    {
      "epoch": 1.7610422838674926,
      "grad_norm": 1.0986114179912403,
      "learning_rate": 3.4830419715969233e-07,
      "loss": 0.2702704906463623,
      "num_input_tokens_seen": 428331144,
      "step": 26155,
      "train_runtime": 50407.0313,
      "train_tokens_per_second": 8497.448
    },
    {
      "epoch": 1.7613789388634529,
      "grad_norm": 1.505175584795199,
      "learning_rate": 3.4733526213095114e-07,
      "loss": 0.2997642993927002,
      "num_input_tokens_seen": 428413064,
      "step": 26160,
      "train_runtime": 50416.5909,
      "train_tokens_per_second": 8497.462
    },
    {
      "epoch": 1.761715593859413,
      "grad_norm": 1.2254947230835869,
      "learning_rate": 3.4636762820506876e-07,
      "loss": 0.27191622257232667,
      "num_input_tokens_seen": 428494984,
      "step": 26165,
      "train_runtime": 50426.1727,
      "train_tokens_per_second": 8497.472
    },
    {
      "epoch": 1.762052248855373,
      "grad_norm": 1.2732886419195062,
      "learning_rate": 3.4540129565263867e-07,
      "loss": 0.26820826530456543,
      "num_input_tokens_seen": 428576904,
      "step": 26170,
      "train_runtime": 50435.7991,
      "train_tokens_per_second": 8497.474
    },
    {
      "epoch": 1.7623889038513332,
      "grad_norm": 1.0630159876956744,
      "learning_rate": 3.444362647438942e-07,
      "loss": 0.28188209533691405,
      "num_input_tokens_seen": 428658824,
      "step": 26175,
      "train_runtime": 50445.526,
      "train_tokens_per_second": 8497.46
    },
    {
      "epoch": 1.7627255588472934,
      "grad_norm": 1.0226439611723943,
      "learning_rate": 3.434725357487029e-07,
      "loss": 0.3018972873687744,
      "num_input_tokens_seen": 428740744,
      "step": 26180,
      "train_runtime": 50455.0739,
      "train_tokens_per_second": 8497.475
    },
    {
      "epoch": 1.7630622138432535,
      "grad_norm": 1.1879074704299086,
      "learning_rate": 3.4251010893656844e-07,
      "loss": 0.32597715854644777,
      "num_input_tokens_seen": 428822664,
      "step": 26185,
      "train_runtime": 50464.7886,
      "train_tokens_per_second": 8497.463
    },
    {
      "epoch": 1.7633988688392135,
      "grad_norm": 1.1885863238716508,
      "learning_rate": 3.4154898457663065e-07,
      "loss": 0.25764706134796145,
      "num_input_tokens_seen": 428904584,
      "step": 26190,
      "train_runtime": 50474.4704,
      "train_tokens_per_second": 8497.456
    },
    {
      "epoch": 1.7637355238351737,
      "grad_norm": 1.0340675175829788,
      "learning_rate": 3.40589162937664e-07,
      "loss": 0.27387542724609376,
      "num_input_tokens_seen": 428986504,
      "step": 26195,
      "train_runtime": 50483.9697,
      "train_tokens_per_second": 8497.48
    },
    {
      "epoch": 1.764072178831134,
      "grad_norm": 1.0236296128002782,
      "learning_rate": 3.3963064428808034e-07,
      "loss": 0.2629204511642456,
      "num_input_tokens_seen": 429068424,
      "step": 26200,
      "train_runtime": 50493.599,
      "train_tokens_per_second": 8497.482
    },
    {
      "epoch": 1.764408833827094,
      "grad_norm": 1.019185071168744,
      "learning_rate": 3.386734288959265e-07,
      "loss": 0.2717123985290527,
      "num_input_tokens_seen": 429150344,
      "step": 26205,
      "train_runtime": 50503.7832,
      "train_tokens_per_second": 8497.39
    },
    {
      "epoch": 1.764745488823054,
      "grad_norm": 1.1036130694074895,
      "learning_rate": 3.377175170288838e-07,
      "loss": 0.265581750869751,
      "num_input_tokens_seen": 429232264,
      "step": 26210,
      "train_runtime": 50513.4556,
      "train_tokens_per_second": 8497.385
    },
    {
      "epoch": 1.7650821438190143,
      "grad_norm": 1.1844831855537283,
      "learning_rate": 3.367629089542712e-07,
      "loss": 0.27230110168457033,
      "num_input_tokens_seen": 429314184,
      "step": 26215,
      "train_runtime": 50523.1683,
      "train_tokens_per_second": 8497.373
    },
    {
      "epoch": 1.7654187988149745,
      "grad_norm": 1.1768252965750052,
      "learning_rate": 3.3580960493904024e-07,
      "loss": 0.30577037334442136,
      "num_input_tokens_seen": 429396104,
      "step": 26220,
      "train_runtime": 50532.8716,
      "train_tokens_per_second": 8497.362
    },
    {
      "epoch": 1.7657554538109346,
      "grad_norm": 1.1124387197710037,
      "learning_rate": 3.3485760524978093e-07,
      "loss": 0.27349367141723635,
      "num_input_tokens_seen": 429478024,
      "step": 26225,
      "train_runtime": 50542.4945,
      "train_tokens_per_second": 8497.365
    },
    {
      "epoch": 1.7660921088068946,
      "grad_norm": 1.0383575483812826,
      "learning_rate": 3.3390691015271646e-07,
      "loss": 0.2823737144470215,
      "num_input_tokens_seen": 429559944,
      "step": 26230,
      "train_runtime": 50552.3126,
      "train_tokens_per_second": 8497.335
    },
    {
      "epoch": 1.7664287638028549,
      "grad_norm": 1.194756672542807,
      "learning_rate": 3.329575199137053e-07,
      "loss": 0.27919902801513674,
      "num_input_tokens_seen": 429641864,
      "step": 26235,
      "train_runtime": 50561.9658,
      "train_tokens_per_second": 8497.333
    },
    {
      "epoch": 1.766765418798815,
      "grad_norm": 1.18335749051627,
      "learning_rate": 3.3200943479824113e-07,
      "loss": 0.275797438621521,
      "num_input_tokens_seen": 429723576,
      "step": 26240,
      "train_runtime": 50571.9585,
      "train_tokens_per_second": 8497.27
    },
    {
      "epoch": 1.7671020737947751,
      "grad_norm": 1.0108183216064217,
      "learning_rate": 3.3106265507145406e-07,
      "loss": 0.22912149429321288,
      "num_input_tokens_seen": 429805496,
      "step": 26245,
      "train_runtime": 50581.4601,
      "train_tokens_per_second": 8497.293
    },
    {
      "epoch": 1.7674387287907352,
      "grad_norm": 1.2774099828063206,
      "learning_rate": 3.3011718099810687e-07,
      "loss": 0.2580888271331787,
      "num_input_tokens_seen": 429887416,
      "step": 26250,
      "train_runtime": 50590.981,
      "train_tokens_per_second": 8497.313
    },
    {
      "epoch": 1.7677753837866954,
      "grad_norm": 1.11394390380549,
      "learning_rate": 3.2917301284259904e-07,
      "loss": 0.2961434364318848,
      "num_input_tokens_seen": 429969336,
      "step": 26255,
      "train_runtime": 50600.4856,
      "train_tokens_per_second": 8497.336
    },
    {
      "epoch": 1.7681120387826557,
      "grad_norm": 1.2859001950658127,
      "learning_rate": 3.2823015086896335e-07,
      "loss": 0.3091093063354492,
      "num_input_tokens_seen": 430051256,
      "step": 26260,
      "train_runtime": 50609.9901,
      "train_tokens_per_second": 8497.359
    },
    {
      "epoch": 1.7684486937786157,
      "grad_norm": 1.0335076161290022,
      "learning_rate": 3.2728859534087e-07,
      "loss": 0.27883837223052976,
      "num_input_tokens_seen": 430133176,
      "step": 26265,
      "train_runtime": 50619.6243,
      "train_tokens_per_second": 8497.36
    },
    {
      "epoch": 1.7687853487745757,
      "grad_norm": 1.1124294108822255,
      "learning_rate": 3.263483465216194e-07,
      "loss": 0.28917133808135986,
      "num_input_tokens_seen": 430215096,
      "step": 26270,
      "train_runtime": 50629.2597,
      "train_tokens_per_second": 8497.361
    },
    {
      "epoch": 1.769122003770536,
      "grad_norm": 1.028908612718888,
      "learning_rate": 3.2540940467415126e-07,
      "loss": 0.2625285148620605,
      "num_input_tokens_seen": 430297016,
      "step": 26275,
      "train_runtime": 50638.94,
      "train_tokens_per_second": 8497.354
    },
    {
      "epoch": 1.7694586587664962,
      "grad_norm": 1.1333611194285766,
      "learning_rate": 3.2447177006103726e-07,
      "loss": 0.2845073461532593,
      "num_input_tokens_seen": 430378720,
      "step": 26280,
      "train_runtime": 50649.0899,
      "train_tokens_per_second": 8497.265
    },
    {
      "epoch": 1.7697953137624562,
      "grad_norm": 1.1508012124971325,
      "learning_rate": 3.235354429444831e-07,
      "loss": 0.2868909120559692,
      "num_input_tokens_seen": 430460640,
      "step": 26285,
      "train_runtime": 50658.6482,
      "train_tokens_per_second": 8497.278
    },
    {
      "epoch": 1.7701319687584163,
      "grad_norm": 1.1706338572036032,
      "learning_rate": 3.226004235863306e-07,
      "loss": 0.28384673595428467,
      "num_input_tokens_seen": 430542560,
      "step": 26290,
      "train_runtime": 50668.2077,
      "train_tokens_per_second": 8497.292
    },
    {
      "epoch": 1.7704686237543765,
      "grad_norm": 1.2460459448082255,
      "learning_rate": 3.2166671224805437e-07,
      "loss": 0.27516367435455324,
      "num_input_tokens_seen": 430624480,
      "step": 26295,
      "train_runtime": 50677.8816,
      "train_tokens_per_second": 8497.287
    },
    {
      "epoch": 1.7708052787503368,
      "grad_norm": 1.1350298283231177,
      "learning_rate": 3.207343091907633e-07,
      "loss": 0.27375354766845705,
      "num_input_tokens_seen": 430706400,
      "step": 26300,
      "train_runtime": 50687.5672,
      "train_tokens_per_second": 8497.279
    },
    {
      "epoch": 1.7711419337462968,
      "grad_norm": 0.9030196892469442,
      "learning_rate": 3.1980321467520226e-07,
      "loss": 0.2697168827056885,
      "num_input_tokens_seen": 430788320,
      "step": 26305,
      "train_runtime": 50697.2529,
      "train_tokens_per_second": 8497.271
    },
    {
      "epoch": 1.7714785887422568,
      "grad_norm": 1.3328326851995116,
      "learning_rate": 3.1887342896174735e-07,
      "loss": 0.3012655019760132,
      "num_input_tokens_seen": 430870240,
      "step": 26310,
      "train_runtime": 50706.7491,
      "train_tokens_per_second": 8497.296
    },
    {
      "epoch": 1.771815243738217,
      "grad_norm": 1.085749809496537,
      "learning_rate": 3.1794495231041276e-07,
      "loss": 0.27461862564086914,
      "num_input_tokens_seen": 430952160,
      "step": 26315,
      "train_runtime": 50716.4437,
      "train_tokens_per_second": 8497.287
    },
    {
      "epoch": 1.7721518987341773,
      "grad_norm": 1.0012025060726033,
      "learning_rate": 3.1701778498084035e-07,
      "loss": 0.27560255527496336,
      "num_input_tokens_seen": 431034080,
      "step": 26320,
      "train_runtime": 50726.1653,
      "train_tokens_per_second": 8497.273
    },
    {
      "epoch": 1.7724885537301374,
      "grad_norm": 1.0105285954365781,
      "learning_rate": 3.160919272323121e-07,
      "loss": 0.29046263694763186,
      "num_input_tokens_seen": 431116000,
      "step": 26325,
      "train_runtime": 50735.8404,
      "train_tokens_per_second": 8497.267
    },
    {
      "epoch": 1.7728252087260974,
      "grad_norm": 1.0037404569522286,
      "learning_rate": 3.1516737932374087e-07,
      "loss": 0.29803028106689455,
      "num_input_tokens_seen": 431197920,
      "step": 26330,
      "train_runtime": 50745.5785,
      "train_tokens_per_second": 8497.251
    },
    {
      "epoch": 1.7731618637220576,
      "grad_norm": 1.1307572967172292,
      "learning_rate": 3.142441415136727e-07,
      "loss": 0.27331621646881105,
      "num_input_tokens_seen": 431279840,
      "step": 26335,
      "train_runtime": 50755.0943,
      "train_tokens_per_second": 8497.272
    },
    {
      "epoch": 1.773498518718018,
      "grad_norm": 1.1193951902823023,
      "learning_rate": 3.133222140602893e-07,
      "loss": 0.2815645456314087,
      "num_input_tokens_seen": 431361760,
      "step": 26340,
      "train_runtime": 50764.7308,
      "train_tokens_per_second": 8497.273
    },
    {
      "epoch": 1.773835173713978,
      "grad_norm": 1.077647542308836,
      "learning_rate": 3.124015972214034e-07,
      "loss": 0.30127880573272703,
      "num_input_tokens_seen": 431443680,
      "step": 26345,
      "train_runtime": 50774.2671,
      "train_tokens_per_second": 8497.29
    },
    {
      "epoch": 1.774171828709938,
      "grad_norm": 1.0851675469876194,
      "learning_rate": 3.114822912544635e-07,
      "loss": 0.2854210376739502,
      "num_input_tokens_seen": 431525600,
      "step": 26350,
      "train_runtime": 50783.8748,
      "train_tokens_per_second": 8497.296
    },
    {
      "epoch": 1.7745084837058982,
      "grad_norm": 1.3152171094496647,
      "learning_rate": 3.1056429641655007e-07,
      "loss": 0.27200775146484374,
      "num_input_tokens_seen": 431607368,
      "step": 26355,
      "train_runtime": 50793.4125,
      "train_tokens_per_second": 8497.31
    },
    {
      "epoch": 1.7748451387018584,
      "grad_norm": 1.1416707822697112,
      "learning_rate": 3.0964761296437665e-07,
      "loss": 0.290425443649292,
      "num_input_tokens_seen": 431689288,
      "step": 26360,
      "train_runtime": 50803.0657,
      "train_tokens_per_second": 8497.308
    },
    {
      "epoch": 1.7751817936978185,
      "grad_norm": 1.2433946577311876,
      "learning_rate": 3.087322411542937e-07,
      "loss": 0.29203214645385744,
      "num_input_tokens_seen": 431771208,
      "step": 26365,
      "train_runtime": 50812.7556,
      "train_tokens_per_second": 8497.3
    },
    {
      "epoch": 1.7755184486937785,
      "grad_norm": 1.020316024340791,
      "learning_rate": 3.0781818124227867e-07,
      "loss": 0.2584686756134033,
      "num_input_tokens_seen": 431853128,
      "step": 26370,
      "train_runtime": 50822.4437,
      "train_tokens_per_second": 8497.292
    },
    {
      "epoch": 1.7758551036897388,
      "grad_norm": 1.1639312525564003,
      "learning_rate": 3.0690543348394776e-07,
      "loss": 0.27327733039855956,
      "num_input_tokens_seen": 431935048,
      "step": 26375,
      "train_runtime": 50832.1343,
      "train_tokens_per_second": 8497.283
    },
    {
      "epoch": 1.776191758685699,
      "grad_norm": 1.1325923638918414,
      "learning_rate": 3.059939981345467e-07,
      "loss": 0.25672330856323244,
      "num_input_tokens_seen": 432016968,
      "step": 26380,
      "train_runtime": 50841.8215,
      "train_tokens_per_second": 8497.276
    },
    {
      "epoch": 1.776528413681659,
      "grad_norm": 1.4001444655374609,
      "learning_rate": 3.050838754489566e-07,
      "loss": 0.28624792098999025,
      "num_input_tokens_seen": 432098888,
      "step": 26385,
      "train_runtime": 50851.5001,
      "train_tokens_per_second": 8497.269
    },
    {
      "epoch": 1.776865068677619,
      "grad_norm": 1.1304315011673653,
      "learning_rate": 3.041750656816894e-07,
      "loss": 0.2950169324874878,
      "num_input_tokens_seen": 432180808,
      "step": 26390,
      "train_runtime": 50861.012,
      "train_tokens_per_second": 8497.291
    },
    {
      "epoch": 1.7772017236735793,
      "grad_norm": 1.049226225305741,
      "learning_rate": 3.0326756908689135e-07,
      "loss": 0.2760596513748169,
      "num_input_tokens_seen": 432262728,
      "step": 26395,
      "train_runtime": 50870.6587,
      "train_tokens_per_second": 8497.29
    },
    {
      "epoch": 1.7775383786695396,
      "grad_norm": 1.1891693695558079,
      "learning_rate": 3.0236138591833995e-07,
      "loss": 0.2689986228942871,
      "num_input_tokens_seen": 432344648,
      "step": 26400,
      "train_runtime": 50889.0843,
      "train_tokens_per_second": 8495.823
    },
    {
      "epoch": 1.7778750336654996,
      "grad_norm": 1.1791127589252715,
      "learning_rate": 3.0145651642944863e-07,
      "loss": 0.2947198152542114,
      "num_input_tokens_seen": 432426568,
      "step": 26405,
      "train_runtime": 50907.421,
      "train_tokens_per_second": 8494.372
    },
    {
      "epoch": 1.7782116886614596,
      "grad_norm": 1.2328268520074952,
      "learning_rate": 3.005529608732588e-07,
      "loss": 0.24716453552246093,
      "num_input_tokens_seen": 432508488,
      "step": 26410,
      "train_runtime": 50916.9126,
      "train_tokens_per_second": 8494.397
    },
    {
      "epoch": 1.7785483436574199,
      "grad_norm": 1.1839700993984397,
      "learning_rate": 2.996507195024495e-07,
      "loss": 0.2994117498397827,
      "num_input_tokens_seen": 432590408,
      "step": 26415,
      "train_runtime": 50926.4005,
      "train_tokens_per_second": 8494.423
    },
    {
      "epoch": 1.7788849986533801,
      "grad_norm": 1.1219515924566985,
      "learning_rate": 2.9874979256932614e-07,
      "loss": 0.3019345760345459,
      "num_input_tokens_seen": 432672328,
      "step": 26420,
      "train_runtime": 50936.0495,
      "train_tokens_per_second": 8494.423
    },
    {
      "epoch": 1.7792216536493402,
      "grad_norm": 1.0858934744736042,
      "learning_rate": 2.9785018032583325e-07,
      "loss": 0.27185649871826173,
      "num_input_tokens_seen": 432754248,
      "step": 26425,
      "train_runtime": 50945.6745,
      "train_tokens_per_second": 8494.426
    },
    {
      "epoch": 1.7795583086453002,
      "grad_norm": 1.45661300850982,
      "learning_rate": 2.969518830235435e-07,
      "loss": 0.30432958602905275,
      "num_input_tokens_seen": 432835792,
      "step": 26430,
      "train_runtime": 50955.677,
      "train_tokens_per_second": 8494.359
    },
    {
      "epoch": 1.7798949636412604,
      "grad_norm": 1.1635134893952788,
      "learning_rate": 2.960549009136626e-07,
      "loss": 0.28301711082458497,
      "num_input_tokens_seen": 432917712,
      "step": 26435,
      "train_runtime": 50965.37,
      "train_tokens_per_second": 8494.35
    },
    {
      "epoch": 1.7802316186372207,
      "grad_norm": 1.2773582980976346,
      "learning_rate": 2.951592342470289e-07,
      "loss": 0.34586129188537595,
      "num_input_tokens_seen": 432999632,
      "step": 26440,
      "train_runtime": 50975.3966,
      "train_tokens_per_second": 8494.287
    },
    {
      "epoch": 1.7805682736331807,
      "grad_norm": 1.1258220110000972,
      "learning_rate": 2.942648832741124e-07,
      "loss": 0.2545922756195068,
      "num_input_tokens_seen": 433081552,
      "step": 26445,
      "train_runtime": 50985.064,
      "train_tokens_per_second": 8494.283
    },
    {
      "epoch": 1.7809049286291407,
      "grad_norm": 1.266465365207741,
      "learning_rate": 2.933718482450154e-07,
      "loss": 0.28557848930358887,
      "num_input_tokens_seen": 433163472,
      "step": 26450,
      "train_runtime": 50994.6868,
      "train_tokens_per_second": 8494.286
    },
    {
      "epoch": 1.781241583625101,
      "grad_norm": 1.4883856444713415,
      "learning_rate": 2.9248012940947423e-07,
      "loss": 0.33151724338531496,
      "num_input_tokens_seen": 433245392,
      "step": 26455,
      "train_runtime": 51004.1969,
      "train_tokens_per_second": 8494.309
    },
    {
      "epoch": 1.7815782386210612,
      "grad_norm": 1.0631577631893447,
      "learning_rate": 2.915897270168522e-07,
      "loss": 0.2597470760345459,
      "num_input_tokens_seen": 433327096,
      "step": 26460,
      "train_runtime": 51014.2323,
      "train_tokens_per_second": 8494.239
    },
    {
      "epoch": 1.7819148936170213,
      "grad_norm": 1.1519035385097742,
      "learning_rate": 2.9070064131615004e-07,
      "loss": 0.2753658056259155,
      "num_input_tokens_seen": 433409016,
      "step": 26465,
      "train_runtime": 51023.7355,
      "train_tokens_per_second": 8494.263
    },
    {
      "epoch": 1.7822515486129813,
      "grad_norm": 1.1335564109751224,
      "learning_rate": 2.8981287255599567e-07,
      "loss": 0.27253284454345705,
      "num_input_tokens_seen": 433490936,
      "step": 26470,
      "train_runtime": 51033.3333,
      "train_tokens_per_second": 8494.27
    },
    {
      "epoch": 1.7825882036089415,
      "grad_norm": 1.3621626386914778,
      "learning_rate": 2.8892642098465205e-07,
      "loss": 0.3192347764968872,
      "num_input_tokens_seen": 433572856,
      "step": 26475,
      "train_runtime": 51042.8534,
      "train_tokens_per_second": 8494.291
    },
    {
      "epoch": 1.7829248586049018,
      "grad_norm": 1.1449869664100547,
      "learning_rate": 2.8804128685001263e-07,
      "loss": 0.2787992000579834,
      "num_input_tokens_seen": 433654776,
      "step": 26480,
      "train_runtime": 51052.4929,
      "train_tokens_per_second": 8494.292
    },
    {
      "epoch": 1.7832615136008618,
      "grad_norm": 1.0050791535648749,
      "learning_rate": 2.8715747039960216e-07,
      "loss": 0.2574162006378174,
      "num_input_tokens_seen": 433736696,
      "step": 26485,
      "train_runtime": 51062.044,
      "train_tokens_per_second": 8494.307
    },
    {
      "epoch": 1.7835981685968219,
      "grad_norm": 1.309625231015288,
      "learning_rate": 2.8627497188057684e-07,
      "loss": 0.302294397354126,
      "num_input_tokens_seen": 433818616,
      "step": 26490,
      "train_runtime": 51071.7613,
      "train_tokens_per_second": 8494.295
    },
    {
      "epoch": 1.783934823592782,
      "grad_norm": 1.2180023599265415,
      "learning_rate": 2.853937915397248e-07,
      "loss": 0.3138465881347656,
      "num_input_tokens_seen": 433900536,
      "step": 26495,
      "train_runtime": 51081.4433,
      "train_tokens_per_second": 8494.289
    },
    {
      "epoch": 1.7842714785887424,
      "grad_norm": 1.0247703349588393,
      "learning_rate": 2.845139296234639e-07,
      "loss": 0.28026957511901857,
      "num_input_tokens_seen": 433982456,
      "step": 26500,
      "train_runtime": 51090.9412,
      "train_tokens_per_second": 8494.313
    },
    {
      "epoch": 1.7846081335847024,
      "grad_norm": 0.9502707155341675,
      "learning_rate": 2.836353863778479e-07,
      "loss": 0.24080278873443603,
      "num_input_tokens_seen": 434064376,
      "step": 26505,
      "train_runtime": 51100.5922,
      "train_tokens_per_second": 8494.312
    },
    {
      "epoch": 1.7849447885806624,
      "grad_norm": 1.0784561333319507,
      "learning_rate": 2.8275816204855534e-07,
      "loss": 0.2796250104904175,
      "num_input_tokens_seen": 434146296,
      "step": 26510,
      "train_runtime": 51110.221,
      "train_tokens_per_second": 8494.315
    },
    {
      "epoch": 1.7852814435766227,
      "grad_norm": 1.147877631800991,
      "learning_rate": 2.818822568809015e-07,
      "loss": 0.26992087364196776,
      "num_input_tokens_seen": 434228216,
      "step": 26515,
      "train_runtime": 51120.0856,
      "train_tokens_per_second": 8494.278
    },
    {
      "epoch": 1.785618098572583,
      "grad_norm": 1.2949868317813065,
      "learning_rate": 2.810076711198284e-07,
      "loss": 0.26367864608764646,
      "num_input_tokens_seen": 434309248,
      "step": 26520,
      "train_runtime": 51131.1547,
      "train_tokens_per_second": 8494.024
    },
    {
      "epoch": 1.785954753568543,
      "grad_norm": 1.0371774389486865,
      "learning_rate": 2.801344050099131e-07,
      "loss": 0.263205361366272,
      "num_input_tokens_seen": 434391168,
      "step": 26525,
      "train_runtime": 51140.7256,
      "train_tokens_per_second": 8494.036
    },
    {
      "epoch": 1.786291408564503,
      "grad_norm": 1.0795490886633534,
      "learning_rate": 2.7926245879536076e-07,
      "loss": 0.28468856811523435,
      "num_input_tokens_seen": 434473088,
      "step": 26530,
      "train_runtime": 51150.2335,
      "train_tokens_per_second": 8494.059
    },
    {
      "epoch": 1.7866280635604632,
      "grad_norm": 1.03071070404919,
      "learning_rate": 2.7839183272000857e-07,
      "loss": 0.2676255226135254,
      "num_input_tokens_seen": 434555008,
      "step": 26535,
      "train_runtime": 51159.8949,
      "train_tokens_per_second": 8494.056
    },
    {
      "epoch": 1.7869647185564235,
      "grad_norm": 1.2545322258511613,
      "learning_rate": 2.775225270273241e-07,
      "loss": 0.2841570615768433,
      "num_input_tokens_seen": 434636928,
      "step": 26540,
      "train_runtime": 51169.4955,
      "train_tokens_per_second": 8494.063
    },
    {
      "epoch": 1.7873013735523835,
      "grad_norm": 1.1657112715178064,
      "learning_rate": 2.7665454196040665e-07,
      "loss": 0.2664470195770264,
      "num_input_tokens_seen": 434718848,
      "step": 26545,
      "train_runtime": 51179.0808,
      "train_tokens_per_second": 8494.073
    },
    {
      "epoch": 1.7876380285483435,
      "grad_norm": 1.132532582389921,
      "learning_rate": 2.7578787776198325e-07,
      "loss": 0.2879770755767822,
      "num_input_tokens_seen": 434800768,
      "step": 26550,
      "train_runtime": 51188.5887,
      "train_tokens_per_second": 8494.096
    },
    {
      "epoch": 1.7879746835443038,
      "grad_norm": 1.2888068239006545,
      "learning_rate": 2.7492253467441775e-07,
      "loss": 0.29546594619750977,
      "num_input_tokens_seen": 434882688,
      "step": 26555,
      "train_runtime": 51198.2893,
      "train_tokens_per_second": 8494.086
    },
    {
      "epoch": 1.788311338540264,
      "grad_norm": 1.1125598450307048,
      "learning_rate": 2.740585129396961e-07,
      "loss": 0.2660548210144043,
      "num_input_tokens_seen": 434964608,
      "step": 26560,
      "train_runtime": 51207.7885,
      "train_tokens_per_second": 8494.11
    },
    {
      "epoch": 1.788647993536224,
      "grad_norm": 1.1250199870753674,
      "learning_rate": 2.731958127994433e-07,
      "loss": 0.2577629804611206,
      "num_input_tokens_seen": 435046528,
      "step": 26565,
      "train_runtime": 51217.2965,
      "train_tokens_per_second": 8494.133
    },
    {
      "epoch": 1.788984648532184,
      "grad_norm": 1.1126439504208372,
      "learning_rate": 2.7233443449490695e-07,
      "loss": 0.2746382236480713,
      "num_input_tokens_seen": 435128448,
      "step": 26570,
      "train_runtime": 51226.9215,
      "train_tokens_per_second": 8494.136
    },
    {
      "epoch": 1.7893213035281443,
      "grad_norm": 1.1092000853130077,
      "learning_rate": 2.7147437826697153e-07,
      "loss": 0.2758479118347168,
      "num_input_tokens_seen": 435210368,
      "step": 26575,
      "train_runtime": 51236.426,
      "train_tokens_per_second": 8494.159
    },
    {
      "epoch": 1.7896579585241046,
      "grad_norm": 1.3334069617653586,
      "learning_rate": 2.706156443561481e-07,
      "loss": 0.268896222114563,
      "num_input_tokens_seen": 435292288,
      "step": 26580,
      "train_runtime": 51245.92,
      "train_tokens_per_second": 8494.184
    },
    {
      "epoch": 1.7899946135200646,
      "grad_norm": 1.3073006363240485,
      "learning_rate": 2.6975823300257884e-07,
      "loss": 0.2692330121994019,
      "num_input_tokens_seen": 435374208,
      "step": 26585,
      "train_runtime": 51255.4216,
      "train_tokens_per_second": 8494.208
    },
    {
      "epoch": 1.7903312685160246,
      "grad_norm": 1.20920447454928,
      "learning_rate": 2.6890214444603604e-07,
      "loss": 0.3003836631774902,
      "num_input_tokens_seen": 435455936,
      "step": 26590,
      "train_runtime": 51265.4082,
      "train_tokens_per_second": 8494.147
    },
    {
      "epoch": 1.790667923511985,
      "grad_norm": 1.1010587249942247,
      "learning_rate": 2.6804737892592193e-07,
      "loss": 0.29770057201385497,
      "num_input_tokens_seen": 435537856,
      "step": 26595,
      "train_runtime": 51274.9702,
      "train_tokens_per_second": 8494.161
    },
    {
      "epoch": 1.7910045785079451,
      "grad_norm": 1.1815278625117964,
      "learning_rate": 2.671939366812687e-07,
      "loss": 0.29491419792175294,
      "num_input_tokens_seen": 435619776,
      "step": 26600,
      "train_runtime": 51284.6153,
      "train_tokens_per_second": 8494.161
    },
    {
      "epoch": 1.7913412335039052,
      "grad_norm": 1.111611402082031,
      "learning_rate": 2.66341817950741e-07,
      "loss": 0.2572668790817261,
      "num_input_tokens_seen": 435701696,
      "step": 26605,
      "train_runtime": 51294.4029,
      "train_tokens_per_second": 8494.137
    },
    {
      "epoch": 1.7916778884998652,
      "grad_norm": 1.2337043660017446,
      "learning_rate": 2.6549102297262764e-07,
      "loss": 0.28863351345062255,
      "num_input_tokens_seen": 435783616,
      "step": 26610,
      "train_runtime": 51304.0099,
      "train_tokens_per_second": 8494.143
    },
    {
      "epoch": 1.7920145434958255,
      "grad_norm": 1.137977437717128,
      "learning_rate": 2.6464155198485387e-07,
      "loss": 0.27512693405151367,
      "num_input_tokens_seen": 435865536,
      "step": 26615,
      "train_runtime": 51313.6546,
      "train_tokens_per_second": 8494.143
    },
    {
      "epoch": 1.7923511984917857,
      "grad_norm": 1.251015915185266,
      "learning_rate": 2.6379340522496853e-07,
      "loss": 0.2557593584060669,
      "num_input_tokens_seen": 435947456,
      "step": 26620,
      "train_runtime": 51323.2004,
      "train_tokens_per_second": 8494.16
    },
    {
      "epoch": 1.7926878534877457,
      "grad_norm": 1.1022942863055467,
      "learning_rate": 2.6294658293015584e-07,
      "loss": 0.3043011903762817,
      "num_input_tokens_seen": 436029376,
      "step": 26625,
      "train_runtime": 51332.8056,
      "train_tokens_per_second": 8494.166
    },
    {
      "epoch": 1.7930245084837058,
      "grad_norm": 1.45762383690386,
      "learning_rate": 2.621010853372258e-07,
      "loss": 0.3160256862640381,
      "num_input_tokens_seen": 436111296,
      "step": 26630,
      "train_runtime": 51342.4879,
      "train_tokens_per_second": 8494.16
    },
    {
      "epoch": 1.793361163479666,
      "grad_norm": 1.246096068273977,
      "learning_rate": 2.612569126826187e-07,
      "loss": 0.29236688613891604,
      "num_input_tokens_seen": 436193216,
      "step": 26635,
      "train_runtime": 51352.1583,
      "train_tokens_per_second": 8494.155
    },
    {
      "epoch": 1.7936978184756263,
      "grad_norm": 1.183341038765162,
      "learning_rate": 2.6041406520240563e-07,
      "loss": 0.3190739393234253,
      "num_input_tokens_seen": 436275136,
      "step": 26640,
      "train_runtime": 51361.7125,
      "train_tokens_per_second": 8494.17
    },
    {
      "epoch": 1.7940344734715863,
      "grad_norm": 1.0266158698143617,
      "learning_rate": 2.5957254313228595e-07,
      "loss": 0.2784675359725952,
      "num_input_tokens_seen": 436357056,
      "step": 26645,
      "train_runtime": 51371.2988,
      "train_tokens_per_second": 8494.18
    },
    {
      "epoch": 1.7943711284675463,
      "grad_norm": 1.0479770617136612,
      "learning_rate": 2.5873234670758753e-07,
      "loss": 0.25803098678588865,
      "num_input_tokens_seen": 436438976,
      "step": 26650,
      "train_runtime": 51380.9308,
      "train_tokens_per_second": 8494.182
    },
    {
      "epoch": 1.7947077834635066,
      "grad_norm": 1.1565283165831723,
      "learning_rate": 2.578934761632712e-07,
      "loss": 0.26105446815490724,
      "num_input_tokens_seen": 436520896,
      "step": 26655,
      "train_runtime": 51390.4323,
      "train_tokens_per_second": 8494.206
    },
    {
      "epoch": 1.7950444384594668,
      "grad_norm": 1.129098517215261,
      "learning_rate": 2.570559317339211e-07,
      "loss": 0.29613828659057617,
      "num_input_tokens_seen": 436602816,
      "step": 26660,
      "train_runtime": 51400.0575,
      "train_tokens_per_second": 8494.209
    },
    {
      "epoch": 1.7953810934554268,
      "grad_norm": 1.2143639262530581,
      "learning_rate": 2.56219713653757e-07,
      "loss": 0.2656237602233887,
      "num_input_tokens_seen": 436684736,
      "step": 26665,
      "train_runtime": 51409.6063,
      "train_tokens_per_second": 8494.224
    },
    {
      "epoch": 1.7957177484513869,
      "grad_norm": 1.1906243458340517,
      "learning_rate": 2.5538482215662187e-07,
      "loss": 0.29483819007873535,
      "num_input_tokens_seen": 436766656,
      "step": 26670,
      "train_runtime": 51419.2045,
      "train_tokens_per_second": 8494.232
    },
    {
      "epoch": 1.7960544034473471,
      "grad_norm": 1.2934577653949344,
      "learning_rate": 2.545512574759923e-07,
      "loss": 0.2979417324066162,
      "num_input_tokens_seen": 436848576,
      "step": 26675,
      "train_runtime": 51428.7172,
      "train_tokens_per_second": 8494.254
    },
    {
      "epoch": 1.7963910584433074,
      "grad_norm": 1.839773633076846,
      "learning_rate": 2.537190198449713e-07,
      "loss": 0.31150362491607664,
      "num_input_tokens_seen": 436930496,
      "step": 26680,
      "train_runtime": 51438.342,
      "train_tokens_per_second": 8494.257
    },
    {
      "epoch": 1.7967277134392674,
      "grad_norm": 1.1802505409148647,
      "learning_rate": 2.5288810949629207e-07,
      "loss": 0.2736582517623901,
      "num_input_tokens_seen": 437012416,
      "step": 26685,
      "train_runtime": 51447.9482,
      "train_tokens_per_second": 8494.263
    },
    {
      "epoch": 1.7970643684352274,
      "grad_norm": 1.14915103266433,
      "learning_rate": 2.5205852666231544e-07,
      "loss": 0.27180042266845705,
      "num_input_tokens_seen": 437094336,
      "step": 26690,
      "train_runtime": 51457.4436,
      "train_tokens_per_second": 8494.288
    },
    {
      "epoch": 1.7974010234311877,
      "grad_norm": 1.0479138146735587,
      "learning_rate": 2.5123027157503144e-07,
      "loss": 0.2544761896133423,
      "num_input_tokens_seen": 437176256,
      "step": 26695,
      "train_runtime": 51467.0493,
      "train_tokens_per_second": 8494.294
    },
    {
      "epoch": 1.797737678427148,
      "grad_norm": 1.0409931092255496,
      "learning_rate": 2.504033444660592e-07,
      "loss": 0.27889993190765383,
      "num_input_tokens_seen": 437258176,
      "step": 26700,
      "train_runtime": 51476.6849,
      "train_tokens_per_second": 8494.296
    },
    {
      "epoch": 1.798074333423108,
      "grad_norm": 0.9104801794151265,
      "learning_rate": 2.4957774556664706e-07,
      "loss": 0.26153564453125,
      "num_input_tokens_seen": 437340096,
      "step": 26705,
      "train_runtime": 51486.3003,
      "train_tokens_per_second": 8494.3
    },
    {
      "epoch": 1.798410988419068,
      "grad_norm": 1.4121530142476384,
      "learning_rate": 2.487534751076692e-07,
      "loss": 0.28719067573547363,
      "num_input_tokens_seen": 437422016,
      "step": 26710,
      "train_runtime": 51495.7962,
      "train_tokens_per_second": 8494.325
    },
    {
      "epoch": 1.7987476434150282,
      "grad_norm": 0.9779401864042444,
      "learning_rate": 2.47930533319633e-07,
      "loss": 0.2542572975158691,
      "num_input_tokens_seen": 437503936,
      "step": 26715,
      "train_runtime": 51505.4076,
      "train_tokens_per_second": 8494.33
    },
    {
      "epoch": 1.7990842984109885,
      "grad_norm": 1.1158590823652859,
      "learning_rate": 2.4710892043266823e-07,
      "loss": 0.31534466743469236,
      "num_input_tokens_seen": 437585856,
      "step": 26720,
      "train_runtime": 51514.9981,
      "train_tokens_per_second": 8494.339
    },
    {
      "epoch": 1.7994209534069485,
      "grad_norm": 1.0413313041812142,
      "learning_rate": 2.462886366765388e-07,
      "loss": 0.2787597417831421,
      "num_input_tokens_seen": 437667776,
      "step": 26725,
      "train_runtime": 51524.5827,
      "train_tokens_per_second": 8494.349
    },
    {
      "epoch": 1.7997576084029085,
      "grad_norm": 1.3572843894806894,
      "learning_rate": 2.454696822806341e-07,
      "loss": 0.2891636848449707,
      "num_input_tokens_seen": 437749696,
      "step": 26730,
      "train_runtime": 51534.0949,
      "train_tokens_per_second": 8494.371
    },
    {
      "epoch": 1.8000942633988688,
      "grad_norm": 1.172373780358597,
      "learning_rate": 2.446520574739708e-07,
      "loss": 0.2570096492767334,
      "num_input_tokens_seen": 437831616,
      "step": 26735,
      "train_runtime": 51543.7101,
      "train_tokens_per_second": 8494.375
    },
    {
      "epoch": 1.800430918394829,
      "grad_norm": 1.3180178584932065,
      "learning_rate": 2.438357624851967e-07,
      "loss": 0.26976377964019777,
      "num_input_tokens_seen": 437913536,
      "step": 26740,
      "train_runtime": 51553.3218,
      "train_tokens_per_second": 8494.381
    },
    {
      "epoch": 1.800767573390789,
      "grad_norm": 0.9132948797377544,
      "learning_rate": 2.430207975425847e-07,
      "loss": 0.2700651168823242,
      "num_input_tokens_seen": 437995456,
      "step": 26745,
      "train_runtime": 51562.9563,
      "train_tokens_per_second": 8494.382
    },
    {
      "epoch": 1.801104228386749,
      "grad_norm": 1.1302595816269405,
      "learning_rate": 2.4220716287403744e-07,
      "loss": 0.2681243419647217,
      "num_input_tokens_seen": 438077376,
      "step": 26750,
      "train_runtime": 51572.4574,
      "train_tokens_per_second": 8494.406
    },
    {
      "epoch": 1.8014408833827094,
      "grad_norm": 1.1979704206809214,
      "learning_rate": 2.413948587070869e-07,
      "loss": 0.28246893882751467,
      "num_input_tokens_seen": 438159296,
      "step": 26755,
      "train_runtime": 51582.066,
      "train_tokens_per_second": 8494.412
    },
    {
      "epoch": 1.8017775383786696,
      "grad_norm": 1.4106861377513595,
      "learning_rate": 2.405838852688891e-07,
      "loss": 0.2931885480880737,
      "num_input_tokens_seen": 438241216,
      "step": 26760,
      "train_runtime": 51591.6595,
      "train_tokens_per_second": 8494.42
    },
    {
      "epoch": 1.8021141933746296,
      "grad_norm": 1.1573524463170204,
      "learning_rate": 2.397742427862315e-07,
      "loss": 0.24038796424865722,
      "num_input_tokens_seen": 438323136,
      "step": 26765,
      "train_runtime": 51601.1687,
      "train_tokens_per_second": 8494.442
    },
    {
      "epoch": 1.8024508483705897,
      "grad_norm": 1.1897423989787854,
      "learning_rate": 2.3896593148552695e-07,
      "loss": 0.2844799280166626,
      "num_input_tokens_seen": 438404816,
      "step": 26770,
      "train_runtime": 51611.2621,
      "train_tokens_per_second": 8494.363
    },
    {
      "epoch": 1.80278750336655,
      "grad_norm": 1.3769598583048284,
      "learning_rate": 2.3815895159281844e-07,
      "loss": 0.2833848476409912,
      "num_input_tokens_seen": 438486736,
      "step": 26775,
      "train_runtime": 51620.9074,
      "train_tokens_per_second": 8494.363
    },
    {
      "epoch": 1.8031241583625102,
      "grad_norm": 1.2869025518258033,
      "learning_rate": 2.3735330333377493e-07,
      "loss": 0.267194128036499,
      "num_input_tokens_seen": 438568656,
      "step": 26780,
      "train_runtime": 51630.642,
      "train_tokens_per_second": 8494.348
    },
    {
      "epoch": 1.8034608133584702,
      "grad_norm": 1.0209092243258466,
      "learning_rate": 2.3654898693369232e-07,
      "loss": 0.29064340591430665,
      "num_input_tokens_seen": 438650576,
      "step": 26785,
      "train_runtime": 51640.3225,
      "train_tokens_per_second": 8494.342
    },
    {
      "epoch": 1.8037974683544302,
      "grad_norm": 1.2142201160579975,
      "learning_rate": 2.3574600261749735e-07,
      "loss": 0.2938575744628906,
      "num_input_tokens_seen": 438732496,
      "step": 26790,
      "train_runtime": 51649.944,
      "train_tokens_per_second": 8494.346
    },
    {
      "epoch": 1.8041341233503905,
      "grad_norm": 1.1201877899804893,
      "learning_rate": 2.3494435060973875e-07,
      "loss": 0.2869978904724121,
      "num_input_tokens_seen": 438814416,
      "step": 26795,
      "train_runtime": 51659.6279,
      "train_tokens_per_second": 8494.339
    },
    {
      "epoch": 1.8044707783463507,
      "grad_norm": 1.1634293617236122,
      "learning_rate": 2.3414403113459937e-07,
      "loss": 0.3070352554321289,
      "num_input_tokens_seen": 438896336,
      "step": 26800,
      "train_runtime": 51669.3127,
      "train_tokens_per_second": 8494.333
    },
    {
      "epoch": 1.8048074333423108,
      "grad_norm": 1.2324386901893671,
      "learning_rate": 2.3334504441588413e-07,
      "loss": 0.2661382436752319,
      "num_input_tokens_seen": 438978256,
      "step": 26805,
      "train_runtime": 51678.9922,
      "train_tokens_per_second": 8494.327
    },
    {
      "epoch": 1.8051440883382708,
      "grad_norm": 1.3261905051657714,
      "learning_rate": 2.3254739067702704e-07,
      "loss": 0.27530899047851565,
      "num_input_tokens_seen": 439060176,
      "step": 26810,
      "train_runtime": 51688.4932,
      "train_tokens_per_second": 8494.35
    },
    {
      "epoch": 1.805480743334231,
      "grad_norm": 1.1242456111728958,
      "learning_rate": 2.3175107014109077e-07,
      "loss": 0.27820146083831787,
      "num_input_tokens_seen": 439142096,
      "step": 26815,
      "train_runtime": 51698.0627,
      "train_tokens_per_second": 8494.363
    },
    {
      "epoch": 1.8058173983301913,
      "grad_norm": 1.2881666127937377,
      "learning_rate": 2.3095608303076223e-07,
      "loss": 0.29860012531280516,
      "num_input_tokens_seen": 439224016,
      "step": 26820,
      "train_runtime": 51707.6862,
      "train_tokens_per_second": 8494.366
    },
    {
      "epoch": 1.8061540533261513,
      "grad_norm": 1.1405391771809161,
      "learning_rate": 2.3016242956835798e-07,
      "loss": 0.30767502784729006,
      "num_input_tokens_seen": 439305936,
      "step": 26825,
      "train_runtime": 51717.3258,
      "train_tokens_per_second": 8494.367
    },
    {
      "epoch": 1.8064907083221113,
      "grad_norm": 1.0301188482854313,
      "learning_rate": 2.2937010997581999e-07,
      "loss": 0.25973167419433596,
      "num_input_tokens_seen": 439387856,
      "step": 26830,
      "train_runtime": 51726.8341,
      "train_tokens_per_second": 8494.389
    },
    {
      "epoch": 1.8068273633180716,
      "grad_norm": 1.257699415264098,
      "learning_rate": 2.2857912447471818e-07,
      "loss": 0.26730971336364745,
      "num_input_tokens_seen": 439469776,
      "step": 26835,
      "train_runtime": 51736.4284,
      "train_tokens_per_second": 8494.397
    },
    {
      "epoch": 1.8071640183140318,
      "grad_norm": 1.2318350279770003,
      "learning_rate": 2.2778947328625012e-07,
      "loss": 0.2674603223800659,
      "num_input_tokens_seen": 439551696,
      "step": 26840,
      "train_runtime": 51746.0475,
      "train_tokens_per_second": 8494.401
    },
    {
      "epoch": 1.8075006733099919,
      "grad_norm": 1.3011830664377977,
      "learning_rate": 2.2700115663123745e-07,
      "loss": 0.30583896636962893,
      "num_input_tokens_seen": 439633568,
      "step": 26845,
      "train_runtime": 51755.5658,
      "train_tokens_per_second": 8494.421
    },
    {
      "epoch": 1.807837328305952,
      "grad_norm": 1.331094579938551,
      "learning_rate": 2.2621417473013164e-07,
      "loss": 0.26637964248657225,
      "num_input_tokens_seen": 439715488,
      "step": 26850,
      "train_runtime": 51765.0789,
      "train_tokens_per_second": 8494.443
    },
    {
      "epoch": 1.8081739833019121,
      "grad_norm": 1.3803625070169516,
      "learning_rate": 2.254285278030094e-07,
      "loss": 0.27629210948944094,
      "num_input_tokens_seen": 439797408,
      "step": 26855,
      "train_runtime": 51774.5814,
      "train_tokens_per_second": 8494.466
    },
    {
      "epoch": 1.8085106382978724,
      "grad_norm": 1.0285453676377598,
      "learning_rate": 2.24644216069575e-07,
      "loss": 0.26798257827758787,
      "num_input_tokens_seen": 439879328,
      "step": 26860,
      "train_runtime": 51784.0969,
      "train_tokens_per_second": 8494.487
    },
    {
      "epoch": 1.8088472932938324,
      "grad_norm": 1.0643411193511547,
      "learning_rate": 2.238612397491574e-07,
      "loss": 0.2657432556152344,
      "num_input_tokens_seen": 439961248,
      "step": 26865,
      "train_runtime": 51793.5974,
      "train_tokens_per_second": 8494.51
    },
    {
      "epoch": 1.8091839482897925,
      "grad_norm": 1.0024247982417844,
      "learning_rate": 2.230795990607143e-07,
      "loss": 0.2779771089553833,
      "num_input_tokens_seen": 440043168,
      "step": 26870,
      "train_runtime": 51803.215,
      "train_tokens_per_second": 8494.515
    },
    {
      "epoch": 1.809520603285753,
      "grad_norm": 1.0556285331110657,
      "learning_rate": 2.2229929422282915e-07,
      "loss": 0.26885759830474854,
      "num_input_tokens_seen": 440124704,
      "step": 26875,
      "train_runtime": 51813.1831,
      "train_tokens_per_second": 8494.454
    },
    {
      "epoch": 1.809857258281713,
      "grad_norm": 1.0402975451071013,
      "learning_rate": 2.215203254537107e-07,
      "loss": 0.2902154207229614,
      "num_input_tokens_seen": 440206624,
      "step": 26880,
      "train_runtime": 51822.8316,
      "train_tokens_per_second": 8494.453
    },
    {
      "epoch": 1.810193913277673,
      "grad_norm": 1.1567907433426845,
      "learning_rate": 2.2074269297119588e-07,
      "loss": 0.29722378253936765,
      "num_input_tokens_seen": 440288544,
      "step": 26885,
      "train_runtime": 51832.4413,
      "train_tokens_per_second": 8494.459
    },
    {
      "epoch": 1.8105305682736332,
      "grad_norm": 1.0022413212208316,
      "learning_rate": 2.199663969927479e-07,
      "loss": 0.24157052040100097,
      "num_input_tokens_seen": 440370464,
      "step": 26890,
      "train_runtime": 51842.0727,
      "train_tokens_per_second": 8494.461
    },
    {
      "epoch": 1.8108672232695935,
      "grad_norm": 1.2144266178931207,
      "learning_rate": 2.1919143773545316e-07,
      "loss": 0.28687777519226076,
      "num_input_tokens_seen": 440452384,
      "step": 26895,
      "train_runtime": 51851.64,
      "train_tokens_per_second": 8494.474
    },
    {
      "epoch": 1.8112038782655535,
      "grad_norm": 1.1716349308764622,
      "learning_rate": 2.184178154160288e-07,
      "loss": 0.2795973300933838,
      "num_input_tokens_seen": 440534152,
      "step": 26900,
      "train_runtime": 51861.7193,
      "train_tokens_per_second": 8494.399
    },
    {
      "epoch": 1.8115405332615135,
      "grad_norm": 1.1346040591529505,
      "learning_rate": 2.1764553025081514e-07,
      "loss": 0.25666677951812744,
      "num_input_tokens_seen": 440616072,
      "step": 26905,
      "train_runtime": 51871.352,
      "train_tokens_per_second": 8494.401
    },
    {
      "epoch": 1.8118771882574738,
      "grad_norm": 1.1827577109203171,
      "learning_rate": 2.168745824557794e-07,
      "loss": 0.26961379051208495,
      "num_input_tokens_seen": 440697992,
      "step": 26910,
      "train_runtime": 51880.9797,
      "train_tokens_per_second": 8494.404
    },
    {
      "epoch": 1.812213843253434,
      "grad_norm": 1.2070625434222129,
      "learning_rate": 2.161049722465136e-07,
      "loss": 0.24811172485351562,
      "num_input_tokens_seen": 440779912,
      "step": 26915,
      "train_runtime": 51890.5342,
      "train_tokens_per_second": 8494.418
    },
    {
      "epoch": 1.812550498249394,
      "grad_norm": 1.331563393295033,
      "learning_rate": 2.1533669983823835e-07,
      "loss": 0.2613528728485107,
      "num_input_tokens_seen": 440861832,
      "step": 26920,
      "train_runtime": 51900.221,
      "train_tokens_per_second": 8494.411
    },
    {
      "epoch": 1.812887153245354,
      "grad_norm": 1.1070044364548977,
      "learning_rate": 2.1456976544579732e-07,
      "loss": 0.2629693031311035,
      "num_input_tokens_seen": 440943752,
      "step": 26925,
      "train_runtime": 51909.7215,
      "train_tokens_per_second": 8494.435
    },
    {
      "epoch": 1.8132238082413143,
      "grad_norm": 1.0957756733929789,
      "learning_rate": 2.138041692836623e-07,
      "loss": 0.27999539375305177,
      "num_input_tokens_seen": 441025024,
      "step": 26930,
      "train_runtime": 51919.4404,
      "train_tokens_per_second": 8494.41
    },
    {
      "epoch": 1.8135604632372746,
      "grad_norm": 1.2555847435993517,
      "learning_rate": 2.1303991156592818e-07,
      "loss": 0.28760876655578616,
      "num_input_tokens_seen": 441106944,
      "step": 26935,
      "train_runtime": 51929.0543,
      "train_tokens_per_second": 8494.415
    },
    {
      "epoch": 1.8138971182332346,
      "grad_norm": 1.2167278954287397,
      "learning_rate": 2.122769925063195e-07,
      "loss": 0.2918863296508789,
      "num_input_tokens_seen": 441188864,
      "step": 26940,
      "train_runtime": 51938.5581,
      "train_tokens_per_second": 8494.438
    },
    {
      "epoch": 1.8142337732291947,
      "grad_norm": 1.2331365609551945,
      "learning_rate": 2.1151541231818174e-07,
      "loss": 0.25782017707824706,
      "num_input_tokens_seen": 441270784,
      "step": 26945,
      "train_runtime": 51948.1569,
      "train_tokens_per_second": 8494.445
    },
    {
      "epoch": 1.814570428225155,
      "grad_norm": 1.332748262276811,
      "learning_rate": 2.1075517121449063e-07,
      "loss": 0.3037875175476074,
      "num_input_tokens_seen": 441352704,
      "step": 26950,
      "train_runtime": 51957.6566,
      "train_tokens_per_second": 8494.469
    },
    {
      "epoch": 1.8149070832211152,
      "grad_norm": 1.1771199690158862,
      "learning_rate": 2.0999626940784334e-07,
      "loss": 0.25164594650268557,
      "num_input_tokens_seen": 441434624,
      "step": 26955,
      "train_runtime": 51967.2981,
      "train_tokens_per_second": 8494.469
    },
    {
      "epoch": 1.8152437382170752,
      "grad_norm": 1.3761829660046418,
      "learning_rate": 2.0923870711046568e-07,
      "loss": 0.27346527576446533,
      "num_input_tokens_seen": 441516544,
      "step": 26960,
      "train_runtime": 51976.8028,
      "train_tokens_per_second": 8494.492
    },
    {
      "epoch": 1.8155803932130352,
      "grad_norm": 1.1175356994372503,
      "learning_rate": 2.0848248453420704e-07,
      "loss": 0.24761862754821778,
      "num_input_tokens_seen": 441598464,
      "step": 26965,
      "train_runtime": 51986.4256,
      "train_tokens_per_second": 8494.496
    },
    {
      "epoch": 1.8159170482089955,
      "grad_norm": 1.278498888658581,
      "learning_rate": 2.077276018905422e-07,
      "loss": 0.2714258670806885,
      "num_input_tokens_seen": 441680384,
      "step": 26970,
      "train_runtime": 51996.1479,
      "train_tokens_per_second": 8494.483
    },
    {
      "epoch": 1.8162537032049557,
      "grad_norm": 1.266708439957689,
      "learning_rate": 2.0697405939057225e-07,
      "loss": 0.27606050968170165,
      "num_input_tokens_seen": 441762304,
      "step": 26975,
      "train_runtime": 52006.2976,
      "train_tokens_per_second": 8494.4
    },
    {
      "epoch": 1.8165903582009157,
      "grad_norm": 1.0253573178294773,
      "learning_rate": 2.0622185724502253e-07,
      "loss": 0.2583176612854004,
      "num_input_tokens_seen": 441844224,
      "step": 26980,
      "train_runtime": 52015.9669,
      "train_tokens_per_second": 8494.396
    },
    {
      "epoch": 1.8169270131968758,
      "grad_norm": 1.0793322272168908,
      "learning_rate": 2.0547099566424367e-07,
      "loss": 0.255094051361084,
      "num_input_tokens_seen": 441926144,
      "step": 26985,
      "train_runtime": 52025.4891,
      "train_tokens_per_second": 8494.416
    },
    {
      "epoch": 1.817263668192836,
      "grad_norm": 1.1238671689549684,
      "learning_rate": 2.0472147485821325e-07,
      "loss": 0.282134485244751,
      "num_input_tokens_seen": 442008064,
      "step": 26990,
      "train_runtime": 52035.1188,
      "train_tokens_per_second": 8494.418
    },
    {
      "epoch": 1.8176003231887963,
      "grad_norm": 1.1040985696052366,
      "learning_rate": 2.0397329503653084e-07,
      "loss": 0.27546072006225586,
      "num_input_tokens_seen": 442089984,
      "step": 26995,
      "train_runtime": 52044.8025,
      "train_tokens_per_second": 8494.412
    },
    {
      "epoch": 1.8179369781847563,
      "grad_norm": 1.0825920275932175,
      "learning_rate": 2.03226456408423e-07,
      "loss": 0.31856465339660645,
      "num_input_tokens_seen": 442171904,
      "step": 27000,
      "train_runtime": 52054.5006,
      "train_tokens_per_second": 8494.403
    },
    {
      "epoch": 1.8182736331807163,
      "grad_norm": 1.13565281838699,
      "learning_rate": 2.024809591827409e-07,
      "loss": 0.2998762369155884,
      "num_input_tokens_seen": 442253824,
      "step": 27005,
      "train_runtime": 52064.2026,
      "train_tokens_per_second": 8494.393
    },
    {
      "epoch": 1.8186102881766766,
      "grad_norm": 1.0641548517253725,
      "learning_rate": 2.0173680356796066e-07,
      "loss": 0.25549869537353515,
      "num_input_tokens_seen": 442335744,
      "step": 27010,
      "train_runtime": 52073.9029,
      "train_tokens_per_second": 8494.384
    },
    {
      "epoch": 1.8189469431726368,
      "grad_norm": 1.0056181880691228,
      "learning_rate": 2.00993989772183e-07,
      "loss": 0.2881868600845337,
      "num_input_tokens_seen": 442417664,
      "step": 27015,
      "train_runtime": 52083.5891,
      "train_tokens_per_second": 8494.377
    },
    {
      "epoch": 1.8192835981685969,
      "grad_norm": 1.0319341900437105,
      "learning_rate": 2.0025251800313284e-07,
      "loss": 0.3028263092041016,
      "num_input_tokens_seen": 442499584,
      "step": 27020,
      "train_runtime": 52093.2637,
      "train_tokens_per_second": 8494.372
    },
    {
      "epoch": 1.8196202531645569,
      "grad_norm": 1.277780988337254,
      "learning_rate": 1.9951238846816045e-07,
      "loss": 0.28616762161254883,
      "num_input_tokens_seen": 442581504,
      "step": 27025,
      "train_runtime": 52102.9705,
      "train_tokens_per_second": 8494.362
    },
    {
      "epoch": 1.8199569081605171,
      "grad_norm": 1.0156758885465393,
      "learning_rate": 1.987736013742414e-07,
      "loss": 0.2605074167251587,
      "num_input_tokens_seen": 442663424,
      "step": 27030,
      "train_runtime": 52112.6629,
      "train_tokens_per_second": 8494.354
    },
    {
      "epoch": 1.8202935631564774,
      "grad_norm": 1.1481512196780654,
      "learning_rate": 1.9803615692797375e-07,
      "loss": 0.23854658603668213,
      "num_input_tokens_seen": 442745344,
      "step": 27035,
      "train_runtime": 52122.1842,
      "train_tokens_per_second": 8494.374
    },
    {
      "epoch": 1.8206302181524374,
      "grad_norm": 1.1258691251875934,
      "learning_rate": 1.9730005533558417e-07,
      "loss": 0.27958581447601316,
      "num_input_tokens_seen": 442827264,
      "step": 27040,
      "train_runtime": 52132.023,
      "train_tokens_per_second": 8494.343
    },
    {
      "epoch": 1.8209668731483974,
      "grad_norm": 1.0346747551904232,
      "learning_rate": 1.9656529680291802e-07,
      "loss": 0.26820006370544436,
      "num_input_tokens_seen": 442909184,
      "step": 27045,
      "train_runtime": 52141.6843,
      "train_tokens_per_second": 8494.34
    },
    {
      "epoch": 1.8213035281443577,
      "grad_norm": 0.999960361770506,
      "learning_rate": 1.9583188153544986e-07,
      "loss": 0.26468691825866697,
      "num_input_tokens_seen": 442991104,
      "step": 27050,
      "train_runtime": 52151.3625,
      "train_tokens_per_second": 8494.334
    },
    {
      "epoch": 1.821640183140318,
      "grad_norm": 1.1511613958115803,
      "learning_rate": 1.9509980973827724e-07,
      "loss": 0.2831921100616455,
      "num_input_tokens_seen": 443073024,
      "step": 27055,
      "train_runtime": 52161.0678,
      "train_tokens_per_second": 8494.324
    },
    {
      "epoch": 1.821976838136278,
      "grad_norm": 1.2641302111465336,
      "learning_rate": 1.943690816161209e-07,
      "loss": 0.2850434064865112,
      "num_input_tokens_seen": 443154944,
      "step": 27060,
      "train_runtime": 52170.7585,
      "train_tokens_per_second": 8494.317
    },
    {
      "epoch": 1.822313493132238,
      "grad_norm": 1.245252406888258,
      "learning_rate": 1.9363969737332677e-07,
      "loss": 0.27038586139678955,
      "num_input_tokens_seen": 443236864,
      "step": 27065,
      "train_runtime": 52180.4367,
      "train_tokens_per_second": 8494.311
    },
    {
      "epoch": 1.8226501481281983,
      "grad_norm": 1.090333170063514,
      "learning_rate": 1.9291165721386507e-07,
      "loss": 0.23222060203552247,
      "num_input_tokens_seen": 443318784,
      "step": 27070,
      "train_runtime": 52190.1269,
      "train_tokens_per_second": 8494.304
    },
    {
      "epoch": 1.8229868031241585,
      "grad_norm": 1.2093765766780489,
      "learning_rate": 1.9218496134132957e-07,
      "loss": 0.29223082065582273,
      "num_input_tokens_seen": 443400592,
      "step": 27075,
      "train_runtime": 52200.3068,
      "train_tokens_per_second": 8494.214
    },
    {
      "epoch": 1.8233234581201185,
      "grad_norm": 1.3033030827441348,
      "learning_rate": 1.9145960995893831e-07,
      "loss": 0.2871379375457764,
      "num_input_tokens_seen": 443480984,
      "step": 27080,
      "train_runtime": 52209.9093,
      "train_tokens_per_second": 8494.192
    },
    {
      "epoch": 1.8236601131160786,
      "grad_norm": 1.1906465395831887,
      "learning_rate": 1.9073560326953344e-07,
      "loss": 0.2633479595184326,
      "num_input_tokens_seen": 443562904,
      "step": 27085,
      "train_runtime": 52219.4148,
      "train_tokens_per_second": 8494.214
    },
    {
      "epoch": 1.8239967681120388,
      "grad_norm": 1.0778874008884227,
      "learning_rate": 1.9001294147558303e-07,
      "loss": 0.2865946054458618,
      "num_input_tokens_seen": 443644376,
      "step": 27090,
      "train_runtime": 52229.3861,
      "train_tokens_per_second": 8494.153
    },
    {
      "epoch": 1.824333423107999,
      "grad_norm": 1.1823155432185448,
      "learning_rate": 1.892916247791743e-07,
      "loss": 0.2810443162918091,
      "num_input_tokens_seen": 443726296,
      "step": 27095,
      "train_runtime": 52241.9281,
      "train_tokens_per_second": 8493.681
    },
    {
      "epoch": 1.824670078103959,
      "grad_norm": 1.201773526088952,
      "learning_rate": 1.885716533820231e-07,
      "loss": 0.2878026247024536,
      "num_input_tokens_seen": 443808216,
      "step": 27100,
      "train_runtime": 52267.8796,
      "train_tokens_per_second": 8491.032
    },
    {
      "epoch": 1.8250067330999191,
      "grad_norm": 1.0769005957203537,
      "learning_rate": 1.878530274854662e-07,
      "loss": 0.26779096126556395,
      "num_input_tokens_seen": 443890136,
      "step": 27105,
      "train_runtime": 52277.451,
      "train_tokens_per_second": 8491.044
    },
    {
      "epoch": 1.8253433880958794,
      "grad_norm": 1.046749752642926,
      "learning_rate": 1.8713574729046557e-07,
      "loss": 0.28159103393554685,
      "num_input_tokens_seen": 443972056,
      "step": 27110,
      "train_runtime": 52286.9358,
      "train_tokens_per_second": 8491.07
    },
    {
      "epoch": 1.8256800430918396,
      "grad_norm": 1.1635359895994954,
      "learning_rate": 1.8641981299760582e-07,
      "loss": 0.26672265529632566,
      "num_input_tokens_seen": 444053976,
      "step": 27115,
      "train_runtime": 52296.4228,
      "train_tokens_per_second": 8491.097
    },
    {
      "epoch": 1.8260166980877997,
      "grad_norm": 1.029649299901841,
      "learning_rate": 1.857052248070962e-07,
      "loss": 0.2774186611175537,
      "num_input_tokens_seen": 444135896,
      "step": 27120,
      "train_runtime": 52306.0662,
      "train_tokens_per_second": 8491.097
    },
    {
      "epoch": 1.8263533530837597,
      "grad_norm": 1.1133444559447474,
      "learning_rate": 1.8499198291876907e-07,
      "loss": 0.26715450286865233,
      "num_input_tokens_seen": 444217816,
      "step": 27125,
      "train_runtime": 52315.5588,
      "train_tokens_per_second": 8491.122
    },
    {
      "epoch": 1.82669000807972,
      "grad_norm": 1.0774702851901783,
      "learning_rate": 1.8428008753207937e-07,
      "loss": 0.3076749324798584,
      "num_input_tokens_seen": 444299736,
      "step": 27130,
      "train_runtime": 52325.193,
      "train_tokens_per_second": 8491.125
    },
    {
      "epoch": 1.8270266630756802,
      "grad_norm": 1.2063865611922595,
      "learning_rate": 1.8356953884610673e-07,
      "loss": 0.30019407272338866,
      "num_input_tokens_seen": 444381656,
      "step": 27135,
      "train_runtime": 52334.8672,
      "train_tokens_per_second": 8491.12
    },
    {
      "epoch": 1.8273633180716402,
      "grad_norm": 1.1640117455397467,
      "learning_rate": 1.8286033705955554e-07,
      "loss": 0.2755995750427246,
      "num_input_tokens_seen": 444463576,
      "step": 27140,
      "train_runtime": 52344.4878,
      "train_tokens_per_second": 8491.125
    },
    {
      "epoch": 1.8276999730676002,
      "grad_norm": 1.074305233556054,
      "learning_rate": 1.8215248237074879e-07,
      "loss": 0.26770973205566406,
      "num_input_tokens_seen": 444545496,
      "step": 27145,
      "train_runtime": 52353.9809,
      "train_tokens_per_second": 8491.15
    },
    {
      "epoch": 1.8280366280635605,
      "grad_norm": 1.0657179808643806,
      "learning_rate": 1.8144597497763816e-07,
      "loss": 0.2614928960800171,
      "num_input_tokens_seen": 444626728,
      "step": 27150,
      "train_runtime": 52364.4934,
      "train_tokens_per_second": 8490.996
    },
    {
      "epoch": 1.8283732830595207,
      "grad_norm": 0.9905949267379227,
      "learning_rate": 1.8074081507779506e-07,
      "loss": 0.2793668031692505,
      "num_input_tokens_seen": 444708648,
      "step": 27155,
      "train_runtime": 52374.0507,
      "train_tokens_per_second": 8491.011
    },
    {
      "epoch": 1.8287099380554808,
      "grad_norm": 1.2281343017723103,
      "learning_rate": 1.8003700286841563e-07,
      "loss": 0.28421022891998293,
      "num_input_tokens_seen": 444790568,
      "step": 27160,
      "train_runtime": 52383.7495,
      "train_tokens_per_second": 8491.003
    },
    {
      "epoch": 1.8290465930514408,
      "grad_norm": 1.024850688679064,
      "learning_rate": 1.793345385463191e-07,
      "loss": 0.27779197692871094,
      "num_input_tokens_seen": 444872488,
      "step": 27165,
      "train_runtime": 52393.4531,
      "train_tokens_per_second": 8490.994
    },
    {
      "epoch": 1.829383248047401,
      "grad_norm": 1.1496050950911527,
      "learning_rate": 1.7863342230794667e-07,
      "loss": 0.3001097202301025,
      "num_input_tokens_seen": 444954408,
      "step": 27170,
      "train_runtime": 52403.145,
      "train_tokens_per_second": 8490.987
    },
    {
      "epoch": 1.8297199030433613,
      "grad_norm": 1.1330268617557742,
      "learning_rate": 1.7793365434936316e-07,
      "loss": 0.24745612144470214,
      "num_input_tokens_seen": 445036328,
      "step": 27175,
      "train_runtime": 52412.7295,
      "train_tokens_per_second": 8490.997
    },
    {
      "epoch": 1.8300565580393213,
      "grad_norm": 1.1114551369554435,
      "learning_rate": 1.7723523486625704e-07,
      "loss": 0.2766072750091553,
      "num_input_tokens_seen": 445118248,
      "step": 27180,
      "train_runtime": 52422.2267,
      "train_tokens_per_second": 8491.021
    },
    {
      "epoch": 1.8303932130352814,
      "grad_norm": 1.258123711056768,
      "learning_rate": 1.7653816405393766e-07,
      "loss": 0.2771531820297241,
      "num_input_tokens_seen": 445200168,
      "step": 27185,
      "train_runtime": 52431.7223,
      "train_tokens_per_second": 8491.046
    },
    {
      "epoch": 1.8307298680312416,
      "grad_norm": 1.1395376118627452,
      "learning_rate": 1.758424421073418e-07,
      "loss": 0.28829591274261473,
      "num_input_tokens_seen": 445282088,
      "step": 27190,
      "train_runtime": 52441.2207,
      "train_tokens_per_second": 8491.07
    },
    {
      "epoch": 1.8310665230272019,
      "grad_norm": 0.986049452016504,
      "learning_rate": 1.7514806922102222e-07,
      "loss": 0.2547647476196289,
      "num_input_tokens_seen": 445364008,
      "step": 27195,
      "train_runtime": 52450.8536,
      "train_tokens_per_second": 8491.073
    },
    {
      "epoch": 1.8314031780231619,
      "grad_norm": 1.2166182263452707,
      "learning_rate": 1.7445504558916027e-07,
      "loss": 0.28039448261260985,
      "num_input_tokens_seen": 445445928,
      "step": 27200,
      "train_runtime": 52460.4753,
      "train_tokens_per_second": 8491.077
    },
    {
      "epoch": 1.831739833019122,
      "grad_norm": 1.0449795560252735,
      "learning_rate": 1.7376337140555754e-07,
      "loss": 0.2490474224090576,
      "num_input_tokens_seen": 445527848,
      "step": 27205,
      "train_runtime": 52470.0957,
      "train_tokens_per_second": 8491.081
    },
    {
      "epoch": 1.8320764880150822,
      "grad_norm": 1.1312290124634028,
      "learning_rate": 1.7307304686363825e-07,
      "loss": 0.26792011260986326,
      "num_input_tokens_seen": 445609768,
      "step": 27210,
      "train_runtime": 52479.5951,
      "train_tokens_per_second": 8491.105
    },
    {
      "epoch": 1.8324131430110424,
      "grad_norm": 1.15492660620784,
      "learning_rate": 1.723840721564496e-07,
      "loss": 0.2589019775390625,
      "num_input_tokens_seen": 445691688,
      "step": 27215,
      "train_runtime": 52489.1739,
      "train_tokens_per_second": 8491.116
    },
    {
      "epoch": 1.8327497980070024,
      "grad_norm": 1.0663567031896801,
      "learning_rate": 1.7169644747666081e-07,
      "loss": 0.2831529140472412,
      "num_input_tokens_seen": 445773608,
      "step": 27220,
      "train_runtime": 52498.7448,
      "train_tokens_per_second": 8491.129
    },
    {
      "epoch": 1.8330864530029625,
      "grad_norm": 1.1428971808721067,
      "learning_rate": 1.710101730165642e-07,
      "loss": 0.2993459939956665,
      "num_input_tokens_seen": 445855528,
      "step": 27225,
      "train_runtime": 52508.3648,
      "train_tokens_per_second": 8491.133
    },
    {
      "epoch": 1.8334231079989227,
      "grad_norm": 1.1029753066906134,
      "learning_rate": 1.7032524896807402e-07,
      "loss": 0.28165464401245116,
      "num_input_tokens_seen": 445937448,
      "step": 27230,
      "train_runtime": 52518.1616,
      "train_tokens_per_second": 8491.109
    },
    {
      "epoch": 1.833759762994883,
      "grad_norm": 0.9981454185748765,
      "learning_rate": 1.6964167552272702e-07,
      "loss": 0.2653217792510986,
      "num_input_tokens_seen": 446019368,
      "step": 27235,
      "train_runtime": 52527.777,
      "train_tokens_per_second": 8491.114
    },
    {
      "epoch": 1.834096417990843,
      "grad_norm": 1.4887821191866006,
      "learning_rate": 1.6895945287168314e-07,
      "loss": 0.30009145736694337,
      "num_input_tokens_seen": 446101288,
      "step": 27240,
      "train_runtime": 52537.2782,
      "train_tokens_per_second": 8491.138
    },
    {
      "epoch": 1.834433072986803,
      "grad_norm": 1.159743590633843,
      "learning_rate": 1.682785812057225e-07,
      "loss": 0.2698985576629639,
      "num_input_tokens_seen": 446183208,
      "step": 27245,
      "train_runtime": 52546.7842,
      "train_tokens_per_second": 8491.161
    },
    {
      "epoch": 1.8347697279827633,
      "grad_norm": 0.9834950955014213,
      "learning_rate": 1.6759906071525e-07,
      "loss": 0.28321170806884766,
      "num_input_tokens_seen": 446265128,
      "step": 27250,
      "train_runtime": 52556.4252,
      "train_tokens_per_second": 8491.162
    },
    {
      "epoch": 1.8351063829787235,
      "grad_norm": 1.2139814484699953,
      "learning_rate": 1.669208915902909e-07,
      "loss": 0.2980930805206299,
      "num_input_tokens_seen": 446347048,
      "step": 27255,
      "train_runtime": 52566.0723,
      "train_tokens_per_second": 8491.162
    },
    {
      "epoch": 1.8354430379746836,
      "grad_norm": 0.9891000283758643,
      "learning_rate": 1.6624407402049237e-07,
      "loss": 0.2622121095657349,
      "num_input_tokens_seen": 446428968,
      "step": 27260,
      "train_runtime": 52575.7891,
      "train_tokens_per_second": 8491.151
    },
    {
      "epoch": 1.8357796929706436,
      "grad_norm": 1.224071912129583,
      "learning_rate": 1.655686081951252e-07,
      "loss": 0.2882040500640869,
      "num_input_tokens_seen": 446510888,
      "step": 27265,
      "train_runtime": 52585.5255,
      "train_tokens_per_second": 8491.137
    },
    {
      "epoch": 1.8361163479666038,
      "grad_norm": 1.3862673543609811,
      "learning_rate": 1.6489449430308058e-07,
      "loss": 0.30556330680847166,
      "num_input_tokens_seen": 446592808,
      "step": 27270,
      "train_runtime": 52595.0298,
      "train_tokens_per_second": 8491.16
    },
    {
      "epoch": 1.836453002962564,
      "grad_norm": 1.1471699975465537,
      "learning_rate": 1.6422173253287323e-07,
      "loss": 0.3083841800689697,
      "num_input_tokens_seen": 446674728,
      "step": 27275,
      "train_runtime": 52604.6677,
      "train_tokens_per_second": 8491.161
    },
    {
      "epoch": 1.8367896579585241,
      "grad_norm": 1.3289123811197125,
      "learning_rate": 1.6355032307263764e-07,
      "loss": 0.27685933113098143,
      "num_input_tokens_seen": 446756648,
      "step": 27280,
      "train_runtime": 52614.2973,
      "train_tokens_per_second": 8491.164
    },
    {
      "epoch": 1.8371263129544841,
      "grad_norm": 1.0912434595422547,
      "learning_rate": 1.628802661101314e-07,
      "loss": 0.24415342807769774,
      "num_input_tokens_seen": 446838568,
      "step": 27285,
      "train_runtime": 52624.0252,
      "train_tokens_per_second": 8491.151
    },
    {
      "epoch": 1.8374629679504444,
      "grad_norm": 1.0297323291371874,
      "learning_rate": 1.6221156183273524e-07,
      "loss": 0.26701865196228025,
      "num_input_tokens_seen": 446920488,
      "step": 27290,
      "train_runtime": 52633.6846,
      "train_tokens_per_second": 8491.15
    },
    {
      "epoch": 1.8377996229464046,
      "grad_norm": 0.9750071363278997,
      "learning_rate": 1.615442104274484e-07,
      "loss": 0.29339659214019775,
      "num_input_tokens_seen": 447002408,
      "step": 27295,
      "train_runtime": 52643.249,
      "train_tokens_per_second": 8491.163
    },
    {
      "epoch": 1.8381362779423647,
      "grad_norm": 1.1371114689697948,
      "learning_rate": 1.6087821208089492e-07,
      "loss": 0.3073129653930664,
      "num_input_tokens_seen": 447084328,
      "step": 27300,
      "train_runtime": 52652.9558,
      "train_tokens_per_second": 8491.153
    },
    {
      "epoch": 1.8384729329383247,
      "grad_norm": 1.0122184709511608,
      "learning_rate": 1.6021356697931866e-07,
      "loss": 0.2558289527893066,
      "num_input_tokens_seen": 447166248,
      "step": 27305,
      "train_runtime": 52662.4645,
      "train_tokens_per_second": 8491.176
    },
    {
      "epoch": 1.838809587934285,
      "grad_norm": 0.9237341250075254,
      "learning_rate": 1.5955027530858537e-07,
      "loss": 0.2593628644943237,
      "num_input_tokens_seen": 447248168,
      "step": 27310,
      "train_runtime": 52672.084,
      "train_tokens_per_second": 8491.18
    },
    {
      "epoch": 1.8391462429302452,
      "grad_norm": 0.9842786794975925,
      "learning_rate": 1.5888833725418275e-07,
      "loss": 0.25969107151031495,
      "num_input_tokens_seen": 447330088,
      "step": 27315,
      "train_runtime": 52681.5939,
      "train_tokens_per_second": 8491.203
    },
    {
      "epoch": 1.8394828979262052,
      "grad_norm": 1.1054205813503302,
      "learning_rate": 1.582277530012194e-07,
      "loss": 0.26453776359558107,
      "num_input_tokens_seen": 447412008,
      "step": 27320,
      "train_runtime": 52691.2137,
      "train_tokens_per_second": 8491.207
    },
    {
      "epoch": 1.8398195529221653,
      "grad_norm": 1.121226762328176,
      "learning_rate": 1.5756852273442536e-07,
      "loss": 0.28026599884033204,
      "num_input_tokens_seen": 447493928,
      "step": 27325,
      "train_runtime": 52700.7123,
      "train_tokens_per_second": 8491.231
    },
    {
      "epoch": 1.8401562079181255,
      "grad_norm": 1.110156822897521,
      "learning_rate": 1.569106466381526e-07,
      "loss": 0.29210872650146485,
      "num_input_tokens_seen": 447575848,
      "step": 27330,
      "train_runtime": 52710.3229,
      "train_tokens_per_second": 8491.237
    },
    {
      "epoch": 1.8404928629140858,
      "grad_norm": 0.9748154145116261,
      "learning_rate": 1.5625412489637337e-07,
      "loss": 0.28329803943634035,
      "num_input_tokens_seen": 447657768,
      "step": 27335,
      "train_runtime": 52719.8246,
      "train_tokens_per_second": 8491.261
    },
    {
      "epoch": 1.8408295179100458,
      "grad_norm": 1.2691699444182276,
      "learning_rate": 1.555989576926842e-07,
      "loss": 0.27708234786987307,
      "num_input_tokens_seen": 447739688,
      "step": 27340,
      "train_runtime": 52729.4217,
      "train_tokens_per_second": 8491.269
    },
    {
      "epoch": 1.8411661729060058,
      "grad_norm": 1.1574548293625584,
      "learning_rate": 1.5494514521029736e-07,
      "loss": 0.29720320701599123,
      "num_input_tokens_seen": 447821608,
      "step": 27345,
      "train_runtime": 52738.9656,
      "train_tokens_per_second": 8491.285
    },
    {
      "epoch": 1.841502827901966,
      "grad_norm": 1.1342182769103342,
      "learning_rate": 1.5429268763205108e-07,
      "loss": 0.2879804611206055,
      "num_input_tokens_seen": 447903528,
      "step": 27350,
      "train_runtime": 52748.5538,
      "train_tokens_per_second": 8491.295
    },
    {
      "epoch": 1.8418394828979263,
      "grad_norm": 1.1813972511926578,
      "learning_rate": 1.5364158514040328e-07,
      "loss": 0.284319543838501,
      "num_input_tokens_seen": 447985448,
      "step": 27355,
      "train_runtime": 52758.172,
      "train_tokens_per_second": 8491.3
    },
    {
      "epoch": 1.8421761378938863,
      "grad_norm": 1.3766911943034994,
      "learning_rate": 1.5299183791743223e-07,
      "loss": 0.28772845268249514,
      "num_input_tokens_seen": 448067368,
      "step": 27360,
      "train_runtime": 52768.0299,
      "train_tokens_per_second": 8491.266
    },
    {
      "epoch": 1.8425127928898464,
      "grad_norm": 1.2591333951862278,
      "learning_rate": 1.523434461448381e-07,
      "loss": 0.2828775405883789,
      "num_input_tokens_seen": 448149288,
      "step": 27365,
      "train_runtime": 52777.559,
      "train_tokens_per_second": 8491.285
    },
    {
      "epoch": 1.8428494478858066,
      "grad_norm": 1.0800310826017607,
      "learning_rate": 1.5169641000394086e-07,
      "loss": 0.27183685302734373,
      "num_input_tokens_seen": 448231208,
      "step": 27370,
      "train_runtime": 52787.2237,
      "train_tokens_per_second": 8491.282
    },
    {
      "epoch": 1.8431861028817669,
      "grad_norm": 1.249178411902346,
      "learning_rate": 1.5105072967568246e-07,
      "loss": 0.2980545997619629,
      "num_input_tokens_seen": 448313128,
      "step": 27375,
      "train_runtime": 52796.8816,
      "train_tokens_per_second": 8491.28
    },
    {
      "epoch": 1.843522757877727,
      "grad_norm": 0.9476693751001515,
      "learning_rate": 1.504064053406251e-07,
      "loss": 0.2505303621292114,
      "num_input_tokens_seen": 448395048,
      "step": 27380,
      "train_runtime": 52806.61,
      "train_tokens_per_second": 8491.267
    },
    {
      "epoch": 1.843859412873687,
      "grad_norm": 1.3923589410866901,
      "learning_rate": 1.4976343717895192e-07,
      "loss": 0.27924845218658445,
      "num_input_tokens_seen": 448476968,
      "step": 27385,
      "train_runtime": 52816.3721,
      "train_tokens_per_second": 8491.249
    },
    {
      "epoch": 1.8441960678696472,
      "grad_norm": 1.2225202940686213,
      "learning_rate": 1.491218253704685e-07,
      "loss": 0.2910027980804443,
      "num_input_tokens_seen": 448558888,
      "step": 27390,
      "train_runtime": 52826.0783,
      "train_tokens_per_second": 8491.24
    },
    {
      "epoch": 1.8445327228656074,
      "grad_norm": 1.1790267083465007,
      "learning_rate": 1.484815700945974e-07,
      "loss": 0.3123300075531006,
      "num_input_tokens_seen": 448640808,
      "step": 27395,
      "train_runtime": 52835.7762,
      "train_tokens_per_second": 8491.232
    },
    {
      "epoch": 1.8448693778615675,
      "grad_norm": 1.3228514376972307,
      "learning_rate": 1.478426715303849e-07,
      "loss": 0.2920565128326416,
      "num_input_tokens_seen": 448722728,
      "step": 27400,
      "train_runtime": 52845.4274,
      "train_tokens_per_second": 8491.231
    },
    {
      "epoch": 1.8452060328575275,
      "grad_norm": 1.001417862105704,
      "learning_rate": 1.4720512985649694e-07,
      "loss": 0.24830999374389648,
      "num_input_tokens_seen": 448804648,
      "step": 27405,
      "train_runtime": 52854.9374,
      "train_tokens_per_second": 8491.253
    },
    {
      "epoch": 1.8455426878534877,
      "grad_norm": 1.2474100288132584,
      "learning_rate": 1.4656894525122034e-07,
      "loss": 0.2777833938598633,
      "num_input_tokens_seen": 448886568,
      "step": 27410,
      "train_runtime": 52864.4539,
      "train_tokens_per_second": 8491.274
    },
    {
      "epoch": 1.845879342849448,
      "grad_norm": 1.1063336989344037,
      "learning_rate": 1.459341178924617e-07,
      "loss": 0.28683035373687743,
      "num_input_tokens_seen": 448968488,
      "step": 27415,
      "train_runtime": 52874.1485,
      "train_tokens_per_second": 8491.267
    },
    {
      "epoch": 1.846215997845408,
      "grad_norm": 1.0683534431109925,
      "learning_rate": 1.453006479577479e-07,
      "loss": 0.2692425727844238,
      "num_input_tokens_seen": 449050024,
      "step": 27420,
      "train_runtime": 52884.947,
      "train_tokens_per_second": 8491.074
    },
    {
      "epoch": 1.846552652841368,
      "grad_norm": 1.0673346363867502,
      "learning_rate": 1.4466853562422778e-07,
      "loss": 0.2639143943786621,
      "num_input_tokens_seen": 449131944,
      "step": 27425,
      "train_runtime": 52894.8379,
      "train_tokens_per_second": 8491.035
    },
    {
      "epoch": 1.8468893078373283,
      "grad_norm": 1.1073321577296205,
      "learning_rate": 1.440377810686694e-07,
      "loss": 0.25376894474029543,
      "num_input_tokens_seen": 449213864,
      "step": 27430,
      "train_runtime": 52904.5724,
      "train_tokens_per_second": 8491.022
    },
    {
      "epoch": 1.8472259628332885,
      "grad_norm": 1.2517332894526798,
      "learning_rate": 1.4340838446746052e-07,
      "loss": 0.2920989990234375,
      "num_input_tokens_seen": 449295784,
      "step": 27435,
      "train_runtime": 52914.1856,
      "train_tokens_per_second": 8491.027
    },
    {
      "epoch": 1.8475626178292486,
      "grad_norm": 1.0127783630699754,
      "learning_rate": 1.4278034599661094e-07,
      "loss": 0.27201528549194337,
      "num_input_tokens_seen": 449377704,
      "step": 27440,
      "train_runtime": 52923.8579,
      "train_tokens_per_second": 8491.023
    },
    {
      "epoch": 1.8478992728252086,
      "grad_norm": 1.0647275828848703,
      "learning_rate": 1.421536658317485e-07,
      "loss": 0.27317070960998535,
      "num_input_tokens_seen": 449459624,
      "step": 27445,
      "train_runtime": 52933.5655,
      "train_tokens_per_second": 8491.014
    },
    {
      "epoch": 1.8482359278211689,
      "grad_norm": 1.112442903194295,
      "learning_rate": 1.41528344148123e-07,
      "loss": 0.27164511680603026,
      "num_input_tokens_seen": 449541544,
      "step": 27450,
      "train_runtime": 52943.241,
      "train_tokens_per_second": 8491.009
    },
    {
      "epoch": 1.848572582817129,
      "grad_norm": 1.194427078679265,
      "learning_rate": 1.40904381120604e-07,
      "loss": 0.2928244829177856,
      "num_input_tokens_seen": 449623464,
      "step": 27455,
      "train_runtime": 52952.7561,
      "train_tokens_per_second": 8491.03
    },
    {
      "epoch": 1.8489092378130891,
      "grad_norm": 1.1263539881757259,
      "learning_rate": 1.4028177692367973e-07,
      "loss": 0.2794511079788208,
      "num_input_tokens_seen": 449705384,
      "step": 27460,
      "train_runtime": 52962.3848,
      "train_tokens_per_second": 8491.034
    },
    {
      "epoch": 1.8492458928090492,
      "grad_norm": 1.1678803591057836,
      "learning_rate": 1.3966053173146032e-07,
      "loss": 0.2991274356842041,
      "num_input_tokens_seen": 449787304,
      "step": 27465,
      "train_runtime": 52971.9908,
      "train_tokens_per_second": 8491.04
    },
    {
      "epoch": 1.8495825478050094,
      "grad_norm": 1.0854816080871712,
      "learning_rate": 1.3904064571767462e-07,
      "loss": 0.2657889127731323,
      "num_input_tokens_seen": 449869224,
      "step": 27470,
      "train_runtime": 52981.5207,
      "train_tokens_per_second": 8491.059
    },
    {
      "epoch": 1.8499192028009697,
      "grad_norm": 1.2161871626207357,
      "learning_rate": 1.3842211905567227e-07,
      "loss": 0.27187204360961914,
      "num_input_tokens_seen": 449951144,
      "step": 27475,
      "train_runtime": 52991.233,
      "train_tokens_per_second": 8491.049
    },
    {
      "epoch": 1.8502558577969297,
      "grad_norm": 1.2730388563565873,
      "learning_rate": 1.378049519184216e-07,
      "loss": 0.2959883689880371,
      "num_input_tokens_seen": 450033064,
      "step": 27480,
      "train_runtime": 53000.7521,
      "train_tokens_per_second": 8491.069
    },
    {
      "epoch": 1.8505925127928897,
      "grad_norm": 1.0332447708845636,
      "learning_rate": 1.3718914447851116e-07,
      "loss": 0.26836867332458497,
      "num_input_tokens_seen": 450114984,
      "step": 27485,
      "train_runtime": 53010.2539,
      "train_tokens_per_second": 8491.093
    },
    {
      "epoch": 1.85092916778885,
      "grad_norm": 1.0294500041841335,
      "learning_rate": 1.3657469690815106e-07,
      "loss": 0.2583444833755493,
      "num_input_tokens_seen": 450196904,
      "step": 27490,
      "train_runtime": 53019.8445,
      "train_tokens_per_second": 8491.102
    },
    {
      "epoch": 1.8512658227848102,
      "grad_norm": 0.9391080536618438,
      "learning_rate": 1.359616093791677e-07,
      "loss": 0.2774820327758789,
      "num_input_tokens_seen": 450278824,
      "step": 27495,
      "train_runtime": 53029.3455,
      "train_tokens_per_second": 8491.125
    },
    {
      "epoch": 1.8516024777807703,
      "grad_norm": 0.9436892750365907,
      "learning_rate": 1.353498820630106e-07,
      "loss": 0.26852192878723147,
      "num_input_tokens_seen": 450360744,
      "step": 27500,
      "train_runtime": 53038.854,
      "train_tokens_per_second": 8491.148
    },
    {
      "epoch": 1.8519391327767303,
      "grad_norm": 1.0758750384480678,
      "learning_rate": 1.3473951513074625e-07,
      "loss": 0.2783402442932129,
      "num_input_tokens_seen": 450442664,
      "step": 27505,
      "train_runtime": 53048.3587,
      "train_tokens_per_second": 8491.171
    },
    {
      "epoch": 1.8522757877726905,
      "grad_norm": 1.2913597574438873,
      "learning_rate": 1.341305087530631e-07,
      "loss": 0.2969644069671631,
      "num_input_tokens_seen": 450524584,
      "step": 27510,
      "train_runtime": 53057.8649,
      "train_tokens_per_second": 8491.193
    },
    {
      "epoch": 1.8526124427686508,
      "grad_norm": 1.229051161935786,
      "learning_rate": 1.3352286310026663e-07,
      "loss": 0.30000307559967043,
      "num_input_tokens_seen": 450606504,
      "step": 27515,
      "train_runtime": 53067.367,
      "train_tokens_per_second": 8491.217
    },
    {
      "epoch": 1.8529490977646108,
      "grad_norm": 1.4899553392577447,
      "learning_rate": 1.329165783422831e-07,
      "loss": 0.3257968664169312,
      "num_input_tokens_seen": 450688424,
      "step": 27520,
      "train_runtime": 53076.8702,
      "train_tokens_per_second": 8491.24
    },
    {
      "epoch": 1.8532857527605708,
      "grad_norm": 1.1465606667118422,
      "learning_rate": 1.3231165464865914e-07,
      "loss": 0.28253166675567626,
      "num_input_tokens_seen": 450770344,
      "step": 27525,
      "train_runtime": 53086.5289,
      "train_tokens_per_second": 8491.238
    },
    {
      "epoch": 1.853622407756531,
      "grad_norm": 1.1804523252915888,
      "learning_rate": 1.3170809218855884e-07,
      "loss": 0.2739693641662598,
      "num_input_tokens_seen": 450852264,
      "step": 27530,
      "train_runtime": 53096.1118,
      "train_tokens_per_second": 8491.248
    },
    {
      "epoch": 1.8539590627524913,
      "grad_norm": 1.001644348936042,
      "learning_rate": 1.3110589113076554e-07,
      "loss": 0.2622534275054932,
      "num_input_tokens_seen": 450934184,
      "step": 27535,
      "train_runtime": 53105.7696,
      "train_tokens_per_second": 8491.247
    },
    {
      "epoch": 1.8542957177484514,
      "grad_norm": 1.2262932834480178,
      "learning_rate": 1.3050505164368566e-07,
      "loss": 0.29141550064086913,
      "num_input_tokens_seen": 451016104,
      "step": 27540,
      "train_runtime": 53115.4466,
      "train_tokens_per_second": 8491.242
    },
    {
      "epoch": 1.8546323727444114,
      "grad_norm": 1.0007054101164123,
      "learning_rate": 1.299055738953392e-07,
      "loss": 0.2619499683380127,
      "num_input_tokens_seen": 451098024,
      "step": 27545,
      "train_runtime": 53124.9564,
      "train_tokens_per_second": 8491.264
    },
    {
      "epoch": 1.8549690277403716,
      "grad_norm": 0.9369319972932061,
      "learning_rate": 1.293074580533704e-07,
      "loss": 0.2778913974761963,
      "num_input_tokens_seen": 451179944,
      "step": 27550,
      "train_runtime": 53134.4615,
      "train_tokens_per_second": 8491.287
    },
    {
      "epoch": 1.855305682736332,
      "grad_norm": 1.2042986167293561,
      "learning_rate": 1.2871070428503885e-07,
      "loss": 0.2869423866271973,
      "num_input_tokens_seen": 451261864,
      "step": 27555,
      "train_runtime": 53143.9628,
      "train_tokens_per_second": 8491.31
    },
    {
      "epoch": 1.855642337732292,
      "grad_norm": 0.9825893050545123,
      "learning_rate": 1.2811531275722543e-07,
      "loss": 0.3102236270904541,
      "num_input_tokens_seen": 451343784,
      "step": 27560,
      "train_runtime": 53153.4603,
      "train_tokens_per_second": 8491.334
    },
    {
      "epoch": 1.855978992728252,
      "grad_norm": 1.0267969961747645,
      "learning_rate": 1.275212836364298e-07,
      "loss": 0.2785353183746338,
      "num_input_tokens_seen": 451425704,
      "step": 27565,
      "train_runtime": 53162.9626,
      "train_tokens_per_second": 8491.357
    },
    {
      "epoch": 1.8563156477242122,
      "grad_norm": 1.240849720427593,
      "learning_rate": 1.2692861708877014e-07,
      "loss": 0.285399866104126,
      "num_input_tokens_seen": 451507624,
      "step": 27570,
      "train_runtime": 53172.5093,
      "train_tokens_per_second": 8491.373
    },
    {
      "epoch": 1.8566523027201725,
      "grad_norm": 1.1523692008244981,
      "learning_rate": 1.2633731327998332e-07,
      "loss": 0.27136542797088625,
      "num_input_tokens_seen": 451589544,
      "step": 27575,
      "train_runtime": 53182.0973,
      "train_tokens_per_second": 8491.383
    },
    {
      "epoch": 1.8569889577161325,
      "grad_norm": 1.088819074125676,
      "learning_rate": 1.25747372375426e-07,
      "loss": 0.2938649654388428,
      "num_input_tokens_seen": 451671464,
      "step": 27580,
      "train_runtime": 53191.662,
      "train_tokens_per_second": 8491.396
    },
    {
      "epoch": 1.8573256127120925,
      "grad_norm": 1.1540666815102651,
      "learning_rate": 1.251587945400723e-07,
      "loss": 0.3031048059463501,
      "num_input_tokens_seen": 451753384,
      "step": 27585,
      "train_runtime": 53201.2806,
      "train_tokens_per_second": 8491.401
    },
    {
      "epoch": 1.8576622677080528,
      "grad_norm": 1.176890476053536,
      "learning_rate": 1.2457157993851832e-07,
      "loss": 0.28957154750823977,
      "num_input_tokens_seen": 451835304,
      "step": 27590,
      "train_runtime": 53210.8602,
      "train_tokens_per_second": 8491.411
    },
    {
      "epoch": 1.857998922704013,
      "grad_norm": 1.274761974250832,
      "learning_rate": 1.2398572873497438e-07,
      "loss": 0.27170577049255373,
      "num_input_tokens_seen": 451917224,
      "step": 27595,
      "train_runtime": 53220.4641,
      "train_tokens_per_second": 8491.418
    },
    {
      "epoch": 1.858335577699973,
      "grad_norm": 1.1016465143028122,
      "learning_rate": 1.2340124109327334e-07,
      "loss": 0.28040645122528074,
      "num_input_tokens_seen": 451999144,
      "step": 27600,
      "train_runtime": 53230.1428,
      "train_tokens_per_second": 8491.413
    },
    {
      "epoch": 1.858672232695933,
      "grad_norm": 1.1906897706608712,
      "learning_rate": 1.22818117176865e-07,
      "loss": 0.28040080070495604,
      "num_input_tokens_seen": 452081064,
      "step": 27605,
      "train_runtime": 53239.7842,
      "train_tokens_per_second": 8491.414
    },
    {
      "epoch": 1.8590088876918933,
      "grad_norm": 1.2034549212325272,
      "learning_rate": 1.2223635714881777e-07,
      "loss": 0.27490849494934083,
      "num_input_tokens_seen": 452162984,
      "step": 27610,
      "train_runtime": 53249.4498,
      "train_tokens_per_second": 8491.411
    },
    {
      "epoch": 1.8593455426878536,
      "grad_norm": 1.108752586681916,
      "learning_rate": 1.2165596117181931e-07,
      "loss": 0.26362311840057373,
      "num_input_tokens_seen": 452244904,
      "step": 27615,
      "train_runtime": 53259.1667,
      "train_tokens_per_second": 8491.4
    },
    {
      "epoch": 1.8596821976838136,
      "grad_norm": 1.0381822996400407,
      "learning_rate": 1.2107692940817594e-07,
      "loss": 0.2844693660736084,
      "num_input_tokens_seen": 452326696,
      "step": 27620,
      "train_runtime": 53269.2646,
      "train_tokens_per_second": 8491.326
    },
    {
      "epoch": 1.8600188526797736,
      "grad_norm": 1.244627578692812,
      "learning_rate": 1.2049926201981088e-07,
      "loss": 0.285171365737915,
      "num_input_tokens_seen": 452408616,
      "step": 27625,
      "train_runtime": 53279.0343,
      "train_tokens_per_second": 8491.307
    },
    {
      "epoch": 1.8603555076757339,
      "grad_norm": 1.1675892247687545,
      "learning_rate": 1.199229591682677e-07,
      "loss": 0.31316025257110597,
      "num_input_tokens_seen": 452490536,
      "step": 27630,
      "train_runtime": 53292.2442,
      "train_tokens_per_second": 8490.739
    },
    {
      "epoch": 1.8606921626716941,
      "grad_norm": 1.506154382902434,
      "learning_rate": 1.193480210147069e-07,
      "loss": 0.29043004512786863,
      "num_input_tokens_seen": 452572456,
      "step": 27635,
      "train_runtime": 53301.745,
      "train_tokens_per_second": 8490.762
    },
    {
      "epoch": 1.8610288176676542,
      "grad_norm": 1.166405210251731,
      "learning_rate": 1.1877444771991042e-07,
      "loss": 0.27990083694458007,
      "num_input_tokens_seen": 452654376,
      "step": 27640,
      "train_runtime": 53311.2569,
      "train_tokens_per_second": 8490.784
    },
    {
      "epoch": 1.8613654726636142,
      "grad_norm": 1.0228724872432124,
      "learning_rate": 1.1820223944427279e-07,
      "loss": 0.2542724132537842,
      "num_input_tokens_seen": 452736296,
      "step": 27645,
      "train_runtime": 53320.7512,
      "train_tokens_per_second": 8490.809
    },
    {
      "epoch": 1.8617021276595744,
      "grad_norm": 1.006289515540561,
      "learning_rate": 1.1763139634781262e-07,
      "loss": 0.26300663948059083,
      "num_input_tokens_seen": 452818216,
      "step": 27650,
      "train_runtime": 53330.2557,
      "train_tokens_per_second": 8490.832
    },
    {
      "epoch": 1.8620387826555347,
      "grad_norm": 1.1333541070344912,
      "learning_rate": 1.1706191859016392e-07,
      "loss": 0.30287947654724123,
      "num_input_tokens_seen": 452900136,
      "step": 27655,
      "train_runtime": 53339.7556,
      "train_tokens_per_second": 8490.855
    },
    {
      "epoch": 1.8623754376514947,
      "grad_norm": 1.3593430425321962,
      "learning_rate": 1.1649380633057927e-07,
      "loss": 0.27471280097961426,
      "num_input_tokens_seen": 452981424,
      "step": 27660,
      "train_runtime": 53349.5452,
      "train_tokens_per_second": 8490.821
    },
    {
      "epoch": 1.8627120926474547,
      "grad_norm": 1.0347197165743813,
      "learning_rate": 1.1592705972792939e-07,
      "loss": 0.27384700775146487,
      "num_input_tokens_seen": 453063344,
      "step": 27665,
      "train_runtime": 53359.0542,
      "train_tokens_per_second": 8490.843
    },
    {
      "epoch": 1.863048747643415,
      "grad_norm": 1.0433155555041158,
      "learning_rate": 1.1536167894070249e-07,
      "loss": 0.29493560791015627,
      "num_input_tokens_seen": 453145264,
      "step": 27670,
      "train_runtime": 53368.7128,
      "train_tokens_per_second": 8490.841
    },
    {
      "epoch": 1.8633854026393752,
      "grad_norm": 1.1356867698604172,
      "learning_rate": 1.1479766412700655e-07,
      "loss": 0.316492486000061,
      "num_input_tokens_seen": 453227184,
      "step": 27675,
      "train_runtime": 53378.43,
      "train_tokens_per_second": 8490.83
    },
    {
      "epoch": 1.8637220576353353,
      "grad_norm": 1.0519017274750986,
      "learning_rate": 1.1423501544456595e-07,
      "loss": 0.2871866703033447,
      "num_input_tokens_seen": 453309104,
      "step": 27680,
      "train_runtime": 53388.1626,
      "train_tokens_per_second": 8490.817
    },
    {
      "epoch": 1.8640587126312953,
      "grad_norm": 0.9896043205380388,
      "learning_rate": 1.1367373305072315e-07,
      "loss": 0.24342472553253175,
      "num_input_tokens_seen": 453391024,
      "step": 27685,
      "train_runtime": 53397.6712,
      "train_tokens_per_second": 8490.839
    },
    {
      "epoch": 1.8643953676272556,
      "grad_norm": 1.0279778250066613,
      "learning_rate": 1.1311381710244096e-07,
      "loss": 0.2632030248641968,
      "num_input_tokens_seen": 453472944,
      "step": 27690,
      "train_runtime": 53407.1731,
      "train_tokens_per_second": 8490.862
    },
    {
      "epoch": 1.8647320226232158,
      "grad_norm": 0.9854189081871758,
      "learning_rate": 1.1255526775629521e-07,
      "loss": 0.2679007053375244,
      "num_input_tokens_seen": 453554864,
      "step": 27695,
      "train_runtime": 53416.7988,
      "train_tokens_per_second": 8490.866
    },
    {
      "epoch": 1.8650686776191758,
      "grad_norm": 1.1496939238135424,
      "learning_rate": 1.1199808516848487e-07,
      "loss": 0.29633369445800783,
      "num_input_tokens_seen": 453636784,
      "step": 27700,
      "train_runtime": 53426.308,
      "train_tokens_per_second": 8490.888
    },
    {
      "epoch": 1.8654053326151359,
      "grad_norm": 1.0110266030685577,
      "learning_rate": 1.1144226949482306e-07,
      "loss": 0.25179405212402345,
      "num_input_tokens_seen": 453718704,
      "step": 27705,
      "train_runtime": 53435.8187,
      "train_tokens_per_second": 8490.91
    },
    {
      "epoch": 1.865741987611096,
      "grad_norm": 1.025755425219254,
      "learning_rate": 1.1088782089074157e-07,
      "loss": 0.2758253335952759,
      "num_input_tokens_seen": 453800624,
      "step": 27710,
      "train_runtime": 53445.4269,
      "train_tokens_per_second": 8490.916
    },
    {
      "epoch": 1.8660786426070564,
      "grad_norm": 1.12176188522495,
      "learning_rate": 1.1033473951129137e-07,
      "loss": 0.2986581325531006,
      "num_input_tokens_seen": 453882544,
      "step": 27715,
      "train_runtime": 53455.1637,
      "train_tokens_per_second": 8490.902
    },
    {
      "epoch": 1.8664152976030164,
      "grad_norm": 1.140683996092135,
      "learning_rate": 1.0978302551113929e-07,
      "loss": 0.27473864555358884,
      "num_input_tokens_seen": 453964464,
      "step": 27720,
      "train_runtime": 53464.9189,
      "train_tokens_per_second": 8490.885
    },
    {
      "epoch": 1.8667519525989764,
      "grad_norm": 1.3818523731435082,
      "learning_rate": 1.0923267904457025e-07,
      "loss": 0.2746377229690552,
      "num_input_tokens_seen": 454046384,
      "step": 27725,
      "train_runtime": 53474.6769,
      "train_tokens_per_second": 8490.867
    },
    {
      "epoch": 1.8670886075949367,
      "grad_norm": 0.9961364855402809,
      "learning_rate": 1.086837002654867e-07,
      "loss": 0.2589722156524658,
      "num_input_tokens_seen": 454128304,
      "step": 27730,
      "train_runtime": 53484.1911,
      "train_tokens_per_second": 8490.889
    },
    {
      "epoch": 1.867425262590897,
      "grad_norm": 1.1931306137379385,
      "learning_rate": 1.0813608932740915e-07,
      "loss": 0.2845525979995728,
      "num_input_tokens_seen": 454210224,
      "step": 27735,
      "train_runtime": 53493.9354,
      "train_tokens_per_second": 8490.873
    },
    {
      "epoch": 1.867761917586857,
      "grad_norm": 1.370372594763464,
      "learning_rate": 1.0758984638347569e-07,
      "loss": 0.2830862522125244,
      "num_input_tokens_seen": 454292144,
      "step": 27740,
      "train_runtime": 53503.5344,
      "train_tokens_per_second": 8490.881
    },
    {
      "epoch": 1.868098572582817,
      "grad_norm": 1.2397445424119269,
      "learning_rate": 1.0704497158644022e-07,
      "loss": 0.32101666927337646,
      "num_input_tokens_seen": 454374064,
      "step": 27745,
      "train_runtime": 53513.0431,
      "train_tokens_per_second": 8490.903
    },
    {
      "epoch": 1.8684352275787772,
      "grad_norm": 0.9024599726011339,
      "learning_rate": 1.0650146508867642e-07,
      "loss": 0.24910287857055663,
      "num_input_tokens_seen": 454455984,
      "step": 27750,
      "train_runtime": 53522.544,
      "train_tokens_per_second": 8490.926
    },
    {
      "epoch": 1.8687718825747375,
      "grad_norm": 1.1623059695172444,
      "learning_rate": 1.0595932704217327e-07,
      "loss": 0.2538156032562256,
      "num_input_tokens_seen": 454537904,
      "step": 27755,
      "train_runtime": 53532.0389,
      "train_tokens_per_second": 8490.951
    },
    {
      "epoch": 1.8691085375706975,
      "grad_norm": 1.2848791379845683,
      "learning_rate": 1.0541855759853837e-07,
      "loss": 0.2784942626953125,
      "num_input_tokens_seen": 454619824,
      "step": 27760,
      "train_runtime": 53541.5515,
      "train_tokens_per_second": 8490.972
    },
    {
      "epoch": 1.8694451925666575,
      "grad_norm": 1.1522940726406916,
      "learning_rate": 1.0487915690899685e-07,
      "loss": 0.30808358192443847,
      "num_input_tokens_seen": 454701744,
      "step": 27765,
      "train_runtime": 53551.2018,
      "train_tokens_per_second": 8490.972
    },
    {
      "epoch": 1.8697818475626178,
      "grad_norm": 1.0266421103679308,
      "learning_rate": 1.0434112512438921e-07,
      "loss": 0.25338168144226075,
      "num_input_tokens_seen": 454783664,
      "step": 27770,
      "train_runtime": 53560.883,
      "train_tokens_per_second": 8490.967
    },
    {
      "epoch": 1.870118502558578,
      "grad_norm": 1.3296656477742126,
      "learning_rate": 1.0380446239517506e-07,
      "loss": 0.2546414613723755,
      "num_input_tokens_seen": 454864344,
      "step": 27775,
      "train_runtime": 53570.8208,
      "train_tokens_per_second": 8490.897
    },
    {
      "epoch": 1.870455157554538,
      "grad_norm": 1.2240753535183655,
      "learning_rate": 1.032691688714299e-07,
      "loss": 0.2746826410293579,
      "num_input_tokens_seen": 454946264,
      "step": 27780,
      "train_runtime": 53580.3356,
      "train_tokens_per_second": 8490.919
    },
    {
      "epoch": 1.870791812550498,
      "grad_norm": 1.2125537375398967,
      "learning_rate": 1.0273524470284734e-07,
      "loss": 0.2896329641342163,
      "num_input_tokens_seen": 455028184,
      "step": 27785,
      "train_runtime": 53589.9592,
      "train_tokens_per_second": 8490.922
    },
    {
      "epoch": 1.8711284675464583,
      "grad_norm": 1.1963766084629714,
      "learning_rate": 1.0220269003873739e-07,
      "loss": 0.24841551780700682,
      "num_input_tokens_seen": 455109904,
      "step": 27790,
      "train_runtime": 53600.1094,
      "train_tokens_per_second": 8490.839
    },
    {
      "epoch": 1.8714651225424186,
      "grad_norm": 1.1960090647374133,
      "learning_rate": 1.0167150502802703e-07,
      "loss": 0.2910590410232544,
      "num_input_tokens_seen": 455191824,
      "step": 27795,
      "train_runtime": 53609.7498,
      "train_tokens_per_second": 8490.84
    },
    {
      "epoch": 1.8718017775383786,
      "grad_norm": 1.1958268280499271,
      "learning_rate": 1.0114168981926075e-07,
      "loss": 0.273318886756897,
      "num_input_tokens_seen": 455273744,
      "step": 27800,
      "train_runtime": 53619.6469,
      "train_tokens_per_second": 8490.801
    },
    {
      "epoch": 1.8721384325343386,
      "grad_norm": 1.3647894644395153,
      "learning_rate": 1.006132445605995e-07,
      "loss": 0.2954791784286499,
      "num_input_tokens_seen": 455355664,
      "step": 27805,
      "train_runtime": 53629.1506,
      "train_tokens_per_second": 8490.824
    },
    {
      "epoch": 1.872475087530299,
      "grad_norm": 1.161718279927382,
      "learning_rate": 1.0008616939982118e-07,
      "loss": 0.2677618980407715,
      "num_input_tokens_seen": 455437584,
      "step": 27810,
      "train_runtime": 53638.7767,
      "train_tokens_per_second": 8490.827
    },
    {
      "epoch": 1.8728117425262591,
      "grad_norm": 1.2886945134166725,
      "learning_rate": 9.956046448432177e-08,
      "loss": 0.2799408435821533,
      "num_input_tokens_seen": 455519504,
      "step": 27815,
      "train_runtime": 53648.3895,
      "train_tokens_per_second": 8490.833
    },
    {
      "epoch": 1.8731483975222192,
      "grad_norm": 0.9524179857946982,
      "learning_rate": 9.90361299611109e-08,
      "loss": 0.2904958248138428,
      "num_input_tokens_seen": 455601424,
      "step": 27820,
      "train_runtime": 53657.9443,
      "train_tokens_per_second": 8490.848
    },
    {
      "epoch": 1.8734850525181792,
      "grad_norm": 1.2698151664127268,
      "learning_rate": 9.851316597681959e-08,
      "loss": 0.30260000228881834,
      "num_input_tokens_seen": 455683344,
      "step": 27825,
      "train_runtime": 53667.5091,
      "train_tokens_per_second": 8490.861
    },
    {
      "epoch": 1.8738217075141395,
      "grad_norm": 1.2510387790105777,
      "learning_rate": 9.799157267769088e-08,
      "loss": 0.30551977157592775,
      "num_input_tokens_seen": 455765264,
      "step": 27830,
      "train_runtime": 53677.0737,
      "train_tokens_per_second": 8490.874
    },
    {
      "epoch": 1.8741583625100997,
      "grad_norm": 1.1625298132972532,
      "learning_rate": 9.747135020958753e-08,
      "loss": 0.27402434349060056,
      "num_input_tokens_seen": 455847184,
      "step": 27835,
      "train_runtime": 53686.6862,
      "train_tokens_per_second": 8490.88
    },
    {
      "epoch": 1.8744950175060597,
      "grad_norm": 1.1372163414264855,
      "learning_rate": 9.695249871798818e-08,
      "loss": 0.2728546142578125,
      "num_input_tokens_seen": 455929104,
      "step": 27840,
      "train_runtime": 53696.3026,
      "train_tokens_per_second": 8490.885
    },
    {
      "epoch": 1.8748316725020198,
      "grad_norm": 1.394712698982084,
      "learning_rate": 9.643501834798786e-08,
      "loss": 0.28392205238342283,
      "num_input_tokens_seen": 456011024,
      "step": 27845,
      "train_runtime": 53705.9581,
      "train_tokens_per_second": 8490.883
    },
    {
      "epoch": 1.8751683274979802,
      "grad_norm": 1.1887083718527394,
      "learning_rate": 9.591890924429858e-08,
      "loss": 0.27073974609375,
      "num_input_tokens_seen": 456092944,
      "step": 27850,
      "train_runtime": 53715.5734,
      "train_tokens_per_second": 8490.888
    },
    {
      "epoch": 1.8755049824939403,
      "grad_norm": 1.059510574296953,
      "learning_rate": 9.540417155124826e-08,
      "loss": 0.269240140914917,
      "num_input_tokens_seen": 456174864,
      "step": 27855,
      "train_runtime": 53725.0799,
      "train_tokens_per_second": 8490.911
    },
    {
      "epoch": 1.8758416374899003,
      "grad_norm": 0.9661160675121866,
      "learning_rate": 9.489080541278117e-08,
      "loss": 0.2680685043334961,
      "num_input_tokens_seen": 456256784,
      "step": 27860,
      "train_runtime": 53734.5891,
      "train_tokens_per_second": 8490.933
    },
    {
      "epoch": 1.8761782924858605,
      "grad_norm": 1.0732021410833805,
      "learning_rate": 9.437881097246026e-08,
      "loss": 0.3041705131530762,
      "num_input_tokens_seen": 456338704,
      "step": 27865,
      "train_runtime": 53744.0913,
      "train_tokens_per_second": 8490.956
    },
    {
      "epoch": 1.8765149474818208,
      "grad_norm": 1.2267654605281453,
      "learning_rate": 9.3868188373461e-08,
      "loss": 0.28890142440795896,
      "num_input_tokens_seen": 456420624,
      "step": 27870,
      "train_runtime": 53753.6004,
      "train_tokens_per_second": 8490.978
    },
    {
      "epoch": 1.8768516024777808,
      "grad_norm": 1.1155996693247008,
      "learning_rate": 9.335893775857918e-08,
      "loss": 0.28839764595031736,
      "num_input_tokens_seen": 456502544,
      "step": 27875,
      "train_runtime": 53763.2411,
      "train_tokens_per_second": 8490.979
    },
    {
      "epoch": 1.8771882574737409,
      "grad_norm": 1.0235363982100751,
      "learning_rate": 9.285105927022309e-08,
      "loss": 0.27278838157653806,
      "num_input_tokens_seen": 456584464,
      "step": 27880,
      "train_runtime": 53772.7415,
      "train_tokens_per_second": 8491.002
    },
    {
      "epoch": 1.877524912469701,
      "grad_norm": 1.0018233231450766,
      "learning_rate": 9.234455305042133e-08,
      "loss": 0.2668973445892334,
      "num_input_tokens_seen": 456666184,
      "step": 27885,
      "train_runtime": 53782.856,
      "train_tokens_per_second": 8490.925
    },
    {
      "epoch": 1.8778615674656614,
      "grad_norm": 1.1167459774503625,
      "learning_rate": 9.183941924081563e-08,
      "loss": 0.2747013568878174,
      "num_input_tokens_seen": 456748104,
      "step": 27890,
      "train_runtime": 53792.589,
      "train_tokens_per_second": 8490.911
    },
    {
      "epoch": 1.8781982224616214,
      "grad_norm": 1.005514052314062,
      "learning_rate": 9.133565798266574e-08,
      "loss": 0.3027233123779297,
      "num_input_tokens_seen": 456830024,
      "step": 27895,
      "train_runtime": 53802.0887,
      "train_tokens_per_second": 8490.935
    },
    {
      "epoch": 1.8785348774575814,
      "grad_norm": 1.3076226077589523,
      "learning_rate": 9.083326941684622e-08,
      "loss": 0.3143908500671387,
      "num_input_tokens_seen": 456911944,
      "step": 27900,
      "train_runtime": 53811.6707,
      "train_tokens_per_second": 8490.945
    },
    {
      "epoch": 1.8788715324535417,
      "grad_norm": 1.0215598512907487,
      "learning_rate": 9.033225368384912e-08,
      "loss": 0.29085168838500974,
      "num_input_tokens_seen": 456993864,
      "step": 27905,
      "train_runtime": 53821.2857,
      "train_tokens_per_second": 8490.95
    },
    {
      "epoch": 1.879208187449502,
      "grad_norm": 1.0375604367024034,
      "learning_rate": 8.98326109237807e-08,
      "loss": 0.2583410978317261,
      "num_input_tokens_seen": 457075784,
      "step": 27910,
      "train_runtime": 53830.8842,
      "train_tokens_per_second": 8490.958
    },
    {
      "epoch": 1.879544842445462,
      "grad_norm": 1.2363958504096308,
      "learning_rate": 8.933434127636586e-08,
      "loss": 0.3034812927246094,
      "num_input_tokens_seen": 457157464,
      "step": 27915,
      "train_runtime": 53840.9811,
      "train_tokens_per_second": 8490.883
    },
    {
      "epoch": 1.879881497441422,
      "grad_norm": 1.0745681190143137,
      "learning_rate": 8.883744488094314e-08,
      "loss": 0.28470730781555176,
      "num_input_tokens_seen": 457239384,
      "step": 27920,
      "train_runtime": 53850.4842,
      "train_tokens_per_second": 8490.906
    },
    {
      "epoch": 1.8802181524373822,
      "grad_norm": 0.9864163398682039,
      "learning_rate": 8.834192187646917e-08,
      "loss": 0.27034156322479247,
      "num_input_tokens_seen": 457321304,
      "step": 27925,
      "train_runtime": 53860.1085,
      "train_tokens_per_second": 8490.909
    },
    {
      "epoch": 1.8805548074333425,
      "grad_norm": 1.0950132839909519,
      "learning_rate": 8.784777240151365e-08,
      "loss": 0.2779674053192139,
      "num_input_tokens_seen": 457403224,
      "step": 27930,
      "train_runtime": 53869.7404,
      "train_tokens_per_second": 8490.912
    },
    {
      "epoch": 1.8808914624293025,
      "grad_norm": 1.0752860348143634,
      "learning_rate": 8.73549965942655e-08,
      "loss": 0.27611875534057617,
      "num_input_tokens_seen": 457485144,
      "step": 27935,
      "train_runtime": 53879.3404,
      "train_tokens_per_second": 8490.92
    },
    {
      "epoch": 1.8812281174252625,
      "grad_norm": 1.0596154864660179,
      "learning_rate": 8.686359459252725e-08,
      "loss": 0.2720787525177002,
      "num_input_tokens_seen": 457567064,
      "step": 27940,
      "train_runtime": 53888.8426,
      "train_tokens_per_second": 8490.942
    },
    {
      "epoch": 1.8815647724212228,
      "grad_norm": 1.0788597529346746,
      "learning_rate": 8.637356653371787e-08,
      "loss": 0.29347991943359375,
      "num_input_tokens_seen": 457648984,
      "step": 27945,
      "train_runtime": 53898.4567,
      "train_tokens_per_second": 8490.948
    },
    {
      "epoch": 1.881901427417183,
      "grad_norm": 1.0513117940484433,
      "learning_rate": 8.588491255487163e-08,
      "loss": 0.24905681610107422,
      "num_input_tokens_seen": 457730904,
      "step": 27950,
      "train_runtime": 53908.036,
      "train_tokens_per_second": 8490.959
    },
    {
      "epoch": 1.882238082413143,
      "grad_norm": 1.0406694518879436,
      "learning_rate": 8.539763279264035e-08,
      "loss": 0.26647510528564455,
      "num_input_tokens_seen": 457812824,
      "step": 27955,
      "train_runtime": 53917.6479,
      "train_tokens_per_second": 8490.964
    },
    {
      "epoch": 1.882574737409103,
      "grad_norm": 0.9914483469929137,
      "learning_rate": 8.49117273832889e-08,
      "loss": 0.2958852767944336,
      "num_input_tokens_seen": 457893808,
      "step": 27960,
      "train_runtime": 53927.3189,
      "train_tokens_per_second": 8490.943
    },
    {
      "epoch": 1.8829113924050633,
      "grad_norm": 1.2156841702242733,
      "learning_rate": 8.442719646270082e-08,
      "loss": 0.27719597816467284,
      "num_input_tokens_seen": 457975728,
      "step": 27965,
      "train_runtime": 53936.9432,
      "train_tokens_per_second": 8490.947
    },
    {
      "epoch": 1.8832480474010236,
      "grad_norm": 1.2046145113852895,
      "learning_rate": 8.394404016637159e-08,
      "loss": 0.25183706283569335,
      "num_input_tokens_seen": 458057648,
      "step": 27970,
      "train_runtime": 53946.5525,
      "train_tokens_per_second": 8490.953
    },
    {
      "epoch": 1.8835847023969836,
      "grad_norm": 1.106156577418649,
      "learning_rate": 8.346225862941592e-08,
      "loss": 0.2844752311706543,
      "num_input_tokens_seen": 458139568,
      "step": 27975,
      "train_runtime": 53956.1619,
      "train_tokens_per_second": 8490.959
    },
    {
      "epoch": 1.8839213573929436,
      "grad_norm": 1.2426453186438544,
      "learning_rate": 8.298185198656161e-08,
      "loss": 0.28389773368835447,
      "num_input_tokens_seen": 458221488,
      "step": 27980,
      "train_runtime": 53965.6865,
      "train_tokens_per_second": 8490.979
    },
    {
      "epoch": 1.884258012388904,
      "grad_norm": 1.1616888770116176,
      "learning_rate": 8.25028203721534e-08,
      "loss": 0.24570350646972655,
      "num_input_tokens_seen": 458303408,
      "step": 27985,
      "train_runtime": 53975.3121,
      "train_tokens_per_second": 8490.982
    },
    {
      "epoch": 1.8845946673848641,
      "grad_norm": 1.1715804193609713,
      "learning_rate": 8.202516392015081e-08,
      "loss": 0.2787142753601074,
      "num_input_tokens_seen": 458385328,
      "step": 27990,
      "train_runtime": 53984.9413,
      "train_tokens_per_second": 8490.985
    },
    {
      "epoch": 1.8849313223808242,
      "grad_norm": 1.1609629278442448,
      "learning_rate": 8.154888276412865e-08,
      "loss": 0.2612344264984131,
      "num_input_tokens_seen": 458467248,
      "step": 27995,
      "train_runtime": 53994.5524,
      "train_tokens_per_second": 8490.991
    },
    {
      "epoch": 1.8852679773767842,
      "grad_norm": 1.13622445276448,
      "learning_rate": 8.10739770372776e-08,
      "loss": 0.2582178831100464,
      "num_input_tokens_seen": 458549168,
      "step": 28000,
      "train_runtime": 54004.06,
      "train_tokens_per_second": 8491.013
    },
    {
      "epoch": 1.8856046323727444,
      "grad_norm": 1.0627308776860567,
      "learning_rate": 8.060044687240365e-08,
      "loss": 0.28805837631225584,
      "num_input_tokens_seen": 458631088,
      "step": 28005,
      "train_runtime": 54013.6529,
      "train_tokens_per_second": 8491.022
    },
    {
      "epoch": 1.8859412873687047,
      "grad_norm": 0.9500042486262122,
      "learning_rate": 8.012829240192754e-08,
      "loss": 0.2683521270751953,
      "num_input_tokens_seen": 458713008,
      "step": 28010,
      "train_runtime": 54023.231,
      "train_tokens_per_second": 8491.032
    },
    {
      "epoch": 1.8862779423646647,
      "grad_norm": 1.1894266209039643,
      "learning_rate": 7.9657513757887e-08,
      "loss": 0.2749659299850464,
      "num_input_tokens_seen": 458794928,
      "step": 28015,
      "train_runtime": 54032.8442,
      "train_tokens_per_second": 8491.038
    },
    {
      "epoch": 1.8866145973606248,
      "grad_norm": 1.1757946532983066,
      "learning_rate": 7.918811107193225e-08,
      "loss": 0.28537869453430176,
      "num_input_tokens_seen": 458876848,
      "step": 28020,
      "train_runtime": 54042.3408,
      "train_tokens_per_second": 8491.062
    },
    {
      "epoch": 1.886951252356585,
      "grad_norm": 1.254831424713069,
      "learning_rate": 7.872008447533109e-08,
      "loss": 0.30893783569335936,
      "num_input_tokens_seen": 458958768,
      "step": 28025,
      "train_runtime": 54051.9641,
      "train_tokens_per_second": 8491.066
    },
    {
      "epoch": 1.8872879073525453,
      "grad_norm": 1.1038536675361779,
      "learning_rate": 7.825343409896547e-08,
      "loss": 0.2977119445800781,
      "num_input_tokens_seen": 459039880,
      "step": 28030,
      "train_runtime": 54061.7642,
      "train_tokens_per_second": 8491.027
    },
    {
      "epoch": 1.8876245623485053,
      "grad_norm": 1.1408923670114228,
      "learning_rate": 7.778816007333268e-08,
      "loss": 0.27144105434417726,
      "num_input_tokens_seen": 459121800,
      "step": 28035,
      "train_runtime": 54071.3726,
      "train_tokens_per_second": 8491.033
    },
    {
      "epoch": 1.8879612173444653,
      "grad_norm": 1.1537437305598441,
      "learning_rate": 7.73242625285453e-08,
      "loss": 0.2768087863922119,
      "num_input_tokens_seen": 459202968,
      "step": 28040,
      "train_runtime": 54081.0852,
      "train_tokens_per_second": 8491.009
    },
    {
      "epoch": 1.8882978723404256,
      "grad_norm": 1.0274398674663405,
      "learning_rate": 7.686174159433069e-08,
      "loss": 0.30141339302062986,
      "num_input_tokens_seen": 459284888,
      "step": 28045,
      "train_runtime": 54090.7101,
      "train_tokens_per_second": 8491.012
    },
    {
      "epoch": 1.8886345273363858,
      "grad_norm": 1.0306769509438516,
      "learning_rate": 7.640059740003091e-08,
      "loss": 0.2704217195510864,
      "num_input_tokens_seen": 459366808,
      "step": 28050,
      "train_runtime": 54100.2443,
      "train_tokens_per_second": 8491.03
    },
    {
      "epoch": 1.8889711823323458,
      "grad_norm": 1.2976890249898017,
      "learning_rate": 7.594083007460396e-08,
      "loss": 0.3097728967666626,
      "num_input_tokens_seen": 459448728,
      "step": 28055,
      "train_runtime": 54109.8432,
      "train_tokens_per_second": 8491.038
    },
    {
      "epoch": 1.8893078373283059,
      "grad_norm": 0.8927092890445534,
      "learning_rate": 7.54824397466225e-08,
      "loss": 0.2865484952926636,
      "num_input_tokens_seen": 459530648,
      "step": 28060,
      "train_runtime": 54119.4443,
      "train_tokens_per_second": 8491.045
    },
    {
      "epoch": 1.8896444923242661,
      "grad_norm": 1.0324258668761892,
      "learning_rate": 7.502542654427347e-08,
      "loss": 0.2673775672912598,
      "num_input_tokens_seen": 459612568,
      "step": 28065,
      "train_runtime": 54128.9749,
      "train_tokens_per_second": 8491.064
    },
    {
      "epoch": 1.8899811473202264,
      "grad_norm": 1.1717548367386503,
      "learning_rate": 7.456979059535907e-08,
      "loss": 0.30215404033660886,
      "num_input_tokens_seen": 459694488,
      "step": 28070,
      "train_runtime": 54138.5755,
      "train_tokens_per_second": 8491.071
    },
    {
      "epoch": 1.8903178023161864,
      "grad_norm": 1.0017066636631373,
      "learning_rate": 7.411553202729682e-08,
      "loss": 0.2833854198455811,
      "num_input_tokens_seen": 459775824,
      "step": 28075,
      "train_runtime": 54148.6942,
      "train_tokens_per_second": 8490.986
    },
    {
      "epoch": 1.8906544573121464,
      "grad_norm": 1.4629196303094665,
      "learning_rate": 7.3662650967119e-08,
      "loss": 0.24695222377777098,
      "num_input_tokens_seen": 459857744,
      "step": 28080,
      "train_runtime": 54158.1917,
      "train_tokens_per_second": 8491.01
    },
    {
      "epoch": 1.8909911123081067,
      "grad_norm": 1.12299453669583,
      "learning_rate": 7.321114754147151e-08,
      "loss": 0.2717560052871704,
      "num_input_tokens_seen": 459939664,
      "step": 28085,
      "train_runtime": 54167.8019,
      "train_tokens_per_second": 8491.016
    },
    {
      "epoch": 1.891327767304067,
      "grad_norm": 1.2108126578477447,
      "learning_rate": 7.276102187661615e-08,
      "loss": 0.2810450315475464,
      "num_input_tokens_seen": 460021584,
      "step": 28090,
      "train_runtime": 54177.3066,
      "train_tokens_per_second": 8491.038
    },
    {
      "epoch": 1.891664422300027,
      "grad_norm": 1.1328783679882288,
      "learning_rate": 7.231227409842944e-08,
      "loss": 0.2897559881210327,
      "num_input_tokens_seen": 460103504,
      "step": 28095,
      "train_runtime": 54186.9169,
      "train_tokens_per_second": 8491.044
    },
    {
      "epoch": 1.892001077295987,
      "grad_norm": 1.1661089630913792,
      "learning_rate": 7.186490433240156e-08,
      "loss": 0.27415130138397215,
      "num_input_tokens_seen": 460185424,
      "step": 28100,
      "train_runtime": 54196.4159,
      "train_tokens_per_second": 8491.067
    },
    {
      "epoch": 1.8923377322919472,
      "grad_norm": 1.2094459904804973,
      "learning_rate": 7.14189127036391e-08,
      "loss": 0.28289358615875243,
      "num_input_tokens_seen": 460267344,
      "step": 28105,
      "train_runtime": 54206.0406,
      "train_tokens_per_second": 8491.071
    },
    {
      "epoch": 1.8926743872879075,
      "grad_norm": 0.93867427167921,
      "learning_rate": 7.097429933686118e-08,
      "loss": 0.28035855293273926,
      "num_input_tokens_seen": 460349264,
      "step": 28110,
      "train_runtime": 54215.5818,
      "train_tokens_per_second": 8491.088
    },
    {
      "epoch": 1.8930110422838675,
      "grad_norm": 1.3819138227579342,
      "learning_rate": 7.053106435640333e-08,
      "loss": 0.30639970302581787,
      "num_input_tokens_seen": 460431184,
      "step": 28115,
      "train_runtime": 54225.1685,
      "train_tokens_per_second": 8491.097
    },
    {
      "epoch": 1.8933476972798275,
      "grad_norm": 1.183297132736644,
      "learning_rate": 7.008920788621366e-08,
      "loss": 0.2886078357696533,
      "num_input_tokens_seen": 460513104,
      "step": 28120,
      "train_runtime": 54234.7214,
      "train_tokens_per_second": 8491.112
    },
    {
      "epoch": 1.8936843522757878,
      "grad_norm": 1.039538105417139,
      "learning_rate": 6.964873004985717e-08,
      "loss": 0.27767484188079833,
      "num_input_tokens_seen": 460595024,
      "step": 28125,
      "train_runtime": 54244.4015,
      "train_tokens_per_second": 8491.107
    },
    {
      "epoch": 1.894021007271748,
      "grad_norm": 1.1969821672349683,
      "learning_rate": 6.920963097051148e-08,
      "loss": 0.277606201171875,
      "num_input_tokens_seen": 460676944,
      "step": 28130,
      "train_runtime": 54254.0875,
      "train_tokens_per_second": 8491.101
    },
    {
      "epoch": 1.894357662267708,
      "grad_norm": 1.318399780889487,
      "learning_rate": 6.877191077096945e-08,
      "loss": 0.2978498458862305,
      "num_input_tokens_seen": 460758864,
      "step": 28135,
      "train_runtime": 54263.5963,
      "train_tokens_per_second": 8491.123
    },
    {
      "epoch": 1.894694317263668,
      "grad_norm": 1.0787992021824082,
      "learning_rate": 6.833556957363818e-08,
      "loss": 0.29288158416748045,
      "num_input_tokens_seen": 460840784,
      "step": 28140,
      "train_runtime": 54273.4692,
      "train_tokens_per_second": 8491.088
    },
    {
      "epoch": 1.8950309722596284,
      "grad_norm": 1.0187704989388429,
      "learning_rate": 6.790060750053895e-08,
      "loss": 0.2675026893615723,
      "num_input_tokens_seen": 460921896,
      "step": 28145,
      "train_runtime": 54283.1924,
      "train_tokens_per_second": 8491.061
    },
    {
      "epoch": 1.8953676272555886,
      "grad_norm": 1.1197058145080006,
      "learning_rate": 6.746702467330723e-08,
      "loss": 0.25696959495544436,
      "num_input_tokens_seen": 461003816,
      "step": 28150,
      "train_runtime": 54292.8853,
      "train_tokens_per_second": 8491.054
    },
    {
      "epoch": 1.8957042822515486,
      "grad_norm": 0.9490884826730286,
      "learning_rate": 6.703482121319383e-08,
      "loss": 0.2868093967437744,
      "num_input_tokens_seen": 461085736,
      "step": 28155,
      "train_runtime": 54302.7486,
      "train_tokens_per_second": 8491.02
    },
    {
      "epoch": 1.8960409372475087,
      "grad_norm": 1.2713546664369275,
      "learning_rate": 6.660399724106204e-08,
      "loss": 0.2908592700958252,
      "num_input_tokens_seen": 461167656,
      "step": 28160,
      "train_runtime": 54312.2473,
      "train_tokens_per_second": 8491.044
    },
    {
      "epoch": 1.896377592243469,
      "grad_norm": 1.1374703517845488,
      "learning_rate": 6.617455287739161e-08,
      "loss": 0.2790527820587158,
      "num_input_tokens_seen": 461249536,
      "step": 28165,
      "train_runtime": 54322.2973,
      "train_tokens_per_second": 8490.98
    },
    {
      "epoch": 1.8967142472394292,
      "grad_norm": 1.3956757437267218,
      "learning_rate": 6.574648824227369e-08,
      "loss": 0.270804238319397,
      "num_input_tokens_seen": 461331456,
      "step": 28170,
      "train_runtime": 54331.802,
      "train_tokens_per_second": 8491.002
    },
    {
      "epoch": 1.8970509022353892,
      "grad_norm": 1.0010092837887639,
      "learning_rate": 6.5319803455417e-08,
      "loss": 0.2813549041748047,
      "num_input_tokens_seen": 461413376,
      "step": 28175,
      "train_runtime": 54341.3478,
      "train_tokens_per_second": 8491.018
    },
    {
      "epoch": 1.8973875572313492,
      "grad_norm": 1.2308557077597104,
      "learning_rate": 6.489449863614106e-08,
      "loss": 0.2947617292404175,
      "num_input_tokens_seen": 461495296,
      "step": 28180,
      "train_runtime": 54350.8477,
      "train_tokens_per_second": 8491.041
    },
    {
      "epoch": 1.8977242122273095,
      "grad_norm": 1.3604055829561257,
      "learning_rate": 6.44705739033824e-08,
      "loss": 0.27008814811706544,
      "num_input_tokens_seen": 461576816,
      "step": 28185,
      "train_runtime": 54360.7031,
      "train_tokens_per_second": 8491.002
    },
    {
      "epoch": 1.8980608672232697,
      "grad_norm": 1.1351299119255913,
      "learning_rate": 6.404802937568899e-08,
      "loss": 0.2749145030975342,
      "num_input_tokens_seen": 461658736,
      "step": 28190,
      "train_runtime": 54370.2054,
      "train_tokens_per_second": 8491.024
    },
    {
      "epoch": 1.8983975222192297,
      "grad_norm": 0.9580744875693412,
      "learning_rate": 6.362686517122463e-08,
      "loss": 0.2791230916976929,
      "num_input_tokens_seen": 461740656,
      "step": 28195,
      "train_runtime": 54379.6972,
      "train_tokens_per_second": 8491.049
    },
    {
      "epoch": 1.8987341772151898,
      "grad_norm": 1.1785982966945485,
      "learning_rate": 6.320708140776676e-08,
      "loss": 0.26555333137512205,
      "num_input_tokens_seen": 461822120,
      "step": 28200,
      "train_runtime": 54389.1989,
      "train_tokens_per_second": 8491.063
    },
    {
      "epoch": 1.89907083221115,
      "grad_norm": 1.1744411633695009,
      "learning_rate": 6.278867820270596e-08,
      "loss": 0.28367593288421633,
      "num_input_tokens_seen": 461904040,
      "step": 28205,
      "train_runtime": 54398.7017,
      "train_tokens_per_second": 8491.086
    },
    {
      "epoch": 1.8994074872071103,
      "grad_norm": 1.2215792247470432,
      "learning_rate": 6.237165567304749e-08,
      "loss": 0.2732261896133423,
      "num_input_tokens_seen": 461985960,
      "step": 28210,
      "train_runtime": 54408.1991,
      "train_tokens_per_second": 8491.109
    },
    {
      "epoch": 1.8997441422030703,
      "grad_norm": 1.1027121827347635,
      "learning_rate": 6.19560139354114e-08,
      "loss": 0.2850240468978882,
      "num_input_tokens_seen": 462067880,
      "step": 28215,
      "train_runtime": 54417.7028,
      "train_tokens_per_second": 8491.132
    },
    {
      "epoch": 1.9000807971990303,
      "grad_norm": 1.1092657414221776,
      "learning_rate": 6.154175310602916e-08,
      "loss": 0.2869659662246704,
      "num_input_tokens_seen": 462149800,
      "step": 28220,
      "train_runtime": 54427.1999,
      "train_tokens_per_second": 8491.155
    },
    {
      "epoch": 1.9004174521949906,
      "grad_norm": 0.9982233885226593,
      "learning_rate": 6.112887330074812e-08,
      "loss": 0.2560394048690796,
      "num_input_tokens_seen": 462230440,
      "step": 28225,
      "train_runtime": 54436.9055,
      "train_tokens_per_second": 8491.123
    },
    {
      "epoch": 1.9007541071909508,
      "grad_norm": 1.1448084867608999,
      "learning_rate": 6.07173746350287e-08,
      "loss": 0.26772949695587156,
      "num_input_tokens_seen": 462312360,
      "step": 28230,
      "train_runtime": 54446.5358,
      "train_tokens_per_second": 8491.125
    },
    {
      "epoch": 1.9010907621869109,
      "grad_norm": 1.283174104194705,
      "learning_rate": 6.030725722394548e-08,
      "loss": 0.28871583938598633,
      "num_input_tokens_seen": 462394280,
      "step": 28235,
      "train_runtime": 54456.2226,
      "train_tokens_per_second": 8491.119
    },
    {
      "epoch": 1.901427417182871,
      "grad_norm": 1.2358384103753886,
      "learning_rate": 5.98985211821862e-08,
      "loss": 0.2933696746826172,
      "num_input_tokens_seen": 462476200,
      "step": 28240,
      "train_runtime": 54465.906,
      "train_tokens_per_second": 8491.114
    },
    {
      "epoch": 1.9017640721788311,
      "grad_norm": 1.7676665695852738,
      "learning_rate": 5.9491166624052185e-08,
      "loss": 0.28604094982147216,
      "num_input_tokens_seen": 462558120,
      "step": 28245,
      "train_runtime": 54475.6042,
      "train_tokens_per_second": 8491.106
    },
    {
      "epoch": 1.9021007271747914,
      "grad_norm": 1.7199228086770877,
      "learning_rate": 5.908519366345955e-08,
      "loss": 0.2909108877182007,
      "num_input_tokens_seen": 462639720,
      "step": 28250,
      "train_runtime": 54485.747,
      "train_tokens_per_second": 8491.023
    },
    {
      "epoch": 1.9024373821707514,
      "grad_norm": 1.2788804296100769,
      "learning_rate": 5.868060241393747e-08,
      "loss": 0.28171658515930176,
      "num_input_tokens_seen": 462721640,
      "step": 28255,
      "train_runtime": 54495.425,
      "train_tokens_per_second": 8491.018
    },
    {
      "epoch": 1.9027740371667115,
      "grad_norm": 1.2206897406396076,
      "learning_rate": 5.8277392988627665e-08,
      "loss": 0.2777735948562622,
      "num_input_tokens_seen": 462803560,
      "step": 28260,
      "train_runtime": 54505.096,
      "train_tokens_per_second": 8491.014
    },
    {
      "epoch": 1.9031106921626717,
      "grad_norm": 0.9804343801703667,
      "learning_rate": 5.787556550028772e-08,
      "loss": 0.27858448028564453,
      "num_input_tokens_seen": 462885480,
      "step": 28265,
      "train_runtime": 54514.7762,
      "train_tokens_per_second": 8491.009
    },
    {
      "epoch": 1.903447347158632,
      "grad_norm": 1.2417472458208023,
      "learning_rate": 5.747512006128553e-08,
      "loss": 0.3124921083450317,
      "num_input_tokens_seen": 462967400,
      "step": 28270,
      "train_runtime": 54524.4681,
      "train_tokens_per_second": 8491.003
    },
    {
      "epoch": 1.903784002154592,
      "grad_norm": 1.2405376373570725,
      "learning_rate": 5.707605678360595e-08,
      "loss": 0.29537644386291506,
      "num_input_tokens_seen": 463049320,
      "step": 28275,
      "train_runtime": 54534.1274,
      "train_tokens_per_second": 8491.001
    },
    {
      "epoch": 1.904120657150552,
      "grad_norm": 1.0283073366443793,
      "learning_rate": 5.6678375778845276e-08,
      "loss": 0.2846913576126099,
      "num_input_tokens_seen": 463131240,
      "step": 28280,
      "train_runtime": 54543.6193,
      "train_tokens_per_second": 8491.025
    },
    {
      "epoch": 1.9044573121465123,
      "grad_norm": 1.2362295277912687,
      "learning_rate": 5.6282077158213435e-08,
      "loss": 0.2613241195678711,
      "num_input_tokens_seen": 463213160,
      "step": 28285,
      "train_runtime": 54553.2645,
      "train_tokens_per_second": 8491.026
    },
    {
      "epoch": 1.9047939671424725,
      "grad_norm": 1.5279623977915384,
      "learning_rate": 5.588716103253511e-08,
      "loss": 0.3043036460876465,
      "num_input_tokens_seen": 463295080,
      "step": 28290,
      "train_runtime": 54562.7684,
      "train_tokens_per_second": 8491.048
    },
    {
      "epoch": 1.9051306221384325,
      "grad_norm": 1.2331831133371096,
      "learning_rate": 5.549362751224585e-08,
      "loss": 0.30846776962280276,
      "num_input_tokens_seen": 463376960,
      "step": 28295,
      "train_runtime": 54572.9604,
      "train_tokens_per_second": 8490.962
    },
    {
      "epoch": 1.9054672771343926,
      "grad_norm": 1.109519290841835,
      "learning_rate": 5.5101476707397075e-08,
      "loss": 0.27219903469085693,
      "num_input_tokens_seen": 463458880,
      "step": 28300,
      "train_runtime": 54582.646,
      "train_tokens_per_second": 8490.957
    },
    {
      "epoch": 1.9058039321303528,
      "grad_norm": 1.150743675907023,
      "learning_rate": 5.471070872765216e-08,
      "loss": 0.29206027984619143,
      "num_input_tokens_seen": 463540800,
      "step": 28305,
      "train_runtime": 54592.149,
      "train_tokens_per_second": 8490.979
    },
    {
      "epoch": 1.906140587126313,
      "grad_norm": 1.089162340552151,
      "learning_rate": 5.432132368228815e-08,
      "loss": 0.30026557445526125,
      "num_input_tokens_seen": 463622720,
      "step": 28310,
      "train_runtime": 54601.6437,
      "train_tokens_per_second": 8491.003
    },
    {
      "epoch": 1.906477242122273,
      "grad_norm": 1.331444115112811,
      "learning_rate": 5.393332168019572e-08,
      "loss": 0.2997517824172974,
      "num_input_tokens_seen": 463704640,
      "step": 28315,
      "train_runtime": 54611.1615,
      "train_tokens_per_second": 8491.023
    },
    {
      "epoch": 1.9068138971182331,
      "grad_norm": 0.9387489875619068,
      "learning_rate": 5.354670282987695e-08,
      "loss": 0.29670684337615966,
      "num_input_tokens_seen": 463786560,
      "step": 28320,
      "train_runtime": 54620.8716,
      "train_tokens_per_second": 8491.014
    },
    {
      "epoch": 1.9071505521141934,
      "grad_norm": 1.3848453890120602,
      "learning_rate": 5.316146723945037e-08,
      "loss": 0.31865606307983396,
      "num_input_tokens_seen": 463868480,
      "step": 28325,
      "train_runtime": 54630.5478,
      "train_tokens_per_second": 8491.009
    },
    {
      "epoch": 1.9074872071101536,
      "grad_norm": 0.9875857832070184,
      "learning_rate": 5.277761501664535e-08,
      "loss": 0.2548397064208984,
      "num_input_tokens_seen": 463950400,
      "step": 28330,
      "train_runtime": 54640.0611,
      "train_tokens_per_second": 8491.03
    },
    {
      "epoch": 1.9078238621061137,
      "grad_norm": 1.2647964456138425,
      "learning_rate": 5.2395146268804934e-08,
      "loss": 0.30023233890533446,
      "num_input_tokens_seen": 464032320,
      "step": 28335,
      "train_runtime": 54649.571,
      "train_tokens_per_second": 8491.051
    },
    {
      "epoch": 1.9081605171020737,
      "grad_norm": 1.2281900419381895,
      "learning_rate": 5.201406110288465e-08,
      "loss": 0.26867079734802246,
      "num_input_tokens_seen": 464114240,
      "step": 28340,
      "train_runtime": 54659.0737,
      "train_tokens_per_second": 8491.074
    },
    {
      "epoch": 1.908497172098034,
      "grad_norm": 1.1690246133318847,
      "learning_rate": 5.1634359625454266e-08,
      "loss": 0.23925013542175294,
      "num_input_tokens_seen": 464196160,
      "step": 28345,
      "train_runtime": 54668.5838,
      "train_tokens_per_second": 8491.095
    },
    {
      "epoch": 1.9088338270939942,
      "grad_norm": 1.2629404372763833,
      "learning_rate": 5.125604194269606e-08,
      "loss": 0.3090235471725464,
      "num_input_tokens_seen": 464278080,
      "step": 28350,
      "train_runtime": 54678.0977,
      "train_tokens_per_second": 8491.116
    },
    {
      "epoch": 1.9091704820899542,
      "grad_norm": 1.3157314152936272,
      "learning_rate": 5.087910816040542e-08,
      "loss": 0.32361364364624023,
      "num_input_tokens_seen": 464360000,
      "step": 28355,
      "train_runtime": 54687.6036,
      "train_tokens_per_second": 8491.138
    },
    {
      "epoch": 1.9095071370859142,
      "grad_norm": 1.1956692989021327,
      "learning_rate": 5.0503558383990235e-08,
      "loss": 0.2813680171966553,
      "num_input_tokens_seen": 464441920,
      "step": 28360,
      "train_runtime": 54697.1118,
      "train_tokens_per_second": 8491.16
    },
    {
      "epoch": 1.9098437920818745,
      "grad_norm": 1.2778633925864897,
      "learning_rate": 5.0129392718472614e-08,
      "loss": 0.3059381008148193,
      "num_input_tokens_seen": 464523840,
      "step": 28365,
      "train_runtime": 54706.6189,
      "train_tokens_per_second": 8491.182
    },
    {
      "epoch": 1.9101804470778347,
      "grad_norm": 0.9905383446639954,
      "learning_rate": 4.9756611268486077e-08,
      "loss": 0.28690781593322756,
      "num_input_tokens_seen": 464605760,
      "step": 28370,
      "train_runtime": 54716.1293,
      "train_tokens_per_second": 8491.203
    },
    {
      "epoch": 1.9105171020737948,
      "grad_norm": 1.1529439015811922,
      "learning_rate": 4.9385214138277795e-08,
      "loss": 0.3206716299057007,
      "num_input_tokens_seen": 464687680,
      "step": 28375,
      "train_runtime": 54725.6415,
      "train_tokens_per_second": 8491.224
    },
    {
      "epoch": 1.9108537570697548,
      "grad_norm": 1.075369913862333,
      "learning_rate": 4.901520143170857e-08,
      "loss": 0.301939582824707,
      "num_input_tokens_seen": 464769600,
      "step": 28380,
      "train_runtime": 54735.1512,
      "train_tokens_per_second": 8491.245
    },
    {
      "epoch": 1.911190412065715,
      "grad_norm": 0.9933848606679463,
      "learning_rate": 4.864657325225064e-08,
      "loss": 0.26233463287353515,
      "num_input_tokens_seen": 464851520,
      "step": 28385,
      "train_runtime": 54744.663,
      "train_tokens_per_second": 8491.266
    },
    {
      "epoch": 1.9115270670616753,
      "grad_norm": 0.9815235884527214,
      "learning_rate": 4.827932970298932e-08,
      "loss": 0.2557141542434692,
      "num_input_tokens_seen": 464933440,
      "step": 28390,
      "train_runtime": 54754.1699,
      "train_tokens_per_second": 8491.288
    },
    {
      "epoch": 1.9118637220576353,
      "grad_norm": 1.0554733722533274,
      "learning_rate": 4.7913470886624125e-08,
      "loss": 0.2641760349273682,
      "num_input_tokens_seen": 465015360,
      "step": 28395,
      "train_runtime": 54763.672,
      "train_tokens_per_second": 8491.311
    },
    {
      "epoch": 1.9122003770535954,
      "grad_norm": 1.2067711020649547,
      "learning_rate": 4.754899690546544e-08,
      "loss": 0.2814201354980469,
      "num_input_tokens_seen": 465097280,
      "step": 28400,
      "train_runtime": 54773.2851,
      "train_tokens_per_second": 8491.316
    },
    {
      "epoch": 1.9125370320495556,
      "grad_norm": 1.2972245385991217,
      "learning_rate": 4.718590786143729e-08,
      "loss": 0.29871549606323244,
      "num_input_tokens_seen": 465179200,
      "step": 28405,
      "train_runtime": 54782.9015,
      "train_tokens_per_second": 8491.321
    },
    {
      "epoch": 1.9128736870455159,
      "grad_norm": 1.1969408016592262,
      "learning_rate": 4.6824203856076776e-08,
      "loss": 0.26626100540161135,
      "num_input_tokens_seen": 465261120,
      "step": 28410,
      "train_runtime": 54792.5401,
      "train_tokens_per_second": 8491.322
    },
    {
      "epoch": 1.9132103420414759,
      "grad_norm": 1.1214207716233087,
      "learning_rate": 4.646388499053356e-08,
      "loss": 0.2568014144897461,
      "num_input_tokens_seen": 465343040,
      "step": 28415,
      "train_runtime": 54802.1788,
      "train_tokens_per_second": 8491.324
    },
    {
      "epoch": 1.913546997037436,
      "grad_norm": 1.1762566789043754,
      "learning_rate": 4.6104951365568696e-08,
      "loss": 0.2669131994247437,
      "num_input_tokens_seen": 465424960,
      "step": 28420,
      "train_runtime": 54811.6826,
      "train_tokens_per_second": 8491.346
    },
    {
      "epoch": 1.9138836520333962,
      "grad_norm": 1.1947167439291444,
      "learning_rate": 4.574740308155801e-08,
      "loss": 0.2845539808273315,
      "num_input_tokens_seen": 465506880,
      "step": 28425,
      "train_runtime": 54821.1954,
      "train_tokens_per_second": 8491.367
    },
    {
      "epoch": 1.9142203070293564,
      "grad_norm": 1.2446127017147575,
      "learning_rate": 4.539124023848762e-08,
      "loss": 0.27053372859954833,
      "num_input_tokens_seen": 465588800,
      "step": 28430,
      "train_runtime": 54830.7141,
      "train_tokens_per_second": 8491.387
    },
    {
      "epoch": 1.9145569620253164,
      "grad_norm": 1.0766577544186886,
      "learning_rate": 4.503646293595787e-08,
      "loss": 0.3139811515808105,
      "num_input_tokens_seen": 465670720,
      "step": 28435,
      "train_runtime": 54840.2224,
      "train_tokens_per_second": 8491.408
    },
    {
      "epoch": 1.9148936170212765,
      "grad_norm": 1.1595174712226006,
      "learning_rate": 4.468307127318105e-08,
      "loss": 0.28277029991149905,
      "num_input_tokens_seen": 465752640,
      "step": 28440,
      "train_runtime": 54849.7517,
      "train_tokens_per_second": 8491.427
    },
    {
      "epoch": 1.9152302720172367,
      "grad_norm": 1.5815151594956316,
      "learning_rate": 4.4331065348982014e-08,
      "loss": 0.2944267511367798,
      "num_input_tokens_seen": 465834560,
      "step": 28445,
      "train_runtime": 54859.387,
      "train_tokens_per_second": 8491.428
    },
    {
      "epoch": 1.915566927013197,
      "grad_norm": 1.209667469248439,
      "learning_rate": 4.3980445261798125e-08,
      "loss": 0.27361724376678465,
      "num_input_tokens_seen": 465916480,
      "step": 28450,
      "train_runtime": 54869.0305,
      "train_tokens_per_second": 8491.429
    },
    {
      "epoch": 1.915903582009157,
      "grad_norm": 1.1070906339681323,
      "learning_rate": 4.363121110967927e-08,
      "loss": 0.2670175313949585,
      "num_input_tokens_seen": 465998400,
      "step": 28455,
      "train_runtime": 54878.5313,
      "train_tokens_per_second": 8491.452
    },
    {
      "epoch": 1.916240237005117,
      "grad_norm": 1.0107706519520236,
      "learning_rate": 4.328336299028735e-08,
      "loss": 0.2385709762573242,
      "num_input_tokens_seen": 466080320,
      "step": 28460,
      "train_runtime": 54888.1426,
      "train_tokens_per_second": 8491.457
    },
    {
      "epoch": 1.9165768920010773,
      "grad_norm": 1.109891333411937,
      "learning_rate": 4.293690100089731e-08,
      "loss": 0.2685267925262451,
      "num_input_tokens_seen": 466162240,
      "step": 28465,
      "train_runtime": 54897.6781,
      "train_tokens_per_second": 8491.475
    },
    {
      "epoch": 1.9169135469970375,
      "grad_norm": 1.4083110054373953,
      "learning_rate": 4.259182523839556e-08,
      "loss": 0.28335726261138916,
      "num_input_tokens_seen": 466243480,
      "step": 28470,
      "train_runtime": 54907.4554,
      "train_tokens_per_second": 8491.442
    },
    {
      "epoch": 1.9172502019929976,
      "grad_norm": 1.235086669853693,
      "learning_rate": 4.224813579928211e-08,
      "loss": 0.3092541456222534,
      "num_input_tokens_seen": 466325400,
      "step": 28475,
      "train_runtime": 54917.0563,
      "train_tokens_per_second": 8491.449
    },
    {
      "epoch": 1.9175868569889576,
      "grad_norm": 1.1610173598396722,
      "learning_rate": 4.190583277966842e-08,
      "loss": 0.2878906488418579,
      "num_input_tokens_seen": 466407320,
      "step": 28480,
      "train_runtime": 54926.5631,
      "train_tokens_per_second": 8491.471
    },
    {
      "epoch": 1.9179235119849178,
      "grad_norm": 1.104121418555774,
      "learning_rate": 4.1564916275278477e-08,
      "loss": 0.27010512351989746,
      "num_input_tokens_seen": 466488728,
      "step": 28485,
      "train_runtime": 54936.4458,
      "train_tokens_per_second": 8491.425
    },
    {
      "epoch": 1.918260166980878,
      "grad_norm": 1.1079156444865457,
      "learning_rate": 4.122538638144824e-08,
      "loss": 0.2685135841369629,
      "num_input_tokens_seen": 466570648,
      "step": 28490,
      "train_runtime": 54946.116,
      "train_tokens_per_second": 8491.422
    },
    {
      "epoch": 1.9185968219768381,
      "grad_norm": 1.1558958742218106,
      "learning_rate": 4.08872431931262e-08,
      "loss": 0.2759772777557373,
      "num_input_tokens_seen": 466652568,
      "step": 28495,
      "train_runtime": 54955.7725,
      "train_tokens_per_second": 8491.42
    },
    {
      "epoch": 1.9189334769727981,
      "grad_norm": 1.254972573400137,
      "learning_rate": 4.055048680487339e-08,
      "loss": 0.2601959228515625,
      "num_input_tokens_seen": 466734488,
      "step": 28500,
      "train_runtime": 54965.4557,
      "train_tokens_per_second": 8491.415
    },
    {
      "epoch": 1.9192701319687584,
      "grad_norm": 1.1372613288847107,
      "learning_rate": 4.0215117310862806e-08,
      "loss": 0.32734317779541017,
      "num_input_tokens_seen": 466815432,
      "step": 28505,
      "train_runtime": 54975.6319,
      "train_tokens_per_second": 8491.315
    },
    {
      "epoch": 1.9196067869647186,
      "grad_norm": 1.2571295055961855,
      "learning_rate": 3.9881134804878316e-08,
      "loss": 0.2761819839477539,
      "num_input_tokens_seen": 466897352,
      "step": 28510,
      "train_runtime": 54985.3323,
      "train_tokens_per_second": 8491.307
    },
    {
      "epoch": 1.9199434419606787,
      "grad_norm": 1.3357246903110338,
      "learning_rate": 3.9548539380318e-08,
      "loss": 0.302858304977417,
      "num_input_tokens_seen": 466979272,
      "step": 28515,
      "train_runtime": 54995.0237,
      "train_tokens_per_second": 8491.3
    },
    {
      "epoch": 1.9202800969566387,
      "grad_norm": 1.152732075357069,
      "learning_rate": 3.921733113019077e-08,
      "loss": 0.2848546028137207,
      "num_input_tokens_seen": 467061192,
      "step": 28520,
      "train_runtime": 55004.7087,
      "train_tokens_per_second": 8491.295
    },
    {
      "epoch": 1.920616751952599,
      "grad_norm": 1.052334922907546,
      "learning_rate": 3.888751014711867e-08,
      "loss": 0.29152746200561525,
      "num_input_tokens_seen": 467143112,
      "step": 28525,
      "train_runtime": 55014.3967,
      "train_tokens_per_second": 8491.288
    },
    {
      "epoch": 1.9209534069485592,
      "grad_norm": 1.1001674687961105,
      "learning_rate": 3.855907652333402e-08,
      "loss": 0.27329463958740235,
      "num_input_tokens_seen": 467225032,
      "step": 28530,
      "train_runtime": 55024.0804,
      "train_tokens_per_second": 8491.283
    },
    {
      "epoch": 1.9212900619445192,
      "grad_norm": 1.2182262066790113,
      "learning_rate": 3.8232030350683344e-08,
      "loss": 0.27783153057098386,
      "num_input_tokens_seen": 467306952,
      "step": 28535,
      "train_runtime": 55033.7774,
      "train_tokens_per_second": 8491.275
    },
    {
      "epoch": 1.9216267169404793,
      "grad_norm": 1.086157681361417,
      "learning_rate": 3.7906371720622946e-08,
      "loss": 0.26311028003692627,
      "num_input_tokens_seen": 467388872,
      "step": 28540,
      "train_runtime": 55043.3588,
      "train_tokens_per_second": 8491.285
    },
    {
      "epoch": 1.9219633719364395,
      "grad_norm": 1.0150492690942599,
      "learning_rate": 3.758210072422275e-08,
      "loss": 0.2619302749633789,
      "num_input_tokens_seen": 467470792,
      "step": 28545,
      "train_runtime": 55052.9745,
      "train_tokens_per_second": 8491.29
    },
    {
      "epoch": 1.9223000269323998,
      "grad_norm": 1.2574510998471955,
      "learning_rate": 3.72592174521641e-08,
      "loss": 0.28816838264465333,
      "num_input_tokens_seen": 467552712,
      "step": 28550,
      "train_runtime": 55062.8214,
      "train_tokens_per_second": 8491.26
    },
    {
      "epoch": 1.9226366819283598,
      "grad_norm": 1.1175358497600962,
      "learning_rate": 3.693772199474033e-08,
      "loss": 0.2837820291519165,
      "num_input_tokens_seen": 467634632,
      "step": 28555,
      "train_runtime": 55072.3266,
      "train_tokens_per_second": 8491.282
    },
    {
      "epoch": 1.9229733369243198,
      "grad_norm": 1.0587556171955357,
      "learning_rate": 3.661761444185674e-08,
      "loss": 0.25052714347839355,
      "num_input_tokens_seen": 467716552,
      "step": 28560,
      "train_runtime": 55081.9516,
      "train_tokens_per_second": 8491.285
    },
    {
      "epoch": 1.92330999192028,
      "grad_norm": 1.1990053222767423,
      "learning_rate": 3.6298894883030066e-08,
      "loss": 0.3037665605545044,
      "num_input_tokens_seen": 467798472,
      "step": 28565,
      "train_runtime": 55091.5955,
      "train_tokens_per_second": 8491.286
    },
    {
      "epoch": 1.9236466469162403,
      "grad_norm": 1.2071879983936786,
      "learning_rate": 3.5981563407389007e-08,
      "loss": 0.29922235012054443,
      "num_input_tokens_seen": 467880392,
      "step": 28570,
      "train_runtime": 55101.1373,
      "train_tokens_per_second": 8491.302
    },
    {
      "epoch": 1.9239833019122003,
      "grad_norm": 1.069665786642348,
      "learning_rate": 3.566562010367536e-08,
      "loss": 0.27229037284851076,
      "num_input_tokens_seen": 467962312,
      "step": 28575,
      "train_runtime": 55110.7894,
      "train_tokens_per_second": 8491.301
    },
    {
      "epoch": 1.9243199569081604,
      "grad_norm": 1.4266237856974324,
      "learning_rate": 3.535106506024011e-08,
      "loss": 0.2777524709701538,
      "num_input_tokens_seen": 468044232,
      "step": 28580,
      "train_runtime": 55120.4304,
      "train_tokens_per_second": 8491.302
    },
    {
      "epoch": 1.9246566119041206,
      "grad_norm": 1.1961918101353803,
      "learning_rate": 3.5037898365049005e-08,
      "loss": 0.24972786903381347,
      "num_input_tokens_seen": 468126152,
      "step": 28585,
      "train_runtime": 55129.9708,
      "train_tokens_per_second": 8491.319
    },
    {
      "epoch": 1.9249932669000809,
      "grad_norm": 1.0449487532518136,
      "learning_rate": 3.472612010567755e-08,
      "loss": 0.2879567384719849,
      "num_input_tokens_seen": 468208072,
      "step": 28590,
      "train_runtime": 55139.5407,
      "train_tokens_per_second": 8491.331
    },
    {
      "epoch": 1.925329921896041,
      "grad_norm": 1.0749680569841524,
      "learning_rate": 3.441573036931323e-08,
      "loss": 0.27460312843322754,
      "num_input_tokens_seen": 468289992,
      "step": 28595,
      "train_runtime": 55149.1547,
      "train_tokens_per_second": 8491.336
    },
    {
      "epoch": 1.925666576892001,
      "grad_norm": 1.0755496122038544,
      "learning_rate": 3.410672924275604e-08,
      "loss": 0.26307001113891604,
      "num_input_tokens_seen": 468371912,
      "step": 28600,
      "train_runtime": 55158.7401,
      "train_tokens_per_second": 8491.345
    },
    {
      "epoch": 1.9260032318879612,
      "grad_norm": 1.256893688590795,
      "learning_rate": 3.3799116812416876e-08,
      "loss": 0.3268022060394287,
      "num_input_tokens_seen": 468453832,
      "step": 28605,
      "train_runtime": 55168.2683,
      "train_tokens_per_second": 8491.364
    },
    {
      "epoch": 1.9263398868839214,
      "grad_norm": 1.2175597476939535,
      "learning_rate": 3.349289316431803e-08,
      "loss": 0.28982784748077395,
      "num_input_tokens_seen": 468535752,
      "step": 28610,
      "train_runtime": 55177.8227,
      "train_tokens_per_second": 8491.378
    },
    {
      "epoch": 1.9266765418798815,
      "grad_norm": 0.9278804347602749,
      "learning_rate": 3.3188058384095446e-08,
      "loss": 0.24699630737304687,
      "num_input_tokens_seen": 468617672,
      "step": 28615,
      "train_runtime": 55187.4304,
      "train_tokens_per_second": 8491.384
    },
    {
      "epoch": 1.9270131968758415,
      "grad_norm": 1.3264769261502707,
      "learning_rate": 3.2884612556993714e-08,
      "loss": 0.2788200378417969,
      "num_input_tokens_seen": 468699592,
      "step": 28620,
      "train_runtime": 55196.9929,
      "train_tokens_per_second": 8491.397
    },
    {
      "epoch": 1.9273498518718017,
      "grad_norm": 1.2449961591878491,
      "learning_rate": 3.258255576787161e-08,
      "loss": 0.28281774520874026,
      "num_input_tokens_seen": 468781512,
      "step": 28625,
      "train_runtime": 55206.5335,
      "train_tokens_per_second": 8491.414
    },
    {
      "epoch": 1.927686506867762,
      "grad_norm": 1.3088033107200754,
      "learning_rate": 3.22818881011977e-08,
      "loss": 0.2840177297592163,
      "num_input_tokens_seen": 468863432,
      "step": 28630,
      "train_runtime": 55216.1028,
      "train_tokens_per_second": 8491.426
    },
    {
      "epoch": 1.928023161863722,
      "grad_norm": 1.139049614384739,
      "learning_rate": 3.198260964105249e-08,
      "loss": 0.2580616235733032,
      "num_input_tokens_seen": 468945352,
      "step": 28635,
      "train_runtime": 55225.7167,
      "train_tokens_per_second": 8491.431
    },
    {
      "epoch": 1.928359816859682,
      "grad_norm": 1.5269888659034323,
      "learning_rate": 3.168472047112903e-08,
      "loss": 0.28037095069885254,
      "num_input_tokens_seen": 469027272,
      "step": 28640,
      "train_runtime": 55235.3389,
      "train_tokens_per_second": 8491.435
    },
    {
      "epoch": 1.9286964718556423,
      "grad_norm": 1.1670626734011653,
      "learning_rate": 3.138822067473068e-08,
      "loss": 0.27768855094909667,
      "num_input_tokens_seen": 469109192,
      "step": 28645,
      "train_runtime": 55244.8527,
      "train_tokens_per_second": 8491.455
    },
    {
      "epoch": 1.9290331268516026,
      "grad_norm": 1.1324031295988692,
      "learning_rate": 3.10931103347728e-08,
      "loss": 0.2924088954925537,
      "num_input_tokens_seen": 469191112,
      "step": 28650,
      "train_runtime": 55254.4751,
      "train_tokens_per_second": 8491.459
    },
    {
      "epoch": 1.9293697818475626,
      "grad_norm": 1.1452173109884018,
      "learning_rate": 3.0799389533781587e-08,
      "loss": 0.314389705657959,
      "num_input_tokens_seen": 469273032,
      "step": 28655,
      "train_runtime": 55264.1391,
      "train_tokens_per_second": 8491.456
    },
    {
      "epoch": 1.9297064368435226,
      "grad_norm": 1.1125918506193657,
      "learning_rate": 3.050705835389578e-08,
      "loss": 0.2544329404830933,
      "num_input_tokens_seen": 469354952,
      "step": 28660,
      "train_runtime": 55273.8652,
      "train_tokens_per_second": 8491.444
    },
    {
      "epoch": 1.9300430918394829,
      "grad_norm": 1.1122876915990636,
      "learning_rate": 3.021611687686443e-08,
      "loss": 0.2763032913208008,
      "num_input_tokens_seen": 469435872,
      "step": 28665,
      "train_runtime": 55283.3826,
      "train_tokens_per_second": 8491.446
    },
    {
      "epoch": 1.9303797468354431,
      "grad_norm": 1.2630740063724977,
      "learning_rate": 2.992656518404857e-08,
      "loss": 0.3142265558242798,
      "num_input_tokens_seen": 469517792,
      "step": 28670,
      "train_runtime": 55292.9988,
      "train_tokens_per_second": 8491.451
    },
    {
      "epoch": 1.9307164018314031,
      "grad_norm": 1.103003520884568,
      "learning_rate": 2.96384033564201e-08,
      "loss": 0.2739395618438721,
      "num_input_tokens_seen": 469599712,
      "step": 28675,
      "train_runtime": 55302.6056,
      "train_tokens_per_second": 8491.457
    },
    {
      "epoch": 1.9310530568273632,
      "grad_norm": 1.1596958867447609,
      "learning_rate": 2.935163147456288e-08,
      "loss": 0.2702032566070557,
      "num_input_tokens_seen": 469681632,
      "step": 28680,
      "train_runtime": 55312.0958,
      "train_tokens_per_second": 8491.481
    },
    {
      "epoch": 1.9313897118233234,
      "grad_norm": 1.0140989402773792,
      "learning_rate": 2.906624961867166e-08,
      "loss": 0.2614999771118164,
      "num_input_tokens_seen": 469763552,
      "step": 28685,
      "train_runtime": 55321.6821,
      "train_tokens_per_second": 8491.491
    },
    {
      "epoch": 1.9317263668192837,
      "grad_norm": 1.079716564479047,
      "learning_rate": 2.8782257868553154e-08,
      "loss": 0.27134683132171633,
      "num_input_tokens_seen": 469845472,
      "step": 28690,
      "train_runtime": 55331.2746,
      "train_tokens_per_second": 8491.499
    },
    {
      "epoch": 1.9320630218152437,
      "grad_norm": 1.0752758357584782,
      "learning_rate": 2.849965630362328e-08,
      "loss": 0.2950048208236694,
      "num_input_tokens_seen": 469927392,
      "step": 28695,
      "train_runtime": 55340.8844,
      "train_tokens_per_second": 8491.505
    },
    {
      "epoch": 1.9323996768112037,
      "grad_norm": 1.1755740243676451,
      "learning_rate": 2.8218445002912153e-08,
      "loss": 0.2792928457260132,
      "num_input_tokens_seen": 470009312,
      "step": 28700,
      "train_runtime": 55350.5193,
      "train_tokens_per_second": 8491.507
    },
    {
      "epoch": 1.932736331807164,
      "grad_norm": 1.1012475709584841,
      "learning_rate": 2.7938624045059094e-08,
      "loss": 0.25181682109832765,
      "num_input_tokens_seen": 470091232,
      "step": 28705,
      "train_runtime": 55360.0583,
      "train_tokens_per_second": 8491.523
    },
    {
      "epoch": 1.9330729868031242,
      "grad_norm": 1.3225003608163617,
      "learning_rate": 2.766019350831428e-08,
      "loss": 0.287015438079834,
      "num_input_tokens_seen": 470173152,
      "step": 28710,
      "train_runtime": 55369.7588,
      "train_tokens_per_second": 8491.515
    },
    {
      "epoch": 1.9334096417990843,
      "grad_norm": 1.2778262790044088,
      "learning_rate": 2.7383153470541546e-08,
      "loss": 0.2764732837677002,
      "num_input_tokens_seen": 470255072,
      "step": 28715,
      "train_runtime": 55379.2641,
      "train_tokens_per_second": 8491.537
    },
    {
      "epoch": 1.9337462967950443,
      "grad_norm": 1.001384170112655,
      "learning_rate": 2.710750400921336e-08,
      "loss": 0.26218609809875487,
      "num_input_tokens_seen": 470336968,
      "step": 28720,
      "train_runtime": 55389.4642,
      "train_tokens_per_second": 8491.452
    },
    {
      "epoch": 1.9340829517910045,
      "grad_norm": 1.0385165067490751,
      "learning_rate": 2.683324520141417e-08,
      "loss": 0.2670066118240356,
      "num_input_tokens_seen": 470418888,
      "step": 28725,
      "train_runtime": 55399.1064,
      "train_tokens_per_second": 8491.453
    },
    {
      "epoch": 1.9344196067869648,
      "grad_norm": 1.0873993510817395,
      "learning_rate": 2.65603771238393e-08,
      "loss": 0.2879924774169922,
      "num_input_tokens_seen": 470500808,
      "step": 28730,
      "train_runtime": 55408.7411,
      "train_tokens_per_second": 8491.455
    },
    {
      "epoch": 1.9347562617829248,
      "grad_norm": 1.170664231361634,
      "learning_rate": 2.6288899852796036e-08,
      "loss": 0.28308007717132566,
      "num_input_tokens_seen": 470582728,
      "step": 28735,
      "train_runtime": 55418.3618,
      "train_tokens_per_second": 8491.459
    },
    {
      "epoch": 1.9350929167788848,
      "grad_norm": 1.1949978373444612,
      "learning_rate": 2.6018813464202543e-08,
      "loss": 0.2826603174209595,
      "num_input_tokens_seen": 470664624,
      "step": 28740,
      "train_runtime": 55427.8615,
      "train_tokens_per_second": 8491.481
    },
    {
      "epoch": 1.935429571774845,
      "grad_norm": 1.1328266177508381,
      "learning_rate": 2.575011803358618e-08,
      "loss": 0.27674098014831544,
      "num_input_tokens_seen": 470746544,
      "step": 28745,
      "train_runtime": 55437.4762,
      "train_tokens_per_second": 8491.486
    },
    {
      "epoch": 1.9357662267708053,
      "grad_norm": 0.9693108337065172,
      "learning_rate": 2.5482813636087955e-08,
      "loss": 0.25964062213897704,
      "num_input_tokens_seen": 470828000,
      "step": 28750,
      "train_runtime": 55447.4252,
      "train_tokens_per_second": 8491.431
    },
    {
      "epoch": 1.9361028817667654,
      "grad_norm": 1.1424191346952883,
      "learning_rate": 2.5216900346458073e-08,
      "loss": 0.2644465446472168,
      "num_input_tokens_seen": 470909920,
      "step": 28755,
      "train_runtime": 55456.9889,
      "train_tokens_per_second": 8491.444
    },
    {
      "epoch": 1.9364395367627254,
      "grad_norm": 1.0466629880136686,
      "learning_rate": 2.4952378239058716e-08,
      "loss": 0.2628424406051636,
      "num_input_tokens_seen": 470991840,
      "step": 28760,
      "train_runtime": 55466.4937,
      "train_tokens_per_second": 8491.466
    },
    {
      "epoch": 1.9367761917586857,
      "grad_norm": 1.158204936236693,
      "learning_rate": 2.4689247387862934e-08,
      "loss": 0.27965245246887205,
      "num_input_tokens_seen": 471073760,
      "step": 28765,
      "train_runtime": 55476.1329,
      "train_tokens_per_second": 8491.467
    },
    {
      "epoch": 1.937112846754646,
      "grad_norm": 0.9691050823575521,
      "learning_rate": 2.4427507866453537e-08,
      "loss": 0.26770570278167727,
      "num_input_tokens_seen": 471155680,
      "step": 28770,
      "train_runtime": 55485.77,
      "train_tokens_per_second": 8491.469
    },
    {
      "epoch": 1.937449501750606,
      "grad_norm": 1.1551112951098998,
      "learning_rate": 2.416715974802586e-08,
      "loss": 0.29672555923461913,
      "num_input_tokens_seen": 471237600,
      "step": 28775,
      "train_runtime": 55495.3057,
      "train_tokens_per_second": 8491.486
    },
    {
      "epoch": 1.937786156746566,
      "grad_norm": 1.008109995306252,
      "learning_rate": 2.3908203105384997e-08,
      "loss": 0.24822678565979003,
      "num_input_tokens_seen": 471319520,
      "step": 28780,
      "train_runtime": 55504.8097,
      "train_tokens_per_second": 8491.508
    },
    {
      "epoch": 1.9381228117425262,
      "grad_norm": 0.9379904135300028,
      "learning_rate": 2.365063801094747e-08,
      "loss": 0.28028130531311035,
      "num_input_tokens_seen": 471401440,
      "step": 28785,
      "train_runtime": 55514.319,
      "train_tokens_per_second": 8491.529
    },
    {
      "epoch": 1.9384594667384865,
      "grad_norm": 1.0498669460869134,
      "learning_rate": 2.3394464536740658e-08,
      "loss": 0.2844208240509033,
      "num_input_tokens_seen": 471483360,
      "step": 28790,
      "train_runtime": 55523.9721,
      "train_tokens_per_second": 8491.528
    },
    {
      "epoch": 1.9387961217344465,
      "grad_norm": 1.2981903949103757,
      "learning_rate": 2.3139682754402816e-08,
      "loss": 0.28132483959197996,
      "num_input_tokens_seen": 471565280,
      "step": 28795,
      "train_runtime": 55533.6035,
      "train_tokens_per_second": 8491.53
    },
    {
      "epoch": 1.9391327767304065,
      "grad_norm": 1.014238939705073,
      "learning_rate": 2.2886292735182502e-08,
      "loss": 0.2593213081359863,
      "num_input_tokens_seen": 471647200,
      "step": 28800,
      "train_runtime": 55543.1509,
      "train_tokens_per_second": 8491.546
    },
    {
      "epoch": 1.9394694317263668,
      "grad_norm": 1.1259268183108198,
      "learning_rate": 2.2634294549939705e-08,
      "loss": 0.269502592086792,
      "num_input_tokens_seen": 471729120,
      "step": 28805,
      "train_runtime": 55552.7226,
      "train_tokens_per_second": 8491.557
    },
    {
      "epoch": 1.939806086722327,
      "grad_norm": 1.2693834769034682,
      "learning_rate": 2.2383688269144723e-08,
      "loss": 0.33671159744262696,
      "num_input_tokens_seen": 471811040,
      "step": 28810,
      "train_runtime": 55562.3309,
      "train_tokens_per_second": 8491.563
    },
    {
      "epoch": 1.940142741718287,
      "grad_norm": 1.1768543525670465,
      "learning_rate": 2.2134473962878712e-08,
      "loss": 0.26990523338317873,
      "num_input_tokens_seen": 471892960,
      "step": 28815,
      "train_runtime": 55571.9483,
      "train_tokens_per_second": 8491.568
    },
    {
      "epoch": 1.940479396714247,
      "grad_norm": 1.37804604771497,
      "learning_rate": 2.1886651700833705e-08,
      "loss": 0.2692497968673706,
      "num_input_tokens_seen": 471974880,
      "step": 28820,
      "train_runtime": 55581.6136,
      "train_tokens_per_second": 8491.565
    },
    {
      "epoch": 1.9408160517102075,
      "grad_norm": 1.0548966791081786,
      "learning_rate": 2.1640221552312603e-08,
      "loss": 0.3004652738571167,
      "num_input_tokens_seen": 472056800,
      "step": 28825,
      "train_runtime": 55591.147,
      "train_tokens_per_second": 8491.582
    },
    {
      "epoch": 1.9411527067061676,
      "grad_norm": 1.05571009237593,
      "learning_rate": 2.1395183586229163e-08,
      "loss": 0.29348387718200686,
      "num_input_tokens_seen": 472138720,
      "step": 28830,
      "train_runtime": 55600.6625,
      "train_tokens_per_second": 8491.602
    },
    {
      "epoch": 1.9414893617021276,
      "grad_norm": 1.314605796346652,
      "learning_rate": 2.1151537871106353e-08,
      "loss": 0.29617109298706057,
      "num_input_tokens_seen": 472219864,
      "step": 28835,
      "train_runtime": 55610.2849,
      "train_tokens_per_second": 8491.592
    },
    {
      "epoch": 1.9418260166980879,
      "grad_norm": 1.1368557041661667,
      "learning_rate": 2.0909284475080226e-08,
      "loss": 0.27585592269897463,
      "num_input_tokens_seen": 472301784,
      "step": 28840,
      "train_runtime": 55619.9957,
      "train_tokens_per_second": 8491.583
    },
    {
      "epoch": 1.942162671694048,
      "grad_norm": 1.1528234724642044,
      "learning_rate": 2.0668423465894928e-08,
      "loss": 0.2776585578918457,
      "num_input_tokens_seen": 472383704,
      "step": 28845,
      "train_runtime": 55629.6816,
      "train_tokens_per_second": 8491.577
    },
    {
      "epoch": 1.9424993266900081,
      "grad_norm": 1.2550871474058019,
      "learning_rate": 2.0428954910907684e-08,
      "loss": 0.27095296382904055,
      "num_input_tokens_seen": 472465624,
      "step": 28850,
      "train_runtime": 55639.3208,
      "train_tokens_per_second": 8491.578
    },
    {
      "epoch": 1.9428359816859682,
      "grad_norm": 1.3509969485077586,
      "learning_rate": 2.019087887708382e-08,
      "loss": 0.30122535228729247,
      "num_input_tokens_seen": 472547544,
      "step": 28855,
      "train_runtime": 55648.9348,
      "train_tokens_per_second": 8491.583
    },
    {
      "epoch": 1.9431726366819284,
      "grad_norm": 1.0786423519691128,
      "learning_rate": 1.9954195431001744e-08,
      "loss": 0.2881383657455444,
      "num_input_tokens_seen": 472629464,
      "step": 28860,
      "train_runtime": 55658.5636,
      "train_tokens_per_second": 8491.586
    },
    {
      "epoch": 1.9435092916778887,
      "grad_norm": 1.1008012393366504,
      "learning_rate": 1.9718904638848513e-08,
      "loss": 0.2590192317962646,
      "num_input_tokens_seen": 472711384,
      "step": 28865,
      "train_runtime": 55668.2241,
      "train_tokens_per_second": 8491.584
    },
    {
      "epoch": 1.9438459466738487,
      "grad_norm": 1.262992305680702,
      "learning_rate": 1.9485006566422602e-08,
      "loss": 0.28837051391601565,
      "num_input_tokens_seen": 472793304,
      "step": 28870,
      "train_runtime": 55677.8915,
      "train_tokens_per_second": 8491.581
    },
    {
      "epoch": 1.9441826016698087,
      "grad_norm": 1.0678046207940153,
      "learning_rate": 1.9252501279132806e-08,
      "loss": 0.26336872577667236,
      "num_input_tokens_seen": 472875224,
      "step": 28875,
      "train_runtime": 55687.5729,
      "train_tokens_per_second": 8491.575
    },
    {
      "epoch": 1.944519256665769,
      "grad_norm": 1.1074800079840958,
      "learning_rate": 1.902138884199878e-08,
      "loss": 0.302351188659668,
      "num_input_tokens_seen": 472957144,
      "step": 28880,
      "train_runtime": 55697.2774,
      "train_tokens_per_second": 8491.567
    },
    {
      "epoch": 1.9448559116617292,
      "grad_norm": 1.048799161612573,
      "learning_rate": 1.879166931964993e-08,
      "loss": 0.26719040870666505,
      "num_input_tokens_seen": 473039064,
      "step": 28885,
      "train_runtime": 55706.7908,
      "train_tokens_per_second": 8491.587
    },
    {
      "epoch": 1.9451925666576892,
      "grad_norm": 1.1133561161976033,
      "learning_rate": 1.856334277632765e-08,
      "loss": 0.2754704236984253,
      "num_input_tokens_seen": 473120984,
      "step": 28890,
      "train_runtime": 55716.3984,
      "train_tokens_per_second": 8491.593
    },
    {
      "epoch": 1.9455292216536493,
      "grad_norm": 1.2350272436127196,
      "learning_rate": 1.8336409275880872e-08,
      "loss": 0.2772572755813599,
      "num_input_tokens_seen": 473202904,
      "step": 28895,
      "train_runtime": 55725.9423,
      "train_tokens_per_second": 8491.609
    },
    {
      "epoch": 1.9458658766496095,
      "grad_norm": 1.2018055624332138,
      "learning_rate": 1.8110868881772713e-08,
      "loss": 0.25763721466064454,
      "num_input_tokens_seen": 473284824,
      "step": 28900,
      "train_runtime": 55735.5116,
      "train_tokens_per_second": 8491.621
    },
    {
      "epoch": 1.9462025316455698,
      "grad_norm": 1.2043177888468108,
      "learning_rate": 1.788672165707328e-08,
      "loss": 0.2609254837036133,
      "num_input_tokens_seen": 473366744,
      "step": 28905,
      "train_runtime": 55745.0172,
      "train_tokens_per_second": 8491.642
    },
    {
      "epoch": 1.9465391866415298,
      "grad_norm": 1.24708953574811,
      "learning_rate": 1.7663967664465763e-08,
      "loss": 0.3025341272354126,
      "num_input_tokens_seen": 473448128,
      "step": 28910,
      "train_runtime": 55754.8905,
      "train_tokens_per_second": 8491.598
    },
    {
      "epoch": 1.9468758416374898,
      "grad_norm": 1.049928038541275,
      "learning_rate": 1.7442606966242005e-08,
      "loss": 0.2944965362548828,
      "num_input_tokens_seen": 473530048,
      "step": 28915,
      "train_runtime": 55764.4925,
      "train_tokens_per_second": 8491.605
    },
    {
      "epoch": 1.94721249663345,
      "grad_norm": 1.1886419885287276,
      "learning_rate": 1.722263962430526e-08,
      "loss": 0.2762741565704346,
      "num_input_tokens_seen": 473611968,
      "step": 28920,
      "train_runtime": 55773.9942,
      "train_tokens_per_second": 8491.627
    },
    {
      "epoch": 1.9475491516294103,
      "grad_norm": 1.1359965723245822,
      "learning_rate": 1.7004065700168547e-08,
      "loss": 0.26909148693084717,
      "num_input_tokens_seen": 473693888,
      "step": 28925,
      "train_runtime": 55783.579,
      "train_tokens_per_second": 8491.637
    },
    {
      "epoch": 1.9478858066253704,
      "grad_norm": 1.146180981941028,
      "learning_rate": 1.6786885254954644e-08,
      "loss": 0.2638939142227173,
      "num_input_tokens_seen": 473775808,
      "step": 28930,
      "train_runtime": 55793.0859,
      "train_tokens_per_second": 8491.658
    },
    {
      "epoch": 1.9482224616213304,
      "grad_norm": 1.1427801443967385,
      "learning_rate": 1.6571098349398296e-08,
      "loss": 0.28124189376831055,
      "num_input_tokens_seen": 473856264,
      "step": 28935,
      "train_runtime": 55802.7736,
      "train_tokens_per_second": 8491.626
    },
    {
      "epoch": 1.9485591166172906,
      "grad_norm": 1.0141378937043803,
      "learning_rate": 1.635670504384346e-08,
      "loss": 0.2985966682434082,
      "num_input_tokens_seen": 473938184,
      "step": 28940,
      "train_runtime": 55812.2735,
      "train_tokens_per_second": 8491.648
    },
    {
      "epoch": 1.948895771613251,
      "grad_norm": 1.2345805169948934,
      "learning_rate": 1.6143705398243837e-08,
      "loss": 0.27730545997619627,
      "num_input_tokens_seen": 474020104,
      "step": 28945,
      "train_runtime": 55821.8735,
      "train_tokens_per_second": 8491.655
    },
    {
      "epoch": 1.949232426609211,
      "grad_norm": 1.192598687324503,
      "learning_rate": 1.5932099472165674e-08,
      "loss": 0.2574120044708252,
      "num_input_tokens_seen": 474102024,
      "step": 28950,
      "train_runtime": 55831.3787,
      "train_tokens_per_second": 8491.677
    },
    {
      "epoch": 1.949569081605171,
      "grad_norm": 1.0723933191899329,
      "learning_rate": 1.572188732478164e-08,
      "loss": 0.27837276458740234,
      "num_input_tokens_seen": 474183944,
      "step": 28955,
      "train_runtime": 55840.9914,
      "train_tokens_per_second": 8491.682
    },
    {
      "epoch": 1.9499057366011312,
      "grad_norm": 1.1225571078299323,
      "learning_rate": 1.551306901487859e-08,
      "loss": 0.26008551120758056,
      "num_input_tokens_seen": 474265864,
      "step": 28960,
      "train_runtime": 55850.5269,
      "train_tokens_per_second": 8491.699
    },
    {
      "epoch": 1.9502423915970915,
      "grad_norm": 1.0977222754434393,
      "learning_rate": 1.5305644600852043e-08,
      "loss": 0.28212289810180663,
      "num_input_tokens_seen": 474347784,
      "step": 28965,
      "train_runtime": 55860.0863,
      "train_tokens_per_second": 8491.712
    },
    {
      "epoch": 1.9505790465930515,
      "grad_norm": 1.1773840823486574,
      "learning_rate": 1.5099614140706154e-08,
      "loss": 0.2699897050857544,
      "num_input_tokens_seen": 474429704,
      "step": 28970,
      "train_runtime": 55869.6873,
      "train_tokens_per_second": 8491.719
    },
    {
      "epoch": 1.9509157015890115,
      "grad_norm": 1.1779189888584287,
      "learning_rate": 1.489497769205761e-08,
      "loss": 0.276621150970459,
      "num_input_tokens_seen": 474511624,
      "step": 28975,
      "train_runtime": 55879.1892,
      "train_tokens_per_second": 8491.741
    },
    {
      "epoch": 1.9512523565849718,
      "grad_norm": 1.1339510913010544,
      "learning_rate": 1.4691735312132304e-08,
      "loss": 0.28543848991394044,
      "num_input_tokens_seen": 474593544,
      "step": 28980,
      "train_runtime": 55888.7757,
      "train_tokens_per_second": 8491.751
    },
    {
      "epoch": 1.951589011580932,
      "grad_norm": 1.0268945864486363,
      "learning_rate": 1.4489887057766439e-08,
      "loss": 0.2698312759399414,
      "num_input_tokens_seen": 474675464,
      "step": 28985,
      "train_runtime": 55898.2739,
      "train_tokens_per_second": 8491.773
    },
    {
      "epoch": 1.951925666576892,
      "grad_norm": 1.1563589420155211,
      "learning_rate": 1.428943298540597e-08,
      "loss": 0.2771999597549438,
      "num_input_tokens_seen": 474757384,
      "step": 28990,
      "train_runtime": 55907.8617,
      "train_tokens_per_second": 8491.782
    },
    {
      "epoch": 1.952262321572852,
      "grad_norm": 1.1217174789385573,
      "learning_rate": 1.4090373151107173e-08,
      "loss": 0.310689115524292,
      "num_input_tokens_seen": 474839304,
      "step": 28995,
      "train_runtime": 55917.3583,
      "train_tokens_per_second": 8491.805
    },
    {
      "epoch": 1.9525989765688123,
      "grad_norm": 1.0360129537219023,
      "learning_rate": 1.3892707610536627e-08,
      "loss": 0.23756856918334962,
      "num_input_tokens_seen": 474921224,
      "step": 29000,
      "train_runtime": 55926.9237,
      "train_tokens_per_second": 8491.817
    },
    {
      "epoch": 1.9529356315647726,
      "grad_norm": 1.0251301499284569,
      "learning_rate": 1.3696436418970672e-08,
      "loss": 0.2746760129928589,
      "num_input_tokens_seen": 475002400,
      "step": 29005,
      "train_runtime": 55936.6347,
      "train_tokens_per_second": 8491.794
    },
    {
      "epoch": 1.9532722865607326,
      "grad_norm": 1.2115218639689207,
      "learning_rate": 1.3501559631296512e-08,
      "loss": 0.28583469390869143,
      "num_input_tokens_seen": 475084320,
      "step": 29010,
      "train_runtime": 55946.1849,
      "train_tokens_per_second": 8491.809
    },
    {
      "epoch": 1.9536089415566926,
      "grad_norm": 1.2376908724967126,
      "learning_rate": 1.3308077302010557e-08,
      "loss": 0.312956714630127,
      "num_input_tokens_seen": 475166240,
      "step": 29015,
      "train_runtime": 55955.7324,
      "train_tokens_per_second": 8491.824
    },
    {
      "epoch": 1.9539455965526529,
      "grad_norm": 1.2132582887832408,
      "learning_rate": 1.3115989485218972e-08,
      "loss": 0.2994795322418213,
      "num_input_tokens_seen": 475248160,
      "step": 29020,
      "train_runtime": 55965.2985,
      "train_tokens_per_second": 8491.836
    },
    {
      "epoch": 1.9542822515486131,
      "grad_norm": 1.1099169919965801,
      "learning_rate": 1.2925296234638784e-08,
      "loss": 0.277925968170166,
      "num_input_tokens_seen": 475330080,
      "step": 29025,
      "train_runtime": 55974.9163,
      "train_tokens_per_second": 8491.841
    },
    {
      "epoch": 1.9546189065445732,
      "grad_norm": 1.126939275602003,
      "learning_rate": 1.2735997603597339e-08,
      "loss": 0.25378167629241943,
      "num_input_tokens_seen": 475412000,
      "step": 29030,
      "train_runtime": 55984.4214,
      "train_tokens_per_second": 8491.862
    },
    {
      "epoch": 1.9549555615405332,
      "grad_norm": 1.1306247902877073,
      "learning_rate": 1.2548093645030623e-08,
      "loss": 0.28356122970581055,
      "num_input_tokens_seen": 475493920,
      "step": 29035,
      "train_runtime": 55994.0123,
      "train_tokens_per_second": 8491.871
    },
    {
      "epoch": 1.9552922165364934,
      "grad_norm": 1.195470005055211,
      "learning_rate": 1.236158441148605e-08,
      "loss": 0.2722217082977295,
      "num_input_tokens_seen": 475575840,
      "step": 29040,
      "train_runtime": 56003.5244,
      "train_tokens_per_second": 8491.891
    },
    {
      "epoch": 1.9556288715324537,
      "grad_norm": 1.2456167290293971,
      "learning_rate": 1.2176469955119119e-08,
      "loss": 0.3343416452407837,
      "num_input_tokens_seen": 475657760,
      "step": 29045,
      "train_runtime": 56013.1092,
      "train_tokens_per_second": 8491.901
    },
    {
      "epoch": 1.9559655265284137,
      "grad_norm": 1.4189184931602614,
      "learning_rate": 1.199275032769842e-08,
      "loss": 0.30692329406738283,
      "num_input_tokens_seen": 475739680,
      "step": 29050,
      "train_runtime": 56022.6062,
      "train_tokens_per_second": 8491.923
    },
    {
      "epoch": 1.9563021815243737,
      "grad_norm": 1.2351599288317856,
      "learning_rate": 1.1810425580598971e-08,
      "loss": 0.3084064483642578,
      "num_input_tokens_seen": 475821600,
      "step": 29055,
      "train_runtime": 56032.2202,
      "train_tokens_per_second": 8491.928
    },
    {
      "epoch": 1.956638836520334,
      "grad_norm": 1.285168905411459,
      "learning_rate": 1.1629495764807763e-08,
      "loss": 0.29056730270385744,
      "num_input_tokens_seen": 475903520,
      "step": 29060,
      "train_runtime": 56041.7689,
      "train_tokens_per_second": 8491.943
    },
    {
      "epoch": 1.9569754915162942,
      "grad_norm": 1.4584283872495394,
      "learning_rate": 1.1449960930921544e-08,
      "loss": 0.28259615898132323,
      "num_input_tokens_seen": 475985440,
      "step": 29065,
      "train_runtime": 56051.2993,
      "train_tokens_per_second": 8491.961
    },
    {
      "epoch": 1.9573121465122543,
      "grad_norm": 1.2246360632168998,
      "learning_rate": 1.1271821129146266e-08,
      "loss": 0.2511638879776001,
      "num_input_tokens_seen": 476067360,
      "step": 29070,
      "train_runtime": 56060.8541,
      "train_tokens_per_second": 8491.975
    },
    {
      "epoch": 1.9576488015082143,
      "grad_norm": 1.1994022156124375,
      "learning_rate": 1.1095076409298189e-08,
      "loss": 0.26639208793640134,
      "num_input_tokens_seen": 476149280,
      "step": 29075,
      "train_runtime": 56070.4256,
      "train_tokens_per_second": 8491.986
    },
    {
      "epoch": 1.9579854565041745,
      "grad_norm": 1.1478947147549998,
      "learning_rate": 1.0919726820803889e-08,
      "loss": 0.29723937511444093,
      "num_input_tokens_seen": 476231200,
      "step": 29080,
      "train_runtime": 56080.0244,
      "train_tokens_per_second": 8491.993
    },
    {
      "epoch": 1.9583221115001348,
      "grad_norm": 1.110414642366881,
      "learning_rate": 1.0745772412698585e-08,
      "loss": 0.29512710571289064,
      "num_input_tokens_seen": 476312976,
      "step": 29085,
      "train_runtime": 56090.0175,
      "train_tokens_per_second": 8491.938
    },
    {
      "epoch": 1.9586587664960948,
      "grad_norm": 1.0981095619477215,
      "learning_rate": 1.057321323362892e-08,
      "loss": 0.2642276048660278,
      "num_input_tokens_seen": 476394896,
      "step": 29090,
      "train_runtime": 56099.5695,
      "train_tokens_per_second": 8491.953
    },
    {
      "epoch": 1.9589954214920549,
      "grad_norm": 1.3002729872363687,
      "learning_rate": 1.0402049331849629e-08,
      "loss": 0.3080998659133911,
      "num_input_tokens_seen": 476476816,
      "step": 29095,
      "train_runtime": 56109.0692,
      "train_tokens_per_second": 8491.975
    },
    {
      "epoch": 1.959332076488015,
      "grad_norm": 1.0925607934356782,
      "learning_rate": 1.0232280755226865e-08,
      "loss": 0.2705387115478516,
      "num_input_tokens_seen": 476558736,
      "step": 29100,
      "train_runtime": 56118.6767,
      "train_tokens_per_second": 8491.981
    },
    {
      "epoch": 1.9596687314839754,
      "grad_norm": 0.8644232983644389,
      "learning_rate": 1.006390755123543e-08,
      "loss": 0.24955995082855226,
      "num_input_tokens_seen": 476640656,
      "step": 29105,
      "train_runtime": 56128.1829,
      "train_tokens_per_second": 8492.002
    },
    {
      "epoch": 1.9600053864799354,
      "grad_norm": 1.096243415526829,
      "learning_rate": 9.89692976696044e-09,
      "loss": 0.2944639205932617,
      "num_input_tokens_seen": 476722224,
      "step": 29110,
      "train_runtime": 56138.1817,
      "train_tokens_per_second": 8491.943
    },
    {
      "epoch": 1.9603420414758954,
      "grad_norm": 1.1624698974258452,
      "learning_rate": 9.731347449097316e-09,
      "loss": 0.29586856365203856,
      "num_input_tokens_seen": 476804144,
      "step": 29115,
      "train_runtime": 56147.7181,
      "train_tokens_per_second": 8491.959
    },
    {
      "epoch": 1.9606786964718557,
      "grad_norm": 1.3055901496899625,
      "learning_rate": 9.56716064394958e-09,
      "loss": 0.2586220741271973,
      "num_input_tokens_seen": 476886064,
      "step": 29120,
      "train_runtime": 56157.2765,
      "train_tokens_per_second": 8491.973
    },
    {
      "epoch": 1.961015351467816,
      "grad_norm": 1.2228198528327847,
      "learning_rate": 9.404369397432166e-09,
      "loss": 0.2650829553604126,
      "num_input_tokens_seen": 476967984,
      "step": 29125,
      "train_runtime": 56166.8204,
      "train_tokens_per_second": 8491.988
    },
    {
      "epoch": 1.961352006463776,
      "grad_norm": 1.184850869864241,
      "learning_rate": 9.242973755068663e-09,
      "loss": 0.27133352756500245,
      "num_input_tokens_seen": 477049904,
      "step": 29130,
      "train_runtime": 56176.3869,
      "train_tokens_per_second": 8492.0
    },
    {
      "epoch": 1.961688661459736,
      "grad_norm": 1.1404877252679915,
      "learning_rate": 9.082973761993518e-09,
      "loss": 0.2787674903869629,
      "num_input_tokens_seen": 477131824,
      "step": 29135,
      "train_runtime": 56185.976,
      "train_tokens_per_second": 8492.009
    },
    {
      "epoch": 1.9620253164556962,
      "grad_norm": 1.0726736912937869,
      "learning_rate": 8.924369462949834e-09,
      "loss": 0.31954593658447267,
      "num_input_tokens_seen": 477213744,
      "step": 29140,
      "train_runtime": 56195.4867,
      "train_tokens_per_second": 8492.03
    },
    {
      "epoch": 1.9623619714516565,
      "grad_norm": 1.164682316979339,
      "learning_rate": 8.767160902291016e-09,
      "loss": 0.2870827913284302,
      "num_input_tokens_seen": 477295664,
      "step": 29145,
      "train_runtime": 56205.0975,
      "train_tokens_per_second": 8492.035
    },
    {
      "epoch": 1.9626986264476165,
      "grad_norm": 1.040166300973189,
      "learning_rate": 8.61134812397968e-09,
      "loss": 0.2838404893875122,
      "num_input_tokens_seen": 477377584,
      "step": 29150,
      "train_runtime": 56214.7123,
      "train_tokens_per_second": 8492.04
    },
    {
      "epoch": 1.9630352814435765,
      "grad_norm": 1.1230827140723436,
      "learning_rate": 8.45693117158819e-09,
      "loss": 0.2685445070266724,
      "num_input_tokens_seen": 477459504,
      "step": 29155,
      "train_runtime": 56224.2157,
      "train_tokens_per_second": 8492.062
    },
    {
      "epoch": 1.9633719364395368,
      "grad_norm": 1.1901690605767008,
      "learning_rate": 8.303910088299228e-09,
      "loss": 0.28833684921264646,
      "num_input_tokens_seen": 477541424,
      "step": 29160,
      "train_runtime": 56233.8476,
      "train_tokens_per_second": 8492.064
    },
    {
      "epoch": 1.963708591435497,
      "grad_norm": 1.0306770869191832,
      "learning_rate": 8.152284916904674e-09,
      "loss": 0.2954511880874634,
      "num_input_tokens_seen": 477623344,
      "step": 29165,
      "train_runtime": 56243.3574,
      "train_tokens_per_second": 8492.085
    },
    {
      "epoch": 1.964045246431457,
      "grad_norm": 0.9836859190214101,
      "learning_rate": 8.002055699805612e-09,
      "loss": 0.2628525972366333,
      "num_input_tokens_seen": 477705264,
      "step": 29170,
      "train_runtime": 56252.9659,
      "train_tokens_per_second": 8492.09
    },
    {
      "epoch": 1.964381901427417,
      "grad_norm": 1.0870262337180219,
      "learning_rate": 7.853222479013434e-09,
      "loss": 0.25813541412353513,
      "num_input_tokens_seen": 477787184,
      "step": 29175,
      "train_runtime": 56262.4693,
      "train_tokens_per_second": 8492.112
    },
    {
      "epoch": 1.9647185564233773,
      "grad_norm": 1.0493101105446943,
      "learning_rate": 7.705785296148737e-09,
      "loss": 0.27676820755004883,
      "num_input_tokens_seen": 477869104,
      "step": 29180,
      "train_runtime": 56272.0884,
      "train_tokens_per_second": 8492.116
    },
    {
      "epoch": 1.9650552114193376,
      "grad_norm": 1.2747652795124207,
      "learning_rate": 7.55974419244132e-09,
      "loss": 0.3056499481201172,
      "num_input_tokens_seen": 477951024,
      "step": 29185,
      "train_runtime": 56281.6385,
      "train_tokens_per_second": 8492.131
    },
    {
      "epoch": 1.9653918664152976,
      "grad_norm": 1.1246184455590087,
      "learning_rate": 7.415099208732402e-09,
      "loss": 0.27393202781677245,
      "num_input_tokens_seen": 478032944,
      "step": 29190,
      "train_runtime": 56291.2107,
      "train_tokens_per_second": 8492.142
    },
    {
      "epoch": 1.9657285214112576,
      "grad_norm": 1.0666148422569175,
      "learning_rate": 7.2718503854707355e-09,
      "loss": 0.2755672693252563,
      "num_input_tokens_seen": 478114632,
      "step": 29195,
      "train_runtime": 56301.256,
      "train_tokens_per_second": 8492.078
    },
    {
      "epoch": 1.966065176407218,
      "grad_norm": 1.10204931060444,
      "learning_rate": 7.129997762715391e-09,
      "loss": 0.2519240856170654,
      "num_input_tokens_seen": 478196552,
      "step": 29200,
      "train_runtime": 56310.8366,
      "train_tokens_per_second": 8492.088
    },
    {
      "epoch": 1.9664018314031781,
      "grad_norm": 1.0613856894782412,
      "learning_rate": 6.98954138013519e-09,
      "loss": 0.26471059322357177,
      "num_input_tokens_seen": 478278472,
      "step": 29205,
      "train_runtime": 56320.44,
      "train_tokens_per_second": 8492.094
    },
    {
      "epoch": 1.9667384863991382,
      "grad_norm": 1.1988953414267163,
      "learning_rate": 6.850481277008159e-09,
      "loss": 0.2982466697692871,
      "num_input_tokens_seen": 478360392,
      "step": 29210,
      "train_runtime": 56329.953,
      "train_tokens_per_second": 8492.114
    },
    {
      "epoch": 1.9670751413950982,
      "grad_norm": 1.2493171265918213,
      "learning_rate": 6.712817492222079e-09,
      "loss": 0.2936177968978882,
      "num_input_tokens_seen": 478442312,
      "step": 29215,
      "train_runtime": 56339.5505,
      "train_tokens_per_second": 8492.122
    },
    {
      "epoch": 1.9674117963910585,
      "grad_norm": 1.0678175716680471,
      "learning_rate": 6.576550064273934e-09,
      "loss": 0.2794138193130493,
      "num_input_tokens_seen": 478524040,
      "step": 29220,
      "train_runtime": 56349.544,
      "train_tokens_per_second": 8492.066
    },
    {
      "epoch": 1.9677484513870187,
      "grad_norm": 1.0919857054457407,
      "learning_rate": 6.44167903127102e-09,
      "loss": 0.28810174465179444,
      "num_input_tokens_seen": 478605960,
      "step": 29225,
      "train_runtime": 56359.1337,
      "train_tokens_per_second": 8492.074
    },
    {
      "epoch": 1.9680851063829787,
      "grad_norm": 1.1339463667013148,
      "learning_rate": 6.3082044309287215e-09,
      "loss": 0.2730569839477539,
      "num_input_tokens_seen": 478687880,
      "step": 29230,
      "train_runtime": 56368.6564,
      "train_tokens_per_second": 8492.093
    },
    {
      "epoch": 1.9684217613789388,
      "grad_norm": 0.9597967158642314,
      "learning_rate": 6.176126300573848e-09,
      "loss": 0.2570362567901611,
      "num_input_tokens_seen": 478769800,
      "step": 29235,
      "train_runtime": 56378.2473,
      "train_tokens_per_second": 8492.102
    },
    {
      "epoch": 1.968758416374899,
      "grad_norm": 1.3414361738970622,
      "learning_rate": 6.045444677140744e-09,
      "loss": 0.26209487915039065,
      "num_input_tokens_seen": 478851720,
      "step": 29240,
      "train_runtime": 56387.7866,
      "train_tokens_per_second": 8492.118
    },
    {
      "epoch": 1.9690950713708593,
      "grad_norm": 1.2271651546659856,
      "learning_rate": 5.91615959717462e-09,
      "loss": 0.2603108882904053,
      "num_input_tokens_seen": 478933640,
      "step": 29245,
      "train_runtime": 56397.3329,
      "train_tokens_per_second": 8492.133
    },
    {
      "epoch": 1.9694317263668193,
      "grad_norm": 0.9448811124311315,
      "learning_rate": 5.78827109682989e-09,
      "loss": 0.2528717517852783,
      "num_input_tokens_seen": 479015560,
      "step": 29250,
      "train_runtime": 56406.9299,
      "train_tokens_per_second": 8492.14
    },
    {
      "epoch": 1.9697683813627793,
      "grad_norm": 1.1926547120156916,
      "learning_rate": 5.661779211869056e-09,
      "loss": 0.29475648403167726,
      "num_input_tokens_seen": 479097480,
      "step": 29255,
      "train_runtime": 56416.4262,
      "train_tokens_per_second": 8492.163
    },
    {
      "epoch": 1.9701050363587396,
      "grad_norm": 1.301074527754965,
      "learning_rate": 5.536683977666601e-09,
      "loss": 0.29985995292663575,
      "num_input_tokens_seen": 479179400,
      "step": 29260,
      "train_runtime": 56425.9692,
      "train_tokens_per_second": 8492.178
    },
    {
      "epoch": 1.9704416913546998,
      "grad_norm": 1.1929332977185736,
      "learning_rate": 5.412985429203988e-09,
      "loss": 0.29673702716827394,
      "num_input_tokens_seen": 479261320,
      "step": 29265,
      "train_runtime": 56435.4801,
      "train_tokens_per_second": 8492.199
    },
    {
      "epoch": 1.9707783463506598,
      "grad_norm": 1.0370505766424243,
      "learning_rate": 5.290683601073543e-09,
      "loss": 0.2650864362716675,
      "num_input_tokens_seen": 479343240,
      "step": 29270,
      "train_runtime": 56445.0748,
      "train_tokens_per_second": 8492.207
    },
    {
      "epoch": 1.9711150013466199,
      "grad_norm": 0.9880486171284946,
      "learning_rate": 5.169778527476243e-09,
      "loss": 0.2686916828155518,
      "num_input_tokens_seen": 479425160,
      "step": 29275,
      "train_runtime": 56454.5758,
      "train_tokens_per_second": 8492.229
    },
    {
      "epoch": 1.9714516563425801,
      "grad_norm": 1.0886919304036013,
      "learning_rate": 5.050270242222821e-09,
      "loss": 0.28948140144348145,
      "num_input_tokens_seen": 479507080,
      "step": 29280,
      "train_runtime": 56464.1518,
      "train_tokens_per_second": 8492.239
    },
    {
      "epoch": 1.9717883113385404,
      "grad_norm": 1.0842037811723346,
      "learning_rate": 4.932158778733765e-09,
      "loss": 0.28247981071472167,
      "num_input_tokens_seen": 479589000,
      "step": 29285,
      "train_runtime": 56473.6595,
      "train_tokens_per_second": 8492.26
    },
    {
      "epoch": 1.9721249663345004,
      "grad_norm": 1.2104703983649792,
      "learning_rate": 4.8154441700387636e-09,
      "loss": 0.30244832038879393,
      "num_input_tokens_seen": 479669624,
      "step": 29290,
      "train_runtime": 56483.7616,
      "train_tokens_per_second": 8492.169
    },
    {
      "epoch": 1.9724616213304604,
      "grad_norm": 1.0951969987124557,
      "learning_rate": 4.7001264487761545e-09,
      "loss": 0.28754215240478515,
      "num_input_tokens_seen": 479751544,
      "step": 29295,
      "train_runtime": 56493.3016,
      "train_tokens_per_second": 8492.185
    },
    {
      "epoch": 1.9727982763264207,
      "grad_norm": 1.3847954644488552,
      "learning_rate": 4.586205647194031e-09,
      "loss": 0.3141560316085815,
      "num_input_tokens_seen": 479833464,
      "step": 29300,
      "train_runtime": 56502.8708,
      "train_tokens_per_second": 8492.196
    },
    {
      "epoch": 1.973134931322381,
      "grad_norm": 1.2758586009496973,
      "learning_rate": 4.4736817971507974e-09,
      "loss": 0.2900221347808838,
      "num_input_tokens_seen": 479915384,
      "step": 29305,
      "train_runtime": 56512.4721,
      "train_tokens_per_second": 8492.203
    },
    {
      "epoch": 1.973471586318341,
      "grad_norm": 1.1463414069267126,
      "learning_rate": 4.362554930112395e-09,
      "loss": 0.2874056339263916,
      "num_input_tokens_seen": 479997304,
      "step": 29310,
      "train_runtime": 56521.9789,
      "train_tokens_per_second": 8492.224
    },
    {
      "epoch": 1.973808241314301,
      "grad_norm": 1.1643776361762153,
      "learning_rate": 4.252825077155631e-09,
      "loss": 0.2943066120147705,
      "num_input_tokens_seen": 480079224,
      "step": 29315,
      "train_runtime": 56531.5813,
      "train_tokens_per_second": 8492.231
    },
    {
      "epoch": 1.9741448963102612,
      "grad_norm": 1.2657975769487622,
      "learning_rate": 4.1444922689665156e-09,
      "loss": 0.2664368152618408,
      "num_input_tokens_seen": 480161144,
      "step": 29320,
      "train_runtime": 56541.0875,
      "train_tokens_per_second": 8492.252
    },
    {
      "epoch": 1.9744815513062215,
      "grad_norm": 1.1003275681451172,
      "learning_rate": 4.037556535839149e-09,
      "loss": 0.26583256721496584,
      "num_input_tokens_seen": 480243064,
      "step": 29325,
      "train_runtime": 56550.677,
      "train_tokens_per_second": 8492.26
    },
    {
      "epoch": 1.9748182063021815,
      "grad_norm": 1.2899318735573024,
      "learning_rate": 3.932017907677943e-09,
      "loss": 0.26120171546936033,
      "num_input_tokens_seen": 480324984,
      "step": 29330,
      "train_runtime": 56560.1868,
      "train_tokens_per_second": 8492.281
    },
    {
      "epoch": 1.9751548612981416,
      "grad_norm": 1.121880262828436,
      "learning_rate": 3.827876413997067e-09,
      "loss": 0.2759050130844116,
      "num_input_tokens_seen": 480406904,
      "step": 29335,
      "train_runtime": 56569.7353,
      "train_tokens_per_second": 8492.295
    },
    {
      "epoch": 1.9754915162941018,
      "grad_norm": 0.8529969052991572,
      "learning_rate": 3.725132083918781e-09,
      "loss": 0.2487175941467285,
      "num_input_tokens_seen": 480488824,
      "step": 29340,
      "train_runtime": 56579.2347,
      "train_tokens_per_second": 8492.317
    },
    {
      "epoch": 1.975828171290062,
      "grad_norm": 1.1880049579855003,
      "learning_rate": 3.623784946175102e-09,
      "loss": 0.30795931816101074,
      "num_input_tokens_seen": 480570744,
      "step": 29345,
      "train_runtime": 56588.8392,
      "train_tokens_per_second": 8492.324
    },
    {
      "epoch": 1.976164826286022,
      "grad_norm": 0.9868372108932001,
      "learning_rate": 3.5238350291083577e-09,
      "loss": 0.24830348491668702,
      "num_input_tokens_seen": 480652664,
      "step": 29350,
      "train_runtime": 56598.3842,
      "train_tokens_per_second": 8492.339
    },
    {
      "epoch": 1.976501481281982,
      "grad_norm": 1.1582051447566661,
      "learning_rate": 3.4252823606678588e-09,
      "loss": 0.2889443874359131,
      "num_input_tokens_seen": 480734584,
      "step": 29355,
      "train_runtime": 56607.959,
      "train_tokens_per_second": 8492.35
    },
    {
      "epoch": 1.9768381362779424,
      "grad_norm": 1.2314546981795853,
      "learning_rate": 3.328126968414336e-09,
      "loss": 0.29518280029296873,
      "num_input_tokens_seen": 480816504,
      "step": 29360,
      "train_runtime": 56617.5136,
      "train_tokens_per_second": 8492.363
    },
    {
      "epoch": 1.9771747912739026,
      "grad_norm": 0.9957267944562501,
      "learning_rate": 3.232368879517167e-09,
      "loss": 0.28371777534484866,
      "num_input_tokens_seen": 480898424,
      "step": 29365,
      "train_runtime": 56627.0067,
      "train_tokens_per_second": 8492.386
    },
    {
      "epoch": 1.9775114462698626,
      "grad_norm": 1.1941335495090237,
      "learning_rate": 3.1380081207543766e-09,
      "loss": 0.29053943157196044,
      "num_input_tokens_seen": 480980344,
      "step": 29370,
      "train_runtime": 56636.5722,
      "train_tokens_per_second": 8492.399
    },
    {
      "epoch": 1.9778481012658227,
      "grad_norm": 1.0899653923513544,
      "learning_rate": 3.0450447185137454e-09,
      "loss": 0.2889542102813721,
      "num_input_tokens_seen": 481062264,
      "step": 29375,
      "train_runtime": 56646.072,
      "train_tokens_per_second": 8492.421
    },
    {
      "epoch": 1.978184756261783,
      "grad_norm": 1.1566605890960562,
      "learning_rate": 2.953478698792256e-09,
      "loss": 0.26831622123718263,
      "num_input_tokens_seen": 481144184,
      "step": 29380,
      "train_runtime": 56655.6155,
      "train_tokens_per_second": 8492.436
    },
    {
      "epoch": 1.9785214112577432,
      "grad_norm": 1.1742510649807782,
      "learning_rate": 2.8633100871960917e-09,
      "loss": 0.2468562602996826,
      "num_input_tokens_seen": 481226104,
      "step": 29385,
      "train_runtime": 56665.1095,
      "train_tokens_per_second": 8492.459
    },
    {
      "epoch": 1.9788580662537032,
      "grad_norm": 1.1626903777254791,
      "learning_rate": 2.774538908940638e-09,
      "loss": 0.28512983322143554,
      "num_input_tokens_seen": 481308024,
      "step": 29390,
      "train_runtime": 56674.6849,
      "train_tokens_per_second": 8492.469
    },
    {
      "epoch": 1.9791947212496632,
      "grad_norm": 1.2210133075875604,
      "learning_rate": 2.687165188849927e-09,
      "loss": 0.26739447116851806,
      "num_input_tokens_seen": 481389944,
      "step": 29395,
      "train_runtime": 56684.1837,
      "train_tokens_per_second": 8492.491
    },
    {
      "epoch": 1.9795313762456235,
      "grad_norm": 1.116196681316709,
      "learning_rate": 2.6011889513588574e-09,
      "loss": 0.26703782081604005,
      "num_input_tokens_seen": 481471864,
      "step": 29400,
      "train_runtime": 56693.7886,
      "train_tokens_per_second": 8492.498
    },
    {
      "epoch": 1.9798680312415837,
      "grad_norm": 1.1696898042999424,
      "learning_rate": 2.5166102205093077e-09,
      "loss": 0.3029764652252197,
      "num_input_tokens_seen": 481553784,
      "step": 29405,
      "train_runtime": 56703.329,
      "train_tokens_per_second": 8492.513
    },
    {
      "epoch": 1.9802046862375438,
      "grad_norm": 1.1316071307568631,
      "learning_rate": 2.4334290199540254e-09,
      "loss": 0.2683378219604492,
      "num_input_tokens_seen": 481635704,
      "step": 29410,
      "train_runtime": 56712.8441,
      "train_tokens_per_second": 8492.533
    },
    {
      "epoch": 1.9805413412335038,
      "grad_norm": 1.5033461343300067,
      "learning_rate": 2.351645372953848e-09,
      "loss": 0.25974164009094236,
      "num_input_tokens_seen": 481717624,
      "step": 29415,
      "train_runtime": 56722.4,
      "train_tokens_per_second": 8492.547
    },
    {
      "epoch": 1.980877996229464,
      "grad_norm": 1.0784792823391598,
      "learning_rate": 2.2712593023804795e-09,
      "loss": 0.26384625434875486,
      "num_input_tokens_seen": 481798064,
      "step": 29420,
      "train_runtime": 56732.5773,
      "train_tokens_per_second": 8492.441
    },
    {
      "epoch": 1.9812146512254243,
      "grad_norm": 1.0116503921421995,
      "learning_rate": 2.1922708307120512e-09,
      "loss": 0.2994229793548584,
      "num_input_tokens_seen": 481879984,
      "step": 29425,
      "train_runtime": 56742.1985,
      "train_tokens_per_second": 8492.445
    },
    {
      "epoch": 1.9815513062213843,
      "grad_norm": 0.9951422586721013,
      "learning_rate": 2.1146799800386698e-09,
      "loss": 0.2882896184921265,
      "num_input_tokens_seen": 481961904,
      "step": 29430,
      "train_runtime": 56751.7112,
      "train_tokens_per_second": 8492.465
    },
    {
      "epoch": 1.9818879612173443,
      "grad_norm": 1.122838355251461,
      "learning_rate": 2.0384867720579794e-09,
      "loss": 0.26388096809387207,
      "num_input_tokens_seen": 482043824,
      "step": 29435,
      "train_runtime": 56761.2675,
      "train_tokens_per_second": 8492.478
    },
    {
      "epoch": 1.9822246162133046,
      "grad_norm": 1.2632569611568454,
      "learning_rate": 1.96369122807738e-09,
      "loss": 0.28839774131774903,
      "num_input_tokens_seen": 482125744,
      "step": 29440,
      "train_runtime": 56770.767,
      "train_tokens_per_second": 8492.5
    },
    {
      "epoch": 1.9825612712092648,
      "grad_norm": 1.2011417179300796,
      "learning_rate": 1.8902933690123637e-09,
      "loss": 0.29303441047668455,
      "num_input_tokens_seen": 482207664,
      "step": 29445,
      "train_runtime": 56780.3606,
      "train_tokens_per_second": 8492.508
    },
    {
      "epoch": 1.9828979262052249,
      "grad_norm": 1.1539997243253324,
      "learning_rate": 1.8182932153892884e-09,
      "loss": 0.2880338430404663,
      "num_input_tokens_seen": 482289584,
      "step": 29450,
      "train_runtime": 56789.8632,
      "train_tokens_per_second": 8492.529
    },
    {
      "epoch": 1.983234581201185,
      "grad_norm": 1.3308249381727715,
      "learning_rate": 1.7476907873426041e-09,
      "loss": 0.2922609567642212,
      "num_input_tokens_seen": 482371504,
      "step": 29455,
      "train_runtime": 56799.4559,
      "train_tokens_per_second": 8492.537
    },
    {
      "epoch": 1.9835712361971451,
      "grad_norm": 1.1059872166582518,
      "learning_rate": 1.6784861046159617e-09,
      "loss": 0.24824082851409912,
      "num_input_tokens_seen": 482453424,
      "step": 29460,
      "train_runtime": 56808.9873,
      "train_tokens_per_second": 8492.555
    },
    {
      "epoch": 1.9839078911931054,
      "grad_norm": 1.2322319807774347,
      "learning_rate": 1.610679186561659e-09,
      "loss": 0.29932150840759275,
      "num_input_tokens_seen": 482535344,
      "step": 29465,
      "train_runtime": 56818.5596,
      "train_tokens_per_second": 8492.566
    },
    {
      "epoch": 1.9842445461890654,
      "grad_norm": 0.8421732156458445,
      "learning_rate": 1.5442700521428599e-09,
      "loss": 0.25074076652526855,
      "num_input_tokens_seen": 482617264,
      "step": 29470,
      "train_runtime": 56828.1101,
      "train_tokens_per_second": 8492.58
    },
    {
      "epoch": 1.9845812011850255,
      "grad_norm": 1.0918835787670027,
      "learning_rate": 1.4792587199297103e-09,
      "loss": 0.2858578681945801,
      "num_input_tokens_seen": 482699184,
      "step": 29475,
      "train_runtime": 56837.676,
      "train_tokens_per_second": 8492.592
    },
    {
      "epoch": 1.9849178561809857,
      "grad_norm": 1.0062692241015425,
      "learning_rate": 1.4156452081026672e-09,
      "loss": 0.2515398979187012,
      "num_input_tokens_seen": 482781104,
      "step": 29480,
      "train_runtime": 56847.2486,
      "train_tokens_per_second": 8492.603
    },
    {
      "epoch": 1.985254511176946,
      "grad_norm": 1.1574935598904534,
      "learning_rate": 1.3534295344513893e-09,
      "loss": 0.274001407623291,
      "num_input_tokens_seen": 482863024,
      "step": 29485,
      "train_runtime": 56856.7651,
      "train_tokens_per_second": 8492.622
    },
    {
      "epoch": 1.985591166172906,
      "grad_norm": 1.2607644535658717,
      "learning_rate": 1.2926117163741813e-09,
      "loss": 0.25748577117919924,
      "num_input_tokens_seen": 482944944,
      "step": 29490,
      "train_runtime": 56866.3655,
      "train_tokens_per_second": 8492.629
    },
    {
      "epoch": 1.985927821168866,
      "grad_norm": 1.112290120540828,
      "learning_rate": 1.233191770877995e-09,
      "loss": 0.27997732162475586,
      "num_input_tokens_seen": 483026864,
      "step": 29495,
      "train_runtime": 56875.8713,
      "train_tokens_per_second": 8492.65
    },
    {
      "epoch": 1.9862644761648263,
      "grad_norm": 1.10028719349929,
      "learning_rate": 1.1751697145800934e-09,
      "loss": 0.26193296909332275,
      "num_input_tokens_seen": 483108784,
      "step": 29500,
      "train_runtime": 56885.4601,
      "train_tokens_per_second": 8492.658
    },
    {
      "epoch": 1.9866011311607865,
      "grad_norm": 1.2109274137982928,
      "learning_rate": 1.1185455637052756e-09,
      "loss": 0.3264944314956665,
      "num_input_tokens_seen": 483190704,
      "step": 29505,
      "train_runtime": 56894.9665,
      "train_tokens_per_second": 8492.679
    },
    {
      "epoch": 1.9869377861567465,
      "grad_norm": 1.1404855267785359,
      "learning_rate": 1.063319334089763e-09,
      "loss": 0.2707049369812012,
      "num_input_tokens_seen": 483272168,
      "step": 29510,
      "train_runtime": 56904.9144,
      "train_tokens_per_second": 8492.626
    },
    {
      "epoch": 1.9872744411527066,
      "grad_norm": 1.1366704377509262,
      "learning_rate": 1.0094910411762027e-09,
      "loss": 0.2745485305786133,
      "num_input_tokens_seen": 483353480,
      "step": 29515,
      "train_runtime": 56914.8124,
      "train_tokens_per_second": 8492.578
    },
    {
      "epoch": 1.9876110961486668,
      "grad_norm": 1.1140182093340152,
      "learning_rate": 9.570607000175536e-10,
      "loss": 0.2902663707733154,
      "num_input_tokens_seen": 483435400,
      "step": 29520,
      "train_runtime": 56924.4203,
      "train_tokens_per_second": 8492.584
    },
    {
      "epoch": 1.987947751144627,
      "grad_norm": 1.2264697546756929,
      "learning_rate": 9.060283252765311e-10,
      "loss": 0.2768577814102173,
      "num_input_tokens_seen": 483517320,
      "step": 29525,
      "train_runtime": 56933.9267,
      "train_tokens_per_second": 8492.604
    },
    {
      "epoch": 1.988284406140587,
      "grad_norm": 0.9721793874587789,
      "learning_rate": 8.563939312233871e-10,
      "loss": 0.2772496700286865,
      "num_input_tokens_seen": 483599240,
      "step": 29530,
      "train_runtime": 56943.5366,
      "train_tokens_per_second": 8492.61
    },
    {
      "epoch": 1.9886210611365471,
      "grad_norm": 1.0193916473735152,
      "learning_rate": 8.081575317386847e-10,
      "loss": 0.2927415132522583,
      "num_input_tokens_seen": 483681160,
      "step": 29535,
      "train_runtime": 56953.1652,
      "train_tokens_per_second": 8492.612
    },
    {
      "epoch": 1.9889577161325074,
      "grad_norm": 1.0794765960591577,
      "learning_rate": 7.613191403116338e-10,
      "loss": 0.27599396705627444,
      "num_input_tokens_seen": 483763080,
      "step": 29540,
      "train_runtime": 56962.7893,
      "train_tokens_per_second": 8492.616
    },
    {
      "epoch": 1.9892943711284676,
      "grad_norm": 1.2340949730346118,
      "learning_rate": 7.158787700400905e-10,
      "loss": 0.28447666168212893,
      "num_input_tokens_seen": 483845000,
      "step": 29545,
      "train_runtime": 56972.299,
      "train_tokens_per_second": 8492.636
    },
    {
      "epoch": 1.9896310261244277,
      "grad_norm": 0.9682752540889712,
      "learning_rate": 6.718364336316674e-10,
      "loss": 0.2769911289215088,
      "num_input_tokens_seen": 483926920,
      "step": 29550,
      "train_runtime": 56981.9738,
      "train_tokens_per_second": 8492.632
    },
    {
      "epoch": 1.9899676811203877,
      "grad_norm": 1.036004411314547,
      "learning_rate": 6.291921434026238e-10,
      "loss": 0.27692337036132814,
      "num_input_tokens_seen": 484008840,
      "step": 29555,
      "train_runtime": 56991.6432,
      "train_tokens_per_second": 8492.628
    },
    {
      "epoch": 1.990304336116348,
      "grad_norm": 1.0131135257800925,
      "learning_rate": 5.879459112778651e-10,
      "loss": 0.23480734825134278,
      "num_input_tokens_seen": 484090760,
      "step": 29560,
      "train_runtime": 57001.1392,
      "train_tokens_per_second": 8492.651
    },
    {
      "epoch": 1.9906409911123082,
      "grad_norm": 1.0377361062733685,
      "learning_rate": 5.480977487920536e-10,
      "loss": 0.2679889440536499,
      "num_input_tokens_seen": 484172680,
      "step": 29565,
      "train_runtime": 57010.7295,
      "train_tokens_per_second": 8492.659
    },
    {
      "epoch": 1.9909776461082682,
      "grad_norm": 1.0377804551468823,
      "learning_rate": 5.096476670890526e-10,
      "loss": 0.27219266891479493,
      "num_input_tokens_seen": 484254600,
      "step": 29570,
      "train_runtime": 57020.2746,
      "train_tokens_per_second": 8492.674
    },
    {
      "epoch": 1.9913143011042282,
      "grad_norm": 1.0461521605084634,
      "learning_rate": 4.725956769202622e-10,
      "loss": 0.2593667984008789,
      "num_input_tokens_seen": 484335792,
      "step": 29575,
      "train_runtime": 57030.0376,
      "train_tokens_per_second": 8492.644
    },
    {
      "epoch": 1.9916509561001885,
      "grad_norm": 1.0463794037593583,
      "learning_rate": 4.369417886485039e-10,
      "loss": 0.275564980506897,
      "num_input_tokens_seen": 484417712,
      "step": 29580,
      "train_runtime": 57039.6353,
      "train_tokens_per_second": 8492.651
    },
    {
      "epoch": 1.9919876110961487,
      "grad_norm": 1.0874427576675012,
      "learning_rate": 4.026860122430254e-10,
      "loss": 0.27674570083618166,
      "num_input_tokens_seen": 484499632,
      "step": 29585,
      "train_runtime": 57049.1343,
      "train_tokens_per_second": 8492.673
    },
    {
      "epoch": 1.9923242660921088,
      "grad_norm": 1.1605554814491619,
      "learning_rate": 3.698283572839412e-10,
      "loss": 0.26306900978088377,
      "num_input_tokens_seen": 484581552,
      "step": 29590,
      "train_runtime": 57058.6919,
      "train_tokens_per_second": 8492.686
    },
    {
      "epoch": 1.9926609210880688,
      "grad_norm": 1.092988106479707,
      "learning_rate": 3.38368832959457e-10,
      "loss": 0.2756309986114502,
      "num_input_tokens_seen": 484663472,
      "step": 29595,
      "train_runtime": 57068.188,
      "train_tokens_per_second": 8492.708
    },
    {
      "epoch": 1.992997576084029,
      "grad_norm": 1.1845749463438988,
      "learning_rate": 3.0830744806753523e-10,
      "loss": 0.27869956493377684,
      "num_input_tokens_seen": 484745392,
      "step": 29600,
      "train_runtime": 57077.7694,
      "train_tokens_per_second": 8492.718
    },
    {
      "epoch": 1.9933342310799893,
      "grad_norm": 1.070897392748756,
      "learning_rate": 2.7964421101478457e-10,
      "loss": 0.27688212394714357,
      "num_input_tokens_seen": 484827312,
      "step": 29605,
      "train_runtime": 57087.2768,
      "train_tokens_per_second": 8492.738
    },
    {
      "epoch": 1.9936708860759493,
      "grad_norm": 1.1918531540959918,
      "learning_rate": 2.5237912981646016e-10,
      "loss": 0.2969186782836914,
      "num_input_tokens_seen": 484909232,
      "step": 29610,
      "train_runtime": 57096.8975,
      "train_tokens_per_second": 8492.742
    },
    {
      "epoch": 1.9940075410719094,
      "grad_norm": 1.2154714781976752,
      "learning_rate": 2.265122120975738e-10,
      "loss": 0.2792654037475586,
      "num_input_tokens_seen": 484991152,
      "step": 29615,
      "train_runtime": 57106.4002,
      "train_tokens_per_second": 8492.764
    },
    {
      "epoch": 1.9943441960678696,
      "grad_norm": 0.9638428952006853,
      "learning_rate": 2.020434650906733e-10,
      "loss": 0.2581496238708496,
      "num_input_tokens_seen": 485073072,
      "step": 29620,
      "train_runtime": 57116.0299,
      "train_tokens_per_second": 8492.766
    },
    {
      "epoch": 1.9946808510638299,
      "grad_norm": 1.0728070251322555,
      "learning_rate": 1.7897289563972854e-10,
      "loss": 0.26364703178405763,
      "num_input_tokens_seen": 485154992,
      "step": 29625,
      "train_runtime": 57125.5336,
      "train_tokens_per_second": 8492.787
    },
    {
      "epoch": 1.99501750605979,
      "grad_norm": 1.3822270810531716,
      "learning_rate": 1.5730051019569037e-10,
      "loss": 0.26564922332763674,
      "num_input_tokens_seen": 485236912,
      "step": 29630,
      "train_runtime": 57135.1269,
      "train_tokens_per_second": 8492.795
    },
    {
      "epoch": 1.99535416105575,
      "grad_norm": 1.1185261046786108,
      "learning_rate": 1.3702631481926633e-10,
      "loss": 0.2823180198669434,
      "num_input_tokens_seen": 485318832,
      "step": 29635,
      "train_runtime": 57144.677,
      "train_tokens_per_second": 8492.809
    },
    {
      "epoch": 1.9956908160517102,
      "grad_norm": 1.0766183356771801,
      "learning_rate": 1.1815031517981023e-10,
      "loss": 0.24714083671569825,
      "num_input_tokens_seen": 485400752,
      "step": 29640,
      "train_runtime": 57154.2307,
      "train_tokens_per_second": 8492.823
    },
    {
      "epoch": 1.9960274710476704,
      "grad_norm": 1.2695557770705146,
      "learning_rate": 1.0067251655643262e-10,
      "loss": 0.27916641235351564,
      "num_input_tokens_seen": 485482672,
      "step": 29645,
      "train_runtime": 57163.832,
      "train_tokens_per_second": 8492.829
    },
    {
      "epoch": 1.9963641260436304,
      "grad_norm": 1.0947680392598016,
      "learning_rate": 8.459292383633522e-11,
      "loss": 0.2621783256530762,
      "num_input_tokens_seen": 485564592,
      "step": 29650,
      "train_runtime": 57173.3313,
      "train_tokens_per_second": 8492.851
    },
    {
      "epoch": 1.9967007810395905,
      "grad_norm": 0.9747519259357583,
      "learning_rate": 6.991154151592128e-11,
      "loss": 0.27343771457672117,
      "num_input_tokens_seen": 485646512,
      "step": 29655,
      "train_runtime": 57182.9181,
      "train_tokens_per_second": 8492.86
    },
    {
      "epoch": 1.9970374360355507,
      "grad_norm": 1.199968954832356,
      "learning_rate": 5.662837370135066e-11,
      "loss": 0.3135037660598755,
      "num_input_tokens_seen": 485728432,
      "step": 29660,
      "train_runtime": 57192.4153,
      "train_tokens_per_second": 8492.882
    },
    {
      "epoch": 1.997374091031511,
      "grad_norm": 1.0238012842240038,
      "learning_rate": 4.47434241074296e-11,
      "loss": 0.28843259811401367,
      "num_input_tokens_seen": 485810352,
      "step": 29665,
      "train_runtime": 57202.0433,
      "train_tokens_per_second": 8492.885
    },
    {
      "epoch": 1.997710746027471,
      "grad_norm": 1.4951456465170827,
      "learning_rate": 3.4256696057055615e-11,
      "loss": 0.2860145092010498,
      "num_input_tokens_seen": 485892272,
      "step": 29670,
      "train_runtime": 57211.5415,
      "train_tokens_per_second": 8492.906
    },
    {
      "epoch": 1.998047401023431,
      "grad_norm": 1.0664702559840347,
      "learning_rate": 2.5168192482882824e-11,
      "loss": 0.28132631778717043,
      "num_input_tokens_seen": 485974192,
      "step": 29675,
      "train_runtime": 57221.1287,
      "train_tokens_per_second": 8492.915
    },
    {
      "epoch": 1.9983840560193913,
      "grad_norm": 1.1269157273613513,
      "learning_rate": 1.747791592676684e-11,
      "loss": 0.2730662107467651,
      "num_input_tokens_seen": 486056112,
      "step": 29680,
      "train_runtime": 57230.6156,
      "train_tokens_per_second": 8492.939
    },
    {
      "epoch": 1.9987207110153515,
      "grad_norm": 1.1223881946663372,
      "learning_rate": 1.1185868539209665e-11,
      "loss": 0.2449695110321045,
      "num_input_tokens_seen": 486138032,
      "step": 29685,
      "train_runtime": 57240.2132,
      "train_tokens_per_second": 8492.946
    },
    {
      "epoch": 1.9990573660113116,
      "grad_norm": 1.2872966003123092,
      "learning_rate": 6.29205207991479e-12,
      "loss": 0.2864171028137207,
      "num_input_tokens_seen": 486219016,
      "step": 29690,
      "train_runtime": 57249.9253,
      "train_tokens_per_second": 8492.92
    },
    {
      "epoch": 1.9993940210072716,
      "grad_norm": 1.018869018003156,
      "learning_rate": 2.7964679166769815e-12,
      "loss": 0.2656317949295044,
      "num_input_tokens_seen": 486300936,
      "step": 29695,
      "train_runtime": 57259.4663,
      "train_tokens_per_second": 8492.935
    },
    {
      "epoch": 1.9997306760032318,
      "grad_norm": 1.2064240858845547,
      "learning_rate": 6.991170281578364e-13,
      "loss": 0.25847413539886477,
      "num_input_tokens_seen": 486382120,
      "step": 29700,
      "train_runtime": 57269.2546,
      "train_tokens_per_second": 8492.901
    },
    {
      "epoch": 2.0,
      "num_input_tokens_seen": 486447656,
      "step": 29704,
      "total_flos": 899609053937664.0,
      "train_loss": 0.35186470192605007,
      "train_runtime": 57316.3921,
      "train_samples_per_second": 2.073,
      "train_steps_per_second": 0.518
    }
  ],
  "logging_steps": 5,
  "max_steps": 29704,
  "num_input_tokens_seen": 486447656,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 899609053937664.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}